(From the Division of Human Genetics, Department of Medicine, Cornell University Medical College, New York 10021) Materials and Methods

Please download to get full document.

View again

of 17
2 views
PDF
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
PREPARATION AND FURTHER CHARACTERIZATION OF THE MN GLYCOPROTEIN OF HUMAN ERYTHROCYTE MEMBRANES*':~ By HARTWIG CLEVE, HIDEO HAMAGUCHI, AXD THOMAS H!~TTEROTH (From the Division of Human Genetics, Department
Document Share
Document Transcript
PREPARATION AND FURTHER CHARACTERIZATION OF THE MN GLYCOPROTEIN OF HUMAN ERYTHROCYTE MEMBRANES*':~ By HARTWIG CLEVE, HIDEO HAMAGUCHI, AXD THOMAS H!~TTEROTH (From the Division of Human Genetics, Department of Medicine, Cornell University Medical College, New York 10021) (Received for publication 5 June 1972) The human erythrocyte membrane contains three sialoglycoproteins which can be distinguished by electrophoresis on acrylamide gels in the presence of sodium dodecyl sulfate (SDS) 1 (1, 2) 2. They appear to extend from the outer surface through the membrane barrier to the interior surface (2, 3). Recently, it was shown that these sialoglycoproteins can be solubilized and recovered almost quantitatively in the aqueous phase after extraction of red cell ghosts with a mixture of chloroform and methanol? In the present report, the preparation of the major glycoprotein, the so-called MN glycoprotein (4-10), from the aqueous phase of chloroform-methanol extracts is described. The partial characterization of this membrane glycoprotein includes the analysis by SDS-acrylamide gel electrophoresis, determination of carbohydrate and amino acid compositions, assays of inhibitory activities for hemagglutination by rabbit M- and N-antisera, influenza virus and Phaseolus vulgaris phytohemagglutinin, and assays of its inhibitory activities in various lymphocyte stimulation systems. Materials and Methods All chemicals employed in these studies were reagent grade. Protein concentrations were determined by the method of Lowry et al. (11) using bovine pancreatic ribonuclease (Worthington Biochemical Corp., Freehold, N. J.) as standard. The microhiuret method was carried out according to Goa (12). Cholesterol was measured by the method of Zlatkis et al. (13), and phosphorus was assayed by the technique of Bartlett (14). Sialic acid was determined by the method described by Warren (15) after hydrolysis in 0.1 g HzSO4 at 80 C fol 1 hr. Hexoses were determined with the orcinol method as described by Vasseur (16). Galactose-mannose ratios were obtained after chromatographic separation according to Spiro (17). Hexosamines were measured with Svennerholm's modification (18) of the Elson-Morgan reaction; galactosa- * Supported by U.S. Public Health Service Grant AM 11796, and aided by a grant from the National Foundation March of Dimes. :~ This wink was presented in part at the 4th International Congress of Human Genetics, Paris, 4-11 September Abbreviations used in this paper: CM, chloroform-methanol; GP, glycoprotein; PAS, periodic acid-schiff; PHA, phytohemagglutinin; SDS, sodium dodecyl sulfate. 2 Hamaguchi, H., and H. Cleve Solubilization of human erythrocyte membrane glycoproteins and separation of the MN glycoprotein from a glycoprotein with I, S and A activity. Biochim. Biophys. Aeta. 278: THE JOURNAL OF EXPERIMENTAL MEDICINE VOLUME 136, 1972 HARTWIG CLEVE, HIDEO HAMAGUCHI, THOMAS HUTTEROTH 1141 mine was assayed by the method of Ludowieg and Benmaman (19). Fucose was measured with a semimicro modification of the method of Dische and Shettles (20). For the chemical analyses the freeze-dried, purified MN glycoprotein was weighed in with a Cahn M I0 electrobalance (Cahn Instruments, Paramount, Calif.). Amino acid analysis was performed on a Beckman Model 120-B automatic amino acid analyzer (Beckman Instruments, Inc., Fullerton, Calif.) using the system of Moore and Stein (21). The protein was hydrolyzed in evacuated sealed tubes in 6 M HC1 at ll0 C for 22 hr. Methionine was determined after performic acid oxidation according to Moore (22). The least-square numerical method devised by Katz (23) was used for calculating the minimal molecular weight from the amino acid analysis data. SDS-acrylamide gel electrophoresis was carried out in 1% SDS-7.5% acrylamide gels as described previously (24). ~ The gels were stained with Coomassie brilliant blue (24) or with the periodic acid-schiff (PAS) reagent according to Zacharius et al. (25). Blood group activities and activities for myxovirus and phytohemagglutinin receptor sites were measured by hemagglufination inhibition tests. Twofold serial dilutions of antigencontaining materials were tested; the specific inhibitory activities were calculated as the smallest amount of material necessary to inhibit completely hemagglutination at four hemagglutinating units of reagent. The quantities were expressed in micrograms of protein per milliliter according to the values found with the Folin method of Lowry et al. (11). Rabbit antisera for M and N, and human antisera for A and B, were purchased from Behring Diagnostics Inc. (Woodbury, N. Y.). For assay of phytohemagglutinin inhibitory activity, PHA-P (Difco Laboratories, Detroit, Mich.) from Phaseolus vulgaris was used. The spot test described by Kornfeld and Kornfeld (26) was employed. Inhibition of myxovirus hemagglutination was determined according to Kathan et al. (4). RI/5+ strain of influenza virus A2 was kindly provided by Dr. Purnell W. Choppin, The Rockefeller University, New York. Inhibition of the mitogenic response of lymphocytes to various phytohemagglutinins was determined by measuring the incorporation of thymidine-~h into DNA by lymphocytes in short-term cultures to which different amounts of purified MN glycoproteins had been added. 1 ml cultures containing 0.2 ( 106 mononuclear cells placed in 13 X 100 mm tissue culture tubes (Falcon Plastics, Oxnard, Calif.) were used. Mter 72 hr, thymidine-3h incorporation was measured according to Hughes and Caspary (27). Phaseolus vulgarls phytohemagglutinins PHA-P (Difco Laboratories, lot No ) and purified phytohemagglutinin (Welleome Research Laboratories, Beckenham, England, lot No. K. 1360) were used in concentrations of 50 #g/ml and 0.5 #g/ml, respectively. Concanavalin A (K and K Laboratories, Inc., Plainview, N. Y., lot No ) was added at 10 #g/ml. RESULTS Preparation of MN Glycoprotein ml of human blood tom individual donors, collected in acid-citrate-dextlose and no more than 2 wk old, was used for the preparation of erythrocyte membranes. Plasma and buffy coat were removed and cells were washed three times with 0.3 ~t dextrose by centrifugation at +4 C for 15 rain at 2000 g. Cells were lysed with 9 vol of distilled water and ghosts were sedimented at 12,000 g for 20 rain. The ghosts were washed four times with 10 m~ tris(hydroxymethyl)aminomethane(tris)-0.1 m~ ethylenediaminetetraacetate (EDTA), ph 7.4, by centrifugation at 25,000 g for 30 min. Glycoproteins were extracted according to Kornfeld and Kornfeld (26) as described in detail elsewhere. ~ To 1 vol of packed ghosts suspended in Tris-EDTA buffer, 9 vol of a mixture of chloroform-methanol (2:1, v/v) weie added with vigorous shaking of the flask. The mixture was stirred for 30 min at room temperature and centrifuged afterwards for 10 min at 1000 g. The aqueous phase was carefully aspirated, centrifuged again in order to remove contaminating interphase components, and concentrated 1142 MN GLYCOPROTEINS OF HUMAN ERYTItROCYTE MEMBRANES to approximately one-fifth of the volume of the original ghost suspension. The aqueous phase of the chloroform-methanol (CM) extracts was, after reduction and alkylation, fraetionated further by gel filtration on Sepharose 4B (Pharmacia Fine Chemicals, Inc., Piscataway, N. J.) columns in the presence of 6 ~ guanidine hydrochloride (28). To the aqueous phase of CM extracts an equal weight of guanidine hydrochloride was added. Disulfide bonds were reduced by fl-mercaptoethanol for 4 hr and alkylation was carried out with 0.1 M iodoacetic acid in the presence of 0.3 ~ Tris for 1 hr. The solution remained clear throughout this procedure. The reduced and alkylated material was applied with 10% (w/v) sucrose to Sepharose 4B columns prepared with 6 M guanldine hydrochloride; two different sizes were used: 5 cm in diameter and 80 cm in length, or 2.5 cm in diameter and 80 cm in length, respectively. The columns were calibrated with purified polypeptide chains of known molecular weights I I I Ji t :L 0.06 o 0.04 ('4 o 0.02 El o 0,00 ::~' i ~ & } i[ I i! f I 2'~ o...,.?... ~ N, L,, ~ I I v'~ - ~ I I. ~ ~...,.. 