A tyrosine kinase associated with the red cell membrane phosphorylates band 3

Phosphorylation of Band 3, the anion transport protein of human erythrocyte membranes, has been studied by incubating isolated ghosts with [gamma-32P]ATP. One of the phosphate-acceptor sites is tyrosine 8 in the NH2-terminal cytoplasmic domain of the Band 3 protein. Seven out of 11 residues in the sequence surrounding the phosphorylated tyrosine are Asp or Glu. It is concluded that the erythrocyte, like other cells, contains a membrane-associated tyrosine kinase which phosphorylates highly anionic peptide acceptor sites.     

In vitro biosynthesis and membrane insertion of gamma-glutamyl transpeptidase

gamma-Glutamyl transpeptidase consists of two polypeptide chains anchored to the kidney brush-border membrane only through a short hydrophobic domain near the NH2-terminal end of the heavy subunit. The two subunits were reported to derive from a single polypeptide precursor by tissue labeling experiments. We have investigated the first steps of GGT biosynthesis and processing in a cell-free system. mRNA was prepared from kidney and enriched in specific sequences by a preparative gel electrophoresis. In vitro translation resulted in the synthesis of a single polypeptide (Mr = 63,000) specifically immunoprecipitated by antibodies raised against the mature dimeric enzyme. Incubation with microsomal membranes resulted in the appearance of a glycosylated form of the propeptide (Mr = 78,000). This latter form was cotranslationally segregated into microsomes and was sensitive to endoglycosidase H. Purified Escherichia coli leader peptidase did not process the primary gamma-glutamyl transpeptidase chain. This ectoprotein therefore appears to be inserted in the phospholipid bilayer without cleavage of a signal peptide, similar to most integral membrane proteins so far studied.     

Reactive sulfhydryl groups of the band 3 polypeptide from human erythroycte membranes. Location in the primary structure.

Human erythrocyte membranes contain a major transmembrane protein, known as Band 3, that is involved in anion transport. This protein contains a total of five reactive sulfhydryl groups, which can be assigned to either of two classes on the basis of their susceptibility to release from the membrane by trypsin. Two of the groups are located in the region COOH-terminal to the extracellular chymotrypsin-sensitive site of the protein and remain with a membrane-bound 55,000-dalton fragment generated by trypsin treatment. The three sulfhydryl groups NH2-terminal to the extracellular chymotrypsin site are released from the cytoplasmic surface of the membrane by trypsin. All three groups are present in a 20,000-dalton tryptic fragment of Band 3. Two of these groups are located very close to the sites of trypsin cleavage that generate the 20,000-dalton fragment. The third reactve group is probably located about 15,000-daltons from the most NH2-terminal sulfhydryl group. Two other well defined fragments of the protein do not contain reactive sulfhydryl groups. They are a 23,000-dalton fragment derived from the NH2-terminal end that is also released by trypsin from the cytoplasmic surface of the membrane and a 19,000-dalton membrane-bound region of the protein that is produced by treatment with chymotrypsin in ghosts. The 20,000-dalton tryptic fragment may, therefore, constitute a sulfhydryl-containing domain of the Band 3 protein.     

Biosynthesis and glycosylation of the human complement regulatory protein decay-accelerating factor

The biosynthesis and oligosaccharide structure of the human complement regulatory glycoprotein decay-accelerating factor (DAF) were studied in erythrocytes and cell lines. Initial information relative to carbohydrate moieties of DAF was obtained by enzymatic digestions. The 74,000 Mr erythrocyte DAF was lowered 3000 by endoglycosidase F, whereas endoglycosidase H had no effect, indicating one N-linked complex-type unit. Treatment with endo-alpha-N-acetylgalactosaminidase to remove O-linked oligosaccharides resulted in a 48,000 Mr molecule (67% of the Mr shift being due to sialic acid), which decreased to 45,000 Mr after sequential endoglycosidase F treatment. To additionally define the oligosaccharide structure and identify precursors in biosynthetic pathways, DAF was studied in the HL-60 cell line differentiated by vitamin D toward monocytes. Pulse-chase experiments with [35S]methionine revealed a precursor species of 43,000 Mr that underwent an early post-translational modification to a 46,000 Mr intermediate, and subsequently was chased into a mature species of 80,000 Mr that aligned with 125I surface-labeled DAF from these cells. All three forms of DAF were approximately 3000 lower in Mr in the presence of tunicamycin. The two lower Mr DAF species were sensitive to endoglycosidases F and H but not to neuraminidase or endo-alpha-N-acetylgalactosaminidase. In summary, DAF is synthesized as a 43,000 Mr precursor species containing one N-linked high-mannose unit. Before entering the central region of the Golgi, it is converted to a 46,000 Mr species by an as yet unknown post-translational modification. The 46,000 Mr form is converted to the 74,000 Mr (erythrocyte) or 80,000 Mr (leukocyte) membrane form of DAF by the addition of multiple, sialylated O-linked oligosaccharide chains (responsible for the large electrophoretic mobility shift) and conversion of the single N-linked high-mannose unit to a complex-type structure. The cell-specific Mr variation between red and white blood cells arises during this post-translational modification from the 46,000 Mr biosynthetic intermediate to the mature DAF species expressed on the cell surface.     

Role of glycosylation in the processing of newly translated insulin proreceptor in 3T3-L1 adipocytes

A procedure was developed for the immunoprecipitation of glycosylated and nonglycosylated forms of the insulin receptor and its precursors without prior purification using lectins. 3T3-L1 adipocytes were labeled with [35S]methionine after which 35S-labeled receptor polypeptides were specifically immunoprecipitated and characterized by sodium dodecyl sulfatepolyacrylamide gel electrophoresis. The first 35S-polypeptide detected was a 190-kDa glycosylated proreceptor which was rapidly (t1/2 approximately equal to 15 min) processed to a 210-kDa intermediate. The latter precursor was more slowly (t1/2 approximately equal to 2 h) proteolytically processed to 125-kDa (alpha') and 83-kDa (beta') precursors of the mature alpha- and beta-receptor subunits. Immediately prior to insertion into the plasma membrane, i.e. about 3 h after translation, the alpha'- and beta'-precursor polypeptides were converted to the mature 135-kDa alpha- and 95-kDa beta-receptor subunits. The characteristics of the oligosaccharide moieties of the receptor precursors and products were investigated. The 210-kDa precursor and its two products, the 125-kDa alpha'- and 83-kDa beta'-species, and the mature alpha- and beta-receptor subunits bind tightly to wheat germ lectin, whereas the 190-kDa proreceptor species is not bound. Upon incubation with endoglycosidase H, both the 210- and 190-kDa species are converted to a 180-kDa species. The 125-kDa alpha'- and 83-kDa beta'-species are also cleaved by endoglycosidase H, being reduced in size to 97 and 79 kDa, respectively. Based on their sensitivity to endoglycosidase H and insensitivity to neuraminidase, the oligosaccharide chains of the receptor precursors (190, 210, 125, and 83 kDa) do not contain terminal sialic acid (or other capping sugars). However, near the time of insertion into the plasma membrane, capping of the alpha'- and beta'-species by sialic acid occurs, giving rise to the mature 135-kDa alpha- and 95-kDa beta-receptor subunits, which are partially endoglycosidase H-resistant and neuraminidase-sensitive. When 3T3-L1 adipocytes are treated with tunicamycin, a 180-kDa proreceptor aglycopolypeptide is synthesized which is incapable of undergoing further processing and proteolytic cleavage to the alpha- and beta (or alpha'- and beta'-)-subunits. The 180-kDa species, which appears to be the aglyco-form of hte 190-kDa proreceptor generated by endoglycosidase H, is resistant to trypsin in the intact cell and apparently has not reached the cell surface. Thus, the oligosaccharide moieties of the insulin receptor precursor are crucial for proper processing, intracellular translocation, and formation of functionally competent insulin re     

Differential processing and turnover of the oncogenically activated neu/erb B2 gene product and its normal cellular counterpart

In nontransformed DHFR/G-8 cells (NIH 3T3 cells transfected with normal rat neu gene), the normal neu gene product was initially synthesized as a 170-kDa protein bearing endoglycosidase H-sensitive oligosaccharide chains and was then processed to a 175-kDa mature form with endoglycosidase H-resistant, endoglycosidase F-sensitive oligosaccharide chains. Most of this 175-kDa mature form appeared on the cell surface 2 h following synthesis and showed a half-life of approximately 3 h. In the presence of a growth factor(s) partially purified from bovine kidney, the half-life of this 175-kDa normal neu gene product was shortened to less than 30 min. In B104-1-1 cells (NIH 3T3 cells transfected with neu gene activated oncogenically by a point mutation that changes a valine residue to a glutamic acid residue in the putative transmembrane region), the oncogenically activated neu gene product was also synthesized as a 170-kDa precursor with endoglycosidase H-sensitive oligosaccharide chains. However, this 170-kDa precursor diminished very fast and was only partially processed to a 185-kDa mature form which exhibited a half-life of less than 30 min. The 185-kDa activated neu gene product possessed an unidentified post-translational modification in addition to N-linked oligosaccharide chains. Both the precursor and mature forms of the mutationally activated neu gene product showed increased tyrosine-specific phosphorylation as compared with those of their normal counterparts in DHFR/G-8 cells. The mutationally activated neu gene product in B104-1-1 cells shared several features which have been reported previously for the ligand-activated platelet-derived growth factor receptor in v-sis- or c-sis-transformed cells. These properties include: 1) accelerated turnover of the precursor and mature forms compared with the rates of turnover of its normal counterparts, 2) insensitivity of this rapid turnover to lysosomotropic amines, and 3) increased in vivo tyrosine-specific phosphorylation of both the precursor and mature forms. These findings suggest that the mutationally activated neu gene product may transform the cells by mimicking ligand-induced activation.     

The processing and intracellular transport of myeloperoxidase. Modulation by lysosomotropic agents and monensin

Myeloperoxidase, stored in azurophil granules of neutrophils, is synthesized in promyelocytes as a larger molecular weight precursor, which is processed to yield a transient Mr 82 000 intermediate and mature polypeptides with molecular weights of 62 000 and 12 000. We have tried to define subcellular sites for processing using metabolic labelling of the promyelocytic leukemia cell line HL-60 in combination with subcellular fractionation on a Percoll gradient. A reasonable separation was achieved between azurophil granules, Golgi elements and endoplasmic reticulum. The finding of almost exclusively fully processed myeloperoxidase in granules and a mixture of unprocessed and processed polypeptