ARCHIVES
Vol.
OF BIOCHEMISTRY
282, No. 1, October,
AND
BIOPHYSICS
pp. 34-38,
1990
Purification and Partial Characterization of Ceruloplasmin Receptors from Rat Liver Endothelium Eijiro
Omoto
University
Received
and Mehdi
of Mississippi
March
Tavassoli’
Medical
8, 1990, and in revised
Center and Veterans Administration
form
May
Center, Jackson, Mississippi
39216
14, 1990
Ceruloplasmin (CP), a circulating glycoprotein, is known for its copper transport. Recently the spectrum of its activity has been increased to include numerous enzymatic functions. CP binds to the liver endothelium and is transported across the cell via a mechanism involving receptor-mediated endocytosis. To isolate CP receptors, we obtained purified preparations of liver endothelium in rats. The membrane was then isolated by ultracentrifugation and solubilized in Triton X- 100. Membrane proteins were labeled with “‘1 and passed through an affinity column in which CP was covalently linked to Sepharose 4B. Most of the radioactivity was eluted with buffer during the first 5 days. When no more radioactivity was eluted with buffer, elution was done either competitively with cold excess CP or 1 M NaCl. By this technique, a sharp single peak of radioactivity was obtained and subjected to sodium dodecyl sulfatepolyacrylamide gel electrophoresis under reducing and nonreducing conditions. Under both conditions receptors appeared as a single band with M, of 35,000 containing 3% carbohydrate and an isoelectric point of 5.2. 0 1990 Academic Press, Inc.
Ceruloplasmin [CP;’ ferroxidase; iron(I1): oxygen oxidoreductase, EC 1.16.3.11 is a blue a-2-globulin in human plasma that binds six copper atoms. This protein is synthesized in the liver as a single protein chain with M, 132,000 and is composed of 1046 amino acids (l-3). Its range of physiological function has recently been considerably expanded to include many enzymatic functions including transport and donation of copper, ferroxidase, amine oxidase, superoxide dismutase, and deaminase ac1 To whom correspondence should be addressed. * Abbreviations used: CP, ceruloplasmin; D-PBS, Dulbecco’s phosphate-buffered saline; BSA, bovine serum albumin; SDS, sodium dodecyl sulfate; 2-ME, 2-mercaptoethanol; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; endo F, endoglycosidase F. 34
Medical
tivities (4-8). In the hereditary disorder Wilson’s disease (hepatolenticular degeneration), serum CP levels are characteristically decreased (9). Specific receptors for CP have been found in membrane fragments from aortic and cardiac tissues (lo), rat liver endothelium (ll-14), erythrocytes (15), monocytes, granulocytes, and lymphocytes (16). It appears that this molecule circulates in the plasma, and wherever its presence is needed, it can enter the cell via a receptor-mediated mechanism. Certain cell types such as CHO cells (17) or K562 cells (18) do not take up CP. We have previously reported that liver endothelium mediates the uptake of CP by hepatocytes (11-14). Endothelium takes up CP on the luminal side through a mechanism involving receptor-mediated endocytosis. The molecule moves across the endothelium via a vesicular pathway and is externalized on the abluminal side into the space of Disse where it is recognized and taken up by hepatocytes, apparently through the mechanism involving asialoglycoprotein receptors (13). CP is a glycoprotein of relatively large molecular mass. Its exquisite vulnerability to proteolysis has hampered studies of its structure, functions, and regulation (l-3,19). Availability of CP receptors could then permit isolation of CP by affinity technique. In the present study we report on the purification and characterization of CP binding molecules from rat liver endothelium. A CP-Sepharose affinity resin was used to obtain highly purified receptors, although in a relatively low yield. MATERIALS
AND
METHODS
Cellpreparation. Liver cell suspensions were prepared from 200- to 250-g male Sprague-Dawley rats by the collagenase perfusion method (11-14). Fractionation of these suspensions was done by centrifugal elutriation. The crude liver suspensions were suspended in Dulbecco’s PBS (D-PBS) and loaded in a Beckman JEG-B standard elutriator rotor (4.2 ml) using a JS-21 centrifuge (Beckman Instruments, Inc., Fullerton, CA). Before cell loading, the rotor was eluted with D-PBS at a flow rate of 11 ml/min. The rotor was then loaded, using a total volume of 4 ml while maintaining the same flow rate. Elutriation was 0003.9861/90 $3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
CERULOPLASMIN
RECEPTORS
done at 650g at 20°C. The flow rate was increased to 22 ml/mm and 400 ml of eluant fluid was collected. The yield was 4 X lo7 endothelial cells per liver with 92% purity as judged by indirect immunofluorescent staining for factor VIII as well as their distinctive features by scanning and transmission electron microscopy (20). Viability was 99% by the trypan blue exclusion test. Cell membrane preparation. To isolate receptors, the cell membrane fraction from elutriator-purified endothelial cells was obtained (21). These cells were remarkably free of hepatocytes. Cells were subjected to hypotonic shock (22) by suspension in a solution containing MgC12 and 10 mM Tris-HCl buffer (pH 7.4, total volume 2 ml containing 5 Kg/ml each of leupeptin, antipain, and pepstatin as protease inhibitors). This was followed by incubation for 30 min at 4°C. The cells were then homogenized in a homogenizer (Polytron Type PT 10 20 350D) for 15 s at 4°C. After homogenization 2 ml of 0.5 M sucrose was added immediately to obtain an isotonic concentration of 0.25 M. The homogenate was then centrifuged at 1OOOg for 10 min at 4°C to remove nuclei and cell debris. The supernate was further ultracentrifuged at 20,OOOg for 20 min at 4°C using a 12 X 51.mm centrifuge tube to remove the microsomal fraction. The sediment was resuspended in 0.25 M sucrose containing protease inhibitors, overlayed on top of 1.3 M sucrose, and subjected to further ultracentrifugation at 100,OOOg for 60 min. The intermediate layer contained the purified membrane fraction as demonstrated by TEM examination. The membrane fraction was then ultracentrifuged at 100,OOOg for 20 min at 4°C in 10 mM Tris buffer (pH 7.4) and the pellet was solubilized with 0.5% Triton X-100 containing protease inhibitors in a total volume of 0.4 ml. Nonsolubilized membrane was then removed by centrifugation at 100,OOOg for 60 min at 4°C and the supernate was harvested. The yield of protein, as determined by the Bradford method (23) was 250 pg from 5 X 10” endothelial cells. Protein iodination. Human CP was obtained from CalbiochemBehring Corp. (San Diego, CA) or Sigma Chemical Co. (St. Louis, MO). The protein was purified in our laboratory on Sephedex G-100 column. The purity was confirmed by isoelectric focusing as described (11, 12, 14). Cell membrane protein or CP was then labeled with r2’I by the lactoperoxidase and glucose oxidase method (24, 25). Briefly, 0.15 ml of immobilized lactoperoxidase (Bio-Rad, Richmond, CA) was added to 0.4 ml of protein solution which contained 250 pg as cell membrane protein or 2 mg as CP. To this was added 2.0 mCi Na ighI (ICN Pharmaceuticals, Inc., Irvine, CA) and 0.2 ml of 1% $-r-glucose and the reaction was permitted to continue for 30 min at room temperature. Immobilized lactoperoxidase beads were then removed by centrifugation at 1OOOg for 10 min, and the supernate was dialyzed for 48 h at 4°C to remove excess unlabeled la?. The specific activity of the protein was 1.6 pCi/pg for cell membrane protein or 0.12 pCi/Fg for CP. The iodinated protein was further checked by isoelectric focusing to assure its purity. Binding assay. The binding assay on purified whole endothelial cells was done as described previously (12). Briefly, 10s cells were incubated at 4°C for 60 min with an increasing concentration of ‘“‘I-CP in 0.25 ml of D-PBS that contained 0.5% BSA. Inhibition was done in the presence of 100.fold excess unlabeled CP. To perform binding assays on the cell membrane fraction, this fraction was resuspended in D-PBS at the concentration of 10 lg protein/ml. To 0.1 ml of this suspension (containing 1 gg protein) was added 0.05 ml of radiolabeled CP in increasing concentrations. The total volume of the incubation was adjusted to 0.25 ml by addition of 0.2 ml of D-PBS. Incubation was carried out in 11 X 60.mm ultracentrifuge tubes at 4°C for 60 min. Nonspecific binding was determined in the presence of 100.fold excess unlabeled CP. Then the pelleted protein was washed twice by ultracentrifugation at 25,OOOg for 20 min, and the bound radioactivity was counted in a gamma counter. Specific binding data were analyzed according to the method of Scatchard (26) to obtain the dissociation constant (&), maximum binding (B,,, ) and the number of receptors per cell or microgram cell membrane.
FROM
RAT
LIVER
35
Affinity chromatography and endoglycosidase treatment. An affinity column was prepared by covalent binding of 10 mg of CP to 1 ml of wet gel of CNBr-activated Sepharose 4B (Sigma) in 0.1 M NaHCO,, pH 8.8. This incubation was done for 16 h at 4°C. Then the CP-Sepharose 4B complex was washed and incubated with 1 M TrisHCl, pH 8.0, for 2 h at room temperature. Wet gel (2.5 ml) was applied to a column of 10 X 100 mm, and washed with 20 bed vol of D-PBS. To this column 0.4 ml of iz51-cell membrane protein was applied and allowed to permeate the gel. After an incubation for 1 h at 4”C, the column was washed with 300 bed vol of D-PBS to elute the labeled protein not bound to CP at a flow rate of 6 ml/h. Then the flow rate was decreased to 2 ml/h, and the column was washed another 20 bed vol of D-PBS. Thereafter, the column was competitively eluted with a solution containing an excess of CP (10 mg/ml in D-PBS) in order to specifically displace labeled bound protein from the CP-Sepharose 4B complex. Fractions of 1.0 ml were collected. The column was also eluted with 1 M NaCl in 0.1 M phosphate buffer, pH 5.5. The radioactivity of all fractions was counted in a gamma counter. An aliquot of the peak fraction that eluted from the affinity column was incubated for 4 h at 37°C with endoglycosidase F (endo F, Sigma) (27). Electrophoresis and autoradiography. Electrophoresis was done on polyacrylamide gels with sodium dodecyl sulfate (SDS) under reducing or nonreducing conditions (28, 29). The separation gel was prepared by using 8 ml of 30% acrylamide (w/v) and 0.8% (w/v) NJ-methylenebisacrylamide in distilled water. To this was added 7.5 ml of 1.5 M Tris-HCI (pH 8.8), and 1 ml of 10 4~ SDS and 13.2 ml of H,O. To the mixture was added 0.015 ml of tetramethylethylenediamine and 1 ml of 1.5% ammonium persulfate (w/v in H,O) to permit polymerization of acrylamide. The final concentration of acrylamide in the gel was 8%, although some experiments were done in 12% gel. The stacking gel was prepared in a similar fashion with a final concentration of 3%. The electrode buffer contained 0.025 M Tris base, 0.192 M glycine, and 0.1% SDS with a final pH of 8.3. Electrophoresis was carried out in gel plates of 16 X 13 cm with an inside diameter of 2 mm (LKB). A few drops of sample buffer were added to the applied sample consisting of 0.0625 M Tris-HC1 (pH 6.8), 2% SDS, 10% glycerol, and 5% 2-mercaptoethanol (2 ME), to provide reducing conditions. In addition, the sample was immersed for 3 min in boiling water. When reducing conditions were not desired (and denaturing conditions were still used), this sample buffer was replaced by a few drops of sucrose solution and the sample was not heated. As a marker 0.001% bromphenol blue dye was used. For molecular mass markers various proteins with M, ranging from 14,400 to 200,000 were used (Bio-Rad). Electrophoresis was done by applying 50 mA per gel and was continued until the marker dye reached the bottom of the gel. The gel was then fixed and stained overnight with 47.5% ethanol and 5% acetic acid containing 0.125% Coomassie brilliant blue R. The gel was then destained in graded alcohol and acetate and dried in a gel dryer for 1 h. Dried gels were exposed to Kodak film for 96 h.
RESULTS
Figure 1A shows the results of binding of CP to rat liver endothelial cell membrane. Nonspecific binding (binding in the presence of excess cold ligand) was linear and when this was subtracted from total binding, specific binding was obtained which was saturable at the ligand concentration of about 5 X lo-” M. The Scatchard analysis of the specific binding data is shown in the inset, demonstrating a single population of receptors with Kd of 1.24 X 10e7 M, which is similar to the Kd of the binding of CP to intact endothelium (Fig. 1B). The estimated number of receptors per 1 pg cell membrane protein was 2.88 X loll.
36
OMOTO
AND
Kdi 1.24~lO-~M Emax:
m 8
478 fmollpg
5 CP concentration
membrane
protein
10 (x 10 -6 M)
Kd: 1~10-~Y
B
Emax: 347 frd1106 ;
1 ,,h
No. of receptws
TAVASSOLI
This radioactive peak so obtained was then collected and concentrated in microconcentrator. It was then subjected to SDS-PAGE followed by autoradiography. Figure 3 shows an autoradiogram of a 12% gel under both reducing and nonreducing conditions. A single band with iVl,35,000 was found. The presence of 2-ME did not change the migration pattern of the band. The isoelectric point of this protein, determined by standard method (ll-14), was 5.2. To determine the degree of glycosylation of protein, the peak obtained from the column was subjected to endoglycosidase treatment. Treatment with endo F led to a reduction in the M,. (Fig. 4). The shift between the M, of the deglycosylated and original protein indicated a 3% carbohydrate content.
cells
I 2.1 x iO%el~s
DISCUSSION
The results presented here suggest that the CP binding molecule on liver endothelial cells is a glycoprotein with a M, of 35,000 and about 3% carbohydrate. It ap-
A
CP concentraHon
I
(x10-% CP
FIG. 1. (A) The binding of ‘251-CP to the endothelial cell membrane fraction. The membrane fraction (pg as protein) was incubated at 4°C for 60 min with an increasing concentration of “‘1-CP in 0.25 ml of DPBS that contained 1% BSA. Nonspecific binding was determined in the presence of 100.fold excess of unlabeled CP. Specific binding was calculated by subtracting nonspecific binding from total binding. The inset shows a scatchard plot constructed from the specific binding data. (B) The binding of 12’I-CP to intact endothelium. Incubations were performed under the same conditions as with the cell membrane fraction.
Our attempts to estimate the molecular mass of the receptor by the crosslinking technique, using the method of Sawyer et al. (30), proved unsuccessful. This might have been caused by protein degradation as a consequence of its fragility. We, therefore, proceeded with an affinity chromatographic approach. Figure 2A shows the elution pattern of affinity column chromatography. When ‘251-labeled solubilized membrane proteins were subjected to these columns, unbound radiolabeled membrane proteins were eluted over a period of 5 days, after which radioactivity was no longer eluted. After this competitive elution with excess unlabeled CP led to the appearance of a sharp single peak of radioactivity, but subsequent elution with 1 M NaCl in 0.1 M phosphate buffer, pH 5.5, did not yield another radioactive peaks. Figure 2B shows the elution pattern with 1 M NaCl. This elution again led to the appearance of a sharp single peak similar to that obtained by the competitive elution, indicating that the receptor can be similarly eluted with excess unlabeled CP or 1 M NaCl.
1 M NaCl
1
Fraction
number
150\ 1M NaCl
f z t IOOP
1
I
50 Fraction
200 number
250
FIG. 2. Elution pattern of lz51-labeled solubilized membrane proteins from CP-Sepharose 4B column. The methods are described in the text. Unlabeled and radiolabeled protein are eluted from the column over a period of 5 days. (A) After this time, competitive elution with excess unlabeled CP led to the appearance of a sharp single peak, but subsequent elution with 1 M NaCl in 0.1 M phosphate buffer, pH 5.5, did not yield another radioactive peak. (B) Elution pattern with 1 M NaCl in 0.1 M phosphate buffer, pH 5.5. This elution again led to the appearance of a sharp single peak similar to that obtained by competitive elution, indicating that the receptor can be similarly eluted with excess unlabeled CP or 1 M NaCl.
CERULOPLASMIN
RECEPTORS
pears as a single chain with no external disulfide bridges that could be removed by the reducing agent 2 ME. Since high purity endothelial cells, free of hepatocytes, were used in these experiments and since endothelial cells do not possess asialoglycoproteins, one is assured that the binding to the column has occurred via affinity for the protein part of the molecule and not its carbohydrate. While the relatively low yield does not permit further characterization at this time and, therefore, the use of the term receptor may not be strictly warranted, it is likely that this molecule represents a CP receptor. In line with this conclusion is the evidence that it is obtained from the plasma membrane of a cell type that can be demonstrated to possess CP receptors. Our affinity chromatographic approach, using solubilized membrane, provides highly purified materials which can be used in further studies. The disadvantage of this technique is that the yield is relatively low. Most CP binding molecules may have eluted with initial excessive washing of the column and, therefore, could not be recovered. This is in part due to the relatively low affinity of the protein for its receptors. Nonetheless, the small proportion of membrane binding molecule that can be recovered is in highly purified form. The glycoprotein nature of CP binding protein is expected, since most membrane proteins are glycosylated. However, the receptor appears to have a relatively low 44, compared to most other membrane receptors. It is
+2ME
-2ME
97,400 66.200
42,699
FROM
RAT
37
LIVER +Endo.
F
-Endo.
F
-
200.000
+
116.250
+
97,400
+
66.200
t
42,699
+
31,000
+
21,500
FIG. 4. The autoradiogram of the band with or without treatment with endoglycosidase F. The treatment leads to a 3% reduction in the molecular mass, indicating 3% carbohydrate content.
possible that what we have recovered here may only be a monomer subunit of a larger molecule which serves as the CP receptor in the intact cell. In agreement with this concept is the work of Barnes and Frieden (15), who have done the only other study of the CP receptor in red cells, and they reported the M, to be -60,000, which may correspond to a dimer of the molecule we have found. Conversely, their results may represent in vitro dimerization of the native molecule after isolation. Unfortunately, in our studies, without the use of SDS, a sharp band could not be detected to confirm or rule out these hypotheses. Protein sequencing analysis and cloning of the receptor gene are currently underway in our laboratory and may elucidate the nature of structure and function of this binding molecule. At any rate, isolation of this binding molecule may help to further elucidate the structure and function of CP itself.
21.500
ACKNOWLEDGMENTS 14,400
This authors
work was supported by NIH Grant DK-30142 to M.T. thank MS .Jackie Jordan for secretarial assistance.
The
REFERENCES FIG. 3.
Autoradiograms of a 12% gel showing the migration of the radioactive peak under reducing or nonreducing conditions. A single band with M, of 35,000 was found. The presence of 2 ME did not change the migration pattern of the band, indicating a single chain with no external disulfide binding. The materials in this figure and Fig. 4 were obtained by elution of affinity column with 1 M NaCl. Elution with excess cold ligand gave similar findings.
1. Kingston, Natl.
Acad.
2. Takashi, Sci.
USA
3. Ortel,
I. B., Kingston, Sci.
USA
B. L., and Putnum,
F. W. (1977)
N., Ortel, T., and Putnum, 8 1,390-394.
F. W. (1984)
Proc. Natl.
T. L., Takashi, N., and Putnum, F. W. (1984) CJSA 81,4761-4765. 4. Frieden, E. (1980) Ciba Found. Symp. 79,9%124. Acad.Sci.
Proc.
74,5377-5381.
Proc.
Acad. Natl.
38
OMOTO
5. Gutteridge,
J. M. C., and Stokes,
J. (1981)
AND
Crit. Rev. Clin. Lab. Sci.
14,257p329. 6. Cousins,
R. J. (1985)
7. Dormandy,
Physiol.
T. L. (1981)
Rev. 65, 238-309.
Agents
Actions
Suppl.
8. Goldstein, I. M., Kaplan, H. B., Edelsen, (1982) Ann. N. Y. Acad. Sci. 389,368-379.
‘79,93-124.
H. S., and Weissman, Science
116,484-485.
10. Stevens, chemistry
M. D., Disilvestro, 23,261-266.
R. A., and Harris,
E. D. (1984)
11. Kataoka,
M.,
M.
9. Scheinberg,
I. H., and Gitlin,
and Tavassoli,
D. (1952)
(1984)
Exp.
Cell Res.
G.
Bio-
155, 232-
240. 12. Tavassoli,
M., Kishimoto,
T., and Katoaka,
M. (1986)
J. Cell Biol.
102,1298-1303. 13. Tavassoli,
M. (1985)
Trans.
14. Irie, S., and Tavassoli,
Assoc. Amer.
M. (1986)
Biochem.
Phys.
98,370-377.
Biophys.
Res. Commun.
140,94-100. 15. Barnes, G., and Frieden, mun. 125,157-162. 16. Kataoka,
810.
M.,
E. (1984)
and Tavassoli,
M.
Biochem. (1985)
Exp.
Biophys. Hematol.
Res. Com-
13, 806-
TAVASSOLI 17. Orena, S. J., Goode, C. A., and Linder, M. C. (1986) Biochem. Biophys. Res. Commun. 129,822-829. 18. Percival, S. S., and Harris, E. D. (1990) Amer. J. Physiol. 258, C140-C146. 19. Aldred, A. R., Grimes, A., Schriber, G., and Mercer, J. F. B. (1987) J. Biol. Chem. 262,2875-2878. 20. Soda, R., and Tavassoli, M. (1983) Exp. Cell Rex 145,389-395. 21. Matsuoka, T., andTavassoli, M. (1989) J. Biol. Chem. 34,20,19320,198. 22. Hapel, A. J., Warren, H. S., and Hume, D. A. (1984) Blood 64, 786-790. 23. Bradford, M. M. (1976) Anal. Biochem. 72,248-254. 24. David, G. S. (1972) Biochem. Biophys. Res. Commun. 48,4644471. 25. Hubbard, A. L., and Cohn, Z. A. (1972) J. Cell Biol. 55,390. 26. Scatchard, G. (1949) Ann. N. Y. Acad. Sci. 51,660-672. 27. Elder, J. H., and Alexander, S. (1982) Proc. Natl. Acad. Sci. USA
79,4540-4544. 28. Laemmli, U. K. (1970)
Nature (London) 227,680-685. 29. Davis, B. J. (1964) Ann. N. Y. Acad. Sci. 121,404-427. 30. Sawyer, S. T., Krantz, S. B., and Luna, J. (1987) Proc. Sci. USA 84,3690-3694.
Natl.
Acad.
Vol.
OF BIOCHEMISTRY
282, No. 1, October,
AND
BIOPHYSICS
pp. 34-38,
1990
Purification and Partial Characterization of Ceruloplasmin Receptors from Rat Liver Endothelium Eijiro
Omoto
University
Received
and Mehdi
of Mississippi
March
Tavassoli’
Medical
8, 1990, and in revised
Center and Veterans Administration
form
May
Center, Jackson, Mississippi
39216
14, 1990
Ceruloplasmin (CP), a circulating glycoprotein, is known for its copper transport. Recently the spectrum of its activity has been increased to include numerous enzymatic functions. CP binds to the liver endothelium and is transported across the cell via a mechanism involving receptor-mediated endocytosis. To isolate CP receptors, we obtained purified preparations of liver endothelium in rats. The membrane was then isolated by ultracentrifugation and solubilized in Triton X- 100. Membrane proteins were labeled with “‘1 and passed through an affinity column in which CP was covalently linked to Sepharose 4B. Most of the radioactivity was eluted with buffer during the first 5 days. When no more radioactivity was eluted with buffer, elution was done either competitively with cold excess CP or 1 M NaCl. By this technique, a sharp single peak of radioactivity was obtained and subjected to sodium dodecyl sulfatepolyacrylamide gel electrophoresis under reducing and nonreducing conditions. Under both conditions receptors appeared as a single band with M, of 35,000 containing 3% carbohydrate and an isoelectric point of 5.2. 0 1990 Academic Press, Inc.
Ceruloplasmin [CP;’ ferroxidase; iron(I1): oxygen oxidoreductase, EC 1.16.3.11 is a blue a-2-globulin in human plasma that binds six copper atoms. This protein is synthesized in the liver as a single protein chain with M, 132,000 and is composed of 1046 amino acids (l-3). Its range of physiological function has recently been considerably expanded to include many enzymatic functions including transport and donation of copper, ferroxidase, amine oxidase, superoxide dismutase, and deaminase ac1 To whom correspondence should be addressed. * Abbreviations used: CP, ceruloplasmin; D-PBS, Dulbecco’s phosphate-buffered saline; BSA, bovine serum albumin; SDS, sodium dodecyl sulfate; 2-ME, 2-mercaptoethanol; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; endo F, endoglycosidase F. 34
Medical
tivities (4-8). In the hereditary disorder Wilson’s disease (hepatolenticular degeneration), serum CP levels are characteristically decreased (9). Specific receptors for CP have been found in membrane fragments from aortic and cardiac tissues (lo), rat liver endothelium (ll-14), erythrocytes (15), monocytes, granulocytes, and lymphocytes (16). It appears that this molecule circulates in the plasma, and wherever its presence is needed, it can enter the cell via a receptor-mediated mechanism. Certain cell types such as CHO cells (17) or K562 cells (18) do not take up CP. We have previously reported that liver endothelium mediates the uptake of CP by hepatocytes (11-14). Endothelium takes up CP on the luminal side through a mechanism involving receptor-mediated endocytosis. The molecule moves across the endothelium via a vesicular pathway and is externalized on the abluminal side into the space of Disse where it is recognized and taken up by hepatocytes, apparently through the mechanism involving asialoglycoprotein receptors (13). CP is a glycoprotein of relatively large molecular mass. Its exquisite vulnerability to proteolysis has hampered studies of its structure, functions, and regulation (l-3,19). Availability of CP receptors could then permit isolation of CP by affinity technique. In the present study we report on the purification and characterization of CP binding molecules from rat liver endothelium. A CP-Sepharose affinity resin was used to obtain highly purified receptors, although in a relatively low yield. MATERIALS
AND
METHODS
Cellpreparation. Liver cell suspensions were prepared from 200- to 250-g male Sprague-Dawley rats by the collagenase perfusion method (11-14). Fractionation of these suspensions was done by centrifugal elutriation. The crude liver suspensions were suspended in Dulbecco’s PBS (D-PBS) and loaded in a Beckman JEG-B standard elutriator rotor (4.2 ml) using a JS-21 centrifuge (Beckman Instruments, Inc., Fullerton, CA). Before cell loading, the rotor was eluted with D-PBS at a flow rate of 11 ml/min. The rotor was then loaded, using a total volume of 4 ml while maintaining the same flow rate. Elutriation was 0003.9861/90 $3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
CERULOPLASMIN
RECEPTORS
done at 650g at 20°C. The flow rate was increased to 22 ml/mm and 400 ml of eluant fluid was collected. The yield was 4 X lo7 endothelial cells per liver with 92% purity as judged by indirect immunofluorescent staining for factor VIII as well as their distinctive features by scanning and transmission electron microscopy (20). Viability was 99% by the trypan blue exclusion test. Cell membrane preparation. To isolate receptors, the cell membrane fraction from elutriator-purified endothelial cells was obtained (21). These cells were remarkably free of hepatocytes. Cells were subjected to hypotonic shock (22) by suspension in a solution containing MgC12 and 10 mM Tris-HCl buffer (pH 7.4, total volume 2 ml containing 5 Kg/ml each of leupeptin, antipain, and pepstatin as protease inhibitors). This was followed by incubation for 30 min at 4°C. The cells were then homogenized in a homogenizer (Polytron Type PT 10 20 350D) for 15 s at 4°C. After homogenization 2 ml of 0.5 M sucrose was added immediately to obtain an isotonic concentration of 0.25 M. The homogenate was then centrifuged at 1OOOg for 10 min at 4°C to remove nuclei and cell debris. The supernate was further ultracentrifuged at 20,OOOg for 20 min at 4°C using a 12 X 51.mm centrifuge tube to remove the microsomal fraction. The sediment was resuspended in 0.25 M sucrose containing protease inhibitors, overlayed on top of 1.3 M sucrose, and subjected to further ultracentrifugation at 100,OOOg for 60 min. The intermediate layer contained the purified membrane fraction as demonstrated by TEM examination. The membrane fraction was then ultracentrifuged at 100,OOOg for 20 min at 4°C in 10 mM Tris buffer (pH 7.4) and the pellet was solubilized with 0.5% Triton X-100 containing protease inhibitors in a total volume of 0.4 ml. Nonsolubilized membrane was then removed by centrifugation at 100,OOOg for 60 min at 4°C and the supernate was harvested. The yield of protein, as determined by the Bradford method (23) was 250 pg from 5 X 10” endothelial cells. Protein iodination. Human CP was obtained from CalbiochemBehring Corp. (San Diego, CA) or Sigma Chemical Co. (St. Louis, MO). The protein was purified in our laboratory on Sephedex G-100 column. The purity was confirmed by isoelectric focusing as described (11, 12, 14). Cell membrane protein or CP was then labeled with r2’I by the lactoperoxidase and glucose oxidase method (24, 25). Briefly, 0.15 ml of immobilized lactoperoxidase (Bio-Rad, Richmond, CA) was added to 0.4 ml of protein solution which contained 250 pg as cell membrane protein or 2 mg as CP. To this was added 2.0 mCi Na ighI (ICN Pharmaceuticals, Inc., Irvine, CA) and 0.2 ml of 1% $-r-glucose and the reaction was permitted to continue for 30 min at room temperature. Immobilized lactoperoxidase beads were then removed by centrifugation at 1OOOg for 10 min, and the supernate was dialyzed for 48 h at 4°C to remove excess unlabeled la?. The specific activity of the protein was 1.6 pCi/pg for cell membrane protein or 0.12 pCi/Fg for CP. The iodinated protein was further checked by isoelectric focusing to assure its purity. Binding assay. The binding assay on purified whole endothelial cells was done as described previously (12). Briefly, 10s cells were incubated at 4°C for 60 min with an increasing concentration of ‘“‘I-CP in 0.25 ml of D-PBS that contained 0.5% BSA. Inhibition was done in the presence of 100.fold excess unlabeled CP. To perform binding assays on the cell membrane fraction, this fraction was resuspended in D-PBS at the concentration of 10 lg protein/ml. To 0.1 ml of this suspension (containing 1 gg protein) was added 0.05 ml of radiolabeled CP in increasing concentrations. The total volume of the incubation was adjusted to 0.25 ml by addition of 0.2 ml of D-PBS. Incubation was carried out in 11 X 60.mm ultracentrifuge tubes at 4°C for 60 min. Nonspecific binding was determined in the presence of 100.fold excess unlabeled CP. Then the pelleted protein was washed twice by ultracentrifugation at 25,OOOg for 20 min, and the bound radioactivity was counted in a gamma counter. Specific binding data were analyzed according to the method of Scatchard (26) to obtain the dissociation constant (&), maximum binding (B,,, ) and the number of receptors per cell or microgram cell membrane.
FROM
RAT
LIVER
35
Affinity chromatography and endoglycosidase treatment. An affinity column was prepared by covalent binding of 10 mg of CP to 1 ml of wet gel of CNBr-activated Sepharose 4B (Sigma) in 0.1 M NaHCO,, pH 8.8. This incubation was done for 16 h at 4°C. Then the CP-Sepharose 4B complex was washed and incubated with 1 M TrisHCl, pH 8.0, for 2 h at room temperature. Wet gel (2.5 ml) was applied to a column of 10 X 100 mm, and washed with 20 bed vol of D-PBS. To this column 0.4 ml of iz51-cell membrane protein was applied and allowed to permeate the gel. After an incubation for 1 h at 4”C, the column was washed with 300 bed vol of D-PBS to elute the labeled protein not bound to CP at a flow rate of 6 ml/h. Then the flow rate was decreased to 2 ml/h, and the column was washed another 20 bed vol of D-PBS. Thereafter, the column was competitively eluted with a solution containing an excess of CP (10 mg/ml in D-PBS) in order to specifically displace labeled bound protein from the CP-Sepharose 4B complex. Fractions of 1.0 ml were collected. The column was also eluted with 1 M NaCl in 0.1 M phosphate buffer, pH 5.5. The radioactivity of all fractions was counted in a gamma counter. An aliquot of the peak fraction that eluted from the affinity column was incubated for 4 h at 37°C with endoglycosidase F (endo F, Sigma) (27). Electrophoresis and autoradiography. Electrophoresis was done on polyacrylamide gels with sodium dodecyl sulfate (SDS) under reducing or nonreducing conditions (28, 29). The separation gel was prepared by using 8 ml of 30% acrylamide (w/v) and 0.8% (w/v) NJ-methylenebisacrylamide in distilled water. To this was added 7.5 ml of 1.5 M Tris-HCI (pH 8.8), and 1 ml of 10 4~ SDS and 13.2 ml of H,O. To the mixture was added 0.015 ml of tetramethylethylenediamine and 1 ml of 1.5% ammonium persulfate (w/v in H,O) to permit polymerization of acrylamide. The final concentration of acrylamide in the gel was 8%, although some experiments were done in 12% gel. The stacking gel was prepared in a similar fashion with a final concentration of 3%. The electrode buffer contained 0.025 M Tris base, 0.192 M glycine, and 0.1% SDS with a final pH of 8.3. Electrophoresis was carried out in gel plates of 16 X 13 cm with an inside diameter of 2 mm (LKB). A few drops of sample buffer were added to the applied sample consisting of 0.0625 M Tris-HC1 (pH 6.8), 2% SDS, 10% glycerol, and 5% 2-mercaptoethanol (2 ME), to provide reducing conditions. In addition, the sample was immersed for 3 min in boiling water. When reducing conditions were not desired (and denaturing conditions were still used), this sample buffer was replaced by a few drops of sucrose solution and the sample was not heated. As a marker 0.001% bromphenol blue dye was used. For molecular mass markers various proteins with M, ranging from 14,400 to 200,000 were used (Bio-Rad). Electrophoresis was done by applying 50 mA per gel and was continued until the marker dye reached the bottom of the gel. The gel was then fixed and stained overnight with 47.5% ethanol and 5% acetic acid containing 0.125% Coomassie brilliant blue R. The gel was then destained in graded alcohol and acetate and dried in a gel dryer for 1 h. Dried gels were exposed to Kodak film for 96 h.
RESULTS
Figure 1A shows the results of binding of CP to rat liver endothelial cell membrane. Nonspecific binding (binding in the presence of excess cold ligand) was linear and when this was subtracted from total binding, specific binding was obtained which was saturable at the ligand concentration of about 5 X lo-” M. The Scatchard analysis of the specific binding data is shown in the inset, demonstrating a single population of receptors with Kd of 1.24 X 10e7 M, which is similar to the Kd of the binding of CP to intact endothelium (Fig. 1B). The estimated number of receptors per 1 pg cell membrane protein was 2.88 X loll.
36
OMOTO
AND
Kdi 1.24~lO-~M Emax:
m 8
478 fmollpg
5 CP concentration
membrane
protein
10 (x 10 -6 M)
Kd: 1~10-~Y
B
Emax: 347 frd1106 ;
1 ,,h
No. of receptws
TAVASSOLI
This radioactive peak so obtained was then collected and concentrated in microconcentrator. It was then subjected to SDS-PAGE followed by autoradiography. Figure 3 shows an autoradiogram of a 12% gel under both reducing and nonreducing conditions. A single band with iVl,35,000 was found. The presence of 2-ME did not change the migration pattern of the band. The isoelectric point of this protein, determined by standard method (ll-14), was 5.2. To determine the degree of glycosylation of protein, the peak obtained from the column was subjected to endoglycosidase treatment. Treatment with endo F led to a reduction in the M,. (Fig. 4). The shift between the M, of the deglycosylated and original protein indicated a 3% carbohydrate content.
cells
I 2.1 x iO%el~s
DISCUSSION
The results presented here suggest that the CP binding molecule on liver endothelial cells is a glycoprotein with a M, of 35,000 and about 3% carbohydrate. It ap-
A
CP concentraHon
I
(x10-% CP
FIG. 1. (A) The binding of ‘251-CP to the endothelial cell membrane fraction. The membrane fraction (pg as protein) was incubated at 4°C for 60 min with an increasing concentration of “‘1-CP in 0.25 ml of DPBS that contained 1% BSA. Nonspecific binding was determined in the presence of 100.fold excess of unlabeled CP. Specific binding was calculated by subtracting nonspecific binding from total binding. The inset shows a scatchard plot constructed from the specific binding data. (B) The binding of 12’I-CP to intact endothelium. Incubations were performed under the same conditions as with the cell membrane fraction.
Our attempts to estimate the molecular mass of the receptor by the crosslinking technique, using the method of Sawyer et al. (30), proved unsuccessful. This might have been caused by protein degradation as a consequence of its fragility. We, therefore, proceeded with an affinity chromatographic approach. Figure 2A shows the elution pattern of affinity column chromatography. When ‘251-labeled solubilized membrane proteins were subjected to these columns, unbound radiolabeled membrane proteins were eluted over a period of 5 days, after which radioactivity was no longer eluted. After this competitive elution with excess unlabeled CP led to the appearance of a sharp single peak of radioactivity, but subsequent elution with 1 M NaCl in 0.1 M phosphate buffer, pH 5.5, did not yield another radioactive peaks. Figure 2B shows the elution pattern with 1 M NaCl. This elution again led to the appearance of a sharp single peak similar to that obtained by the competitive elution, indicating that the receptor can be similarly eluted with excess unlabeled CP or 1 M NaCl.
1 M NaCl
1
Fraction
number
150\ 1M NaCl
f z t IOOP
1
I
50 Fraction
200 number
250
FIG. 2. Elution pattern of lz51-labeled solubilized membrane proteins from CP-Sepharose 4B column. The methods are described in the text. Unlabeled and radiolabeled protein are eluted from the column over a period of 5 days. (A) After this time, competitive elution with excess unlabeled CP led to the appearance of a sharp single peak, but subsequent elution with 1 M NaCl in 0.1 M phosphate buffer, pH 5.5, did not yield another radioactive peak. (B) Elution pattern with 1 M NaCl in 0.1 M phosphate buffer, pH 5.5. This elution again led to the appearance of a sharp single peak similar to that obtained by competitive elution, indicating that the receptor can be similarly eluted with excess unlabeled CP or 1 M NaCl.
CERULOPLASMIN
RECEPTORS
pears as a single chain with no external disulfide bridges that could be removed by the reducing agent 2 ME. Since high purity endothelial cells, free of hepatocytes, were used in these experiments and since endothelial cells do not possess asialoglycoproteins, one is assured that the binding to the column has occurred via affinity for the protein part of the molecule and not its carbohydrate. While the relatively low yield does not permit further characterization at this time and, therefore, the use of the term receptor may not be strictly warranted, it is likely that this molecule represents a CP receptor. In line with this conclusion is the evidence that it is obtained from the plasma membrane of a cell type that can be demonstrated to possess CP receptors. Our affinity chromatographic approach, using solubilized membrane, provides highly purified materials which can be used in further studies. The disadvantage of this technique is that the yield is relatively low. Most CP binding molecules may have eluted with initial excessive washing of the column and, therefore, could not be recovered. This is in part due to the relatively low affinity of the protein for its receptors. Nonetheless, the small proportion of membrane binding molecule that can be recovered is in highly purified form. The glycoprotein nature of CP binding protein is expected, since most membrane proteins are glycosylated. However, the receptor appears to have a relatively low 44, compared to most other membrane receptors. It is
+2ME
-2ME
97,400 66.200
42,699
FROM
RAT
37
LIVER +Endo.
F
-Endo.
F
-
200.000
+
116.250
+
97,400
+
66.200
t
42,699
+
31,000
+
21,500
FIG. 4. The autoradiogram of the band with or without treatment with endoglycosidase F. The treatment leads to a 3% reduction in the molecular mass, indicating 3% carbohydrate content.
possible that what we have recovered here may only be a monomer subunit of a larger molecule which serves as the CP receptor in the intact cell. In agreement with this concept is the work of Barnes and Frieden (15), who have done the only other study of the CP receptor in red cells, and they reported the M, to be -60,000, which may correspond to a dimer of the molecule we have found. Conversely, their results may represent in vitro dimerization of the native molecule after isolation. Unfortunately, in our studies, without the use of SDS, a sharp band could not be detected to confirm or rule out these hypotheses. Protein sequencing analysis and cloning of the receptor gene are currently underway in our laboratory and may elucidate the nature of structure and function of this binding molecule. At any rate, isolation of this binding molecule may help to further elucidate the structure and function of CP itself.
21.500
ACKNOWLEDGMENTS 14,400
This authors
work was supported by NIH Grant DK-30142 to M.T. thank MS .Jackie Jordan for secretarial assistance.
The
REFERENCES FIG. 3.
Autoradiograms of a 12% gel showing the migration of the radioactive peak under reducing or nonreducing conditions. A single band with M, of 35,000 was found. The presence of 2 ME did not change the migration pattern of the band, indicating a single chain with no external disulfide binding. The materials in this figure and Fig. 4 were obtained by elution of affinity column with 1 M NaCl. Elution with excess cold ligand gave similar findings.
1. Kingston, Natl.
Acad.
2. Takashi, Sci.
USA
3. Ortel,
I. B., Kingston, Sci.
USA
B. L., and Putnum,
F. W. (1977)
N., Ortel, T., and Putnum, 8 1,390-394.
F. W. (1984)
Proc. Natl.
T. L., Takashi, N., and Putnum, F. W. (1984) CJSA 81,4761-4765. 4. Frieden, E. (1980) Ciba Found. Symp. 79,9%124. Acad.Sci.
Proc.
74,5377-5381.
Proc.
Acad. Natl.
38
OMOTO
5. Gutteridge,
J. M. C., and Stokes,
J. (1981)
AND
Crit. Rev. Clin. Lab. Sci.
14,257p329. 6. Cousins,
R. J. (1985)
7. Dormandy,
Physiol.
T. L. (1981)
Rev. 65, 238-309.
Agents
Actions
Suppl.
8. Goldstein, I. M., Kaplan, H. B., Edelsen, (1982) Ann. N. Y. Acad. Sci. 389,368-379.
‘79,93-124.
H. S., and Weissman, Science
116,484-485.
10. Stevens, chemistry
M. D., Disilvestro, 23,261-266.
R. A., and Harris,
E. D. (1984)
11. Kataoka,
M.,
M.
9. Scheinberg,
I. H., and Gitlin,
and Tavassoli,
D. (1952)
(1984)
Exp.
Cell Res.
G.
Bio-
155, 232-
240. 12. Tavassoli,
M., Kishimoto,
T., and Katoaka,
M. (1986)
J. Cell Biol.
102,1298-1303. 13. Tavassoli,
M. (1985)
Trans.
14. Irie, S., and Tavassoli,
Assoc. Amer.
M. (1986)
Biochem.
Phys.
98,370-377.
Biophys.
Res. Commun.
140,94-100. 15. Barnes, G., and Frieden, mun. 125,157-162. 16. Kataoka,
810.
M.,
E. (1984)
and Tavassoli,
M.
Biochem. (1985)
Exp.
Biophys. Hematol.
Res. Com-
13, 806-
TAVASSOLI 17. Orena, S. J., Goode, C. A., and Linder, M. C. (1986) Biochem. Biophys. Res. Commun. 129,822-829. 18. Percival, S. S., and Harris, E. D. (1990) Amer. J. Physiol. 258, C140-C146. 19. Aldred, A. R., Grimes, A., Schriber, G., and Mercer, J. F. B. (1987) J. Biol. Chem. 262,2875-2878. 20. Soda, R., and Tavassoli, M. (1983) Exp. Cell Rex 145,389-395. 21. Matsuoka, T., andTavassoli, M. (1989) J. Biol. Chem. 34,20,19320,198. 22. Hapel, A. J., Warren, H. S., and Hume, D. A. (1984) Blood 64, 786-790. 23. Bradford, M. M. (1976) Anal. Biochem. 72,248-254. 24. David, G. S. (1972) Biochem. Biophys. Res. Commun. 48,4644471. 25. Hubbard, A. L., and Cohn, Z. A. (1972) J. Cell Biol. 55,390. 26. Scatchard, G. (1949) Ann. N. Y. Acad. Sci. 51,660-672. 27. Elder, J. H., and Alexander, S. (1982) Proc. Natl. Acad. Sci. USA
79,4540-4544. 28. Laemmli, U. K. (1970)
Nature (London) 227,680-685. 29. Davis, B. J. (1964) Ann. N. Y. Acad. Sci. 121,404-427. 30. Sawyer, S. T., Krantz, S. B., and Luna, J. (1987) Proc. Sci. USA 84,3690-3694.
Natl.
Acad.