50 I Effluent Volume (ml) FIG. t. Gel filtration on Sepharose 4B in 6 M guanidine hydrochloride. O... O, elution of reduced and alkylated human erythrocyte ghost proteins. --, elution of human erythrocyte membrane glycoproteins from aqueous phase after chloroform-methanol extraction. Fig. 1 demonstrates a representative elution diagram. The proteins of the aqueous phase of CM extracts are separated into two fractions. The elution pattern of total human erythrocyte ghost proteins on this column is given for comparison (Fig. 1, dotted line). The ghosts were solubilized with 6 M guanidine hydrochloride according to Gwynne and Tanford (28). The red cell ghost proteins were separated into five major fractions. The fractions of the two peaks from the aqueous phase of CM extracts were pooled as indicated, dialyzed for 4 days at +4 C against distilled water, and subsequently concentrated by vacuum ultrafiltration. Analysis of these two fractions were carried out by SDS-polyacrylamide gel electrophoresis (Fig. 2). The aqueous phase of CM extracts contains the glycoproteins GP I, GP II, and GP III and the glycoprotein component B as demonstrated previously. 2 The first peak from the Sepharose 4B-guanidine hydrochloride column is composed of several proteins: some high HARTWIG CLEVE, HIDEO HAMAGUCIII~ THOMAS IYUTTEROTH 1143 molecular weight components penetrate the gel only a few millimeters; these components may represent aggregated glycoproteins. One component migrates slightly slower than GP I; there is also a limited amount of GP I. In addition, substantial amounts of GP II and a fair amount of GP III are found. The second peak contains a single component. Although this material is eluted from the column late, that is at an effluent volume of approximately 250 ml, this glycoprotein GP I migrates relatively slowly on SDS-polyacrylamide gel electrophoresis. In the majority of preparations, the second peak contained only this single component. In the remainder of preparations, this peak contained very FIG. 2. Electrophoresis on 1% SDS-7.5% acrylamide gels. 1-3 stained by Coomassie brilliant blue: (1) aqueous phase of chloroform-methanol extracts, (2) fraction I from gel filtration on Sepharose 4B in 6 ~t guanidine hydrochloride, (3) fraction II from gel filtration on Sepharose 4B in 6 guanidine hydrochloride. 4-6 stained by periodic acid-schiff (PAS): (4) aqueous phase, (5) fraction I, (6) fraction II. GP, glycoprotein; GP I, II, and III refer to the three glycoproteins present in the aqueous phase of chloroform-methanol extracts. small amounts of contaminating GP II and GP III which could be disclosed when large amounts of protein (80 #g) were applied to the gels. Purity of these preparations was tested also by polyacrylamide gel electrophoresis without SDS in the presence of 8 M urea, and by electrophoresis on 5 % acrylamide gels, without SDS and without urea, using a 5 m~ Tris-38 m~ glycine buffer, ph 8.3, both as gel buffer and as electrode buffer (24). With both systems single components were observed which, however, ga~e slightly blurred patterns at the cathodal end of the protein band while the anodal front of the protein bands was sharply defined. The recovery of purified GP I, the so-called MN glycoprotein, was on the average approximately 5 mg of purified and freeze-dried protein/150 ml blood. It was noticed that the procedure could not be scaled up easily since the capacity 1144 MN GLYCOPROTEINS OF HUMAN ERYTHROCYTE MEMBRANES of the Sepharose 4B-6 M guanidine hydrochloride column appeared to be rather limited. Application of larger amounts of proteins from the aqueous phase of of CM extracts did not lead to higher yields of purified MN glycoprotein but to an increase of the first peak which contained relatively large amounts of GP I which were disporprotionately increased, than when smaller quantities of protein were applied to the column. Analysis of MN Glycoprotein.--The chemical composition of three MN glycoprotein prepalations is shown in Table I. Results are given in grams per 100 g dry weight of purified freeze-dried MN glycoprotein. The results of one preparation from each genotype, MlVl, MN, and NN, are presented. The material contains approximately 40% protein. It is noteworthy that the determination with the phenol reagent of Folin and Ciocalteau gave consistently lower protein estimates than the determination with the biuret reaction. Approximately 40-50% of the material is carbohydrate; approximately 20% of the TABLE I Chemical Composition oj Human Erythrocyte Membrane Glycoproteins MM, MN, and NN Constituent MM MN NN ~;/100 g g/lo0 g g/lo0 g Protein (biuret) Protein (folin) Sialic acid Hexosamine Hexose Fucose Phosphorus total is sialic acid. Traces of phosphorus were detected indicating that these preparations contain phospholipids despite preceding extraction by chloroformmethanol, subsequent treatment with 6 M guanidine hydrochloride, and separation by gel filtration on Sepharose 4B-guanidine hydrochloride columns. The phospholipid content may vary from approximately 20% in the MM preparation to less than 10% in the MN preparation (Table I). Further carbohydrate analysis showed galactosamine concentrations in the MM, MN, and NN preparations of 3.2, 6.4, and 3.4 g/100 g, respectively. The galactose/mannose ratios in the MM, MN, and NN preparations were found to be 3.0, 7.5, and 6.9, respectively. In Fig. 3 the patterns on 1% SDS-7.5 % polyacrylamide gel electrophoresis are shown. Differences in eleetrophoretic migration rates of the purified glycoproteins of types MM, MN, and NN are not revealed. Fig. 3 also illustrates the varying degrees of purity of different preparations. Whereas the MN glycoprotein shown is free of detectable contaminants, the MM preparation contains one and the NN preparation two minor faster migrating contaminants. Appar- HARTWIG CLEVE, HIDEO HAMAGUCHI, THOMAS HUTTEROTH 1145 ent molecular weights were determined from the migration rates accmding to Weber and Osborn (29). The apparent molecular weight of the MN glycoprotein was with this method found to be 58,000 daltons. Apparent molecular weights were also estimated from the elution volumes on Sepharose 4B-guanidine hydrochloride columns, as described by Fish, Mann, and Tanford (30). The column was calibrated with purified polypeptide chains; human immunoglobulin G light- and heavy-chains and human haptoglobin 1-1 a- and fl-chains were used (Fig. 4). The molecular weight of the MN glycoprotein estimated from the elution volume was found to be 24,000 daltons. Amino acid composition was determined by duplicate analyses of prepara- FIG. 3. Electrophoresis on 1% SDS-7.5% acrylamide gels. Purified MN glycoproteins: 1-3 stained by Coomassie brilliant blue; 4-6 stained by PAS reagent. (1 and 4) MM glycoprotein, (2 and 5) MN glycoprotein, and (3 and 6) NN glycoprotein. tions from three different individuals of each of the three genotypes. The results from these nine preparations are given in Table II. Significant differences among the genotypes MM, MN, and NN were not observed. The content of threonine serine, and glutamine is relatively high; the amount of methionine and phenylalanine is relatively low; half-cysteine residues were not discovered. The amino acid data were used to calculate the minimal molecular weight of the MN glycoprotein (23). The results are given in Fig. 5. The amino acid data are compatible with three different molecular weights for the peptide part of the MN glycoprotein. The molecular weights are 7000, 14,500, and 21,000, respectively. Assuming a peptide content of 40 % these values correspond to molecular weights of 17,500, 36,250, and 52,000, respectively for the total glycoprotein. In Table III the number of amino acid residues per peptide has been calculated for the three most compatible molecular weights. The total number of residues was 1146 JaN GLYCOPROTEINS OF HUMAN ERYTHROCYTE MEMBRANES estimated to be 54, 114, or 164, respectively. The statistical analysis is given by the values of Z (DI/R~) ~ X 10 ~, where the subscript i represents the ith kind of amino acid residue. Di is the difference between the analytical composition of an amino acid and its nearest integer for a trial molecular weight, and Ri is moles of i in 10 ~ g protein. I ~IgGH 3O 'o_ 20 x /1~Ig G L MN glycoprotein'._ lid a 5 I I I I Z Ve/Vo * FIo. 4. Determination of apparent molecular weight by gel filtration on Sepharose 4B in 6 ~ guanidine hydrochloride. For the calibration of the column, reduced and a]kylated purified polypeptide chains were used: IgG heavy-chain, haptoglobin ~-chain, IgG light-chain, and haptoglobin al chain ~ 2o Lo x ~,1 IO 5 I, ', l,,, I, ' ' l,, l l l l l l l l l l l l l l l l l l l l l l k l l l l k l l l l l l l l v- ~ ~, ~, ~- ~ ~- Molecular Weight Fic. 5. Calculation of minimal molecular weight by a least-square numerical method using amino acid analysis data. ]Plot of Y,(D~/R~)2 X 104 (see text) vs. trial molecular weight for MN glycoprotein. HARTWIG CLEVE, ~IIDEO HAMAGUCHI, THOMAS H-UTTEROTH 1147 Immunologic Activities of MN Glycoprotein.--The serologic activities of a total of 14 preparations are summarized in Table IV. Purified MM glycoprotein inhibits rabbit-anti-m serum at a concentration of approximately 5 #g/ml; if very large quantities of MM preparations are employed, 3000 #g/ml, a rabbitanti-n serum can also be inhibited. Purified NN glycoprotein inhibits a rabbitanti-n serum at a concentration of approximately 20 #g/ml. The rabbit-anti-m serum cannot be inhibited by this material. Purified preparations from individ- TABLE II Amino Acid Composition of MN Glycoproteins* Moles % MM MN NN Range Mean Range Mean Range Mean Lys , His Arg , Asp Thr Set Glu Pro Gly Ala Cys/ Val Mete: l Ile Leu Tyr , Phe , * For each genotype three preparations from three different individuals were analyzed in duplicates. Hydrolysis in 6 N HC1 fo~ 22 hr at 110 C. :~ Met after performic acid oxidation: 2.10 moles per cent (average of two preparations analyzed in duplicates). uals heterozygous for MN require more than twice the quantity of either homozygous preparation to inhibit M- or N-rabbit antisera. Table IV shows that from the MN preparation approximately 15/~g/ml are required to inhibit the M-, and that approximately 65/zg/ml are necessary to inhibit the N-antiserum. From 5.2 to 42.4 t~g/ml of the purified glycoprotein are necessary to inhibit myxovirus hemagglutination. Significant differences between the MN genotypes were not observed, although the quantities required from MM glycoprotein tended to be lower. The purified MN glycoprotein preparations also act as potent inhibitors for Phaseolus vulgaris phytohemagglutination. The specific inhibitory activity was found to be approximately 13/xg/ml with a relatively 1148 MN GLYCOPROTEINS OF HUN[AN ERYTHROCYTE MEMBRANES narrow range from 8.1 to 19.1 #g/ml. Differences in inhibitory activities for PHA were not observed for the three genotypes, MM, MN, and NN. The preparations were also tested for other serologic activities; a total of 15 preparations obtained from donors with blood group A were tested for the presence of the A antigen; 10 preparations were negative for A antigenic activity. In five preparations traces of A activity could be found. The specific inhibitory activities were calculated to be 45, 50, 380, 400, and 1850 t~g/ml, respectively. A total TABLE III Calculated Number of AA Residues Assumed molecular weight ,500 21,000 Lys His Arg Asp Thr Ser Glu Pro Gly , Ala Cys/ Val Met Ile Leu Tyr Phe \ed Total number of AA residues Data from duplicate analyses of three MN preparations from three different individuals were used. of 11 preparations were tested for the ability to inhibit hemagglutination by concanavalin A. All 11 preparations were found to
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks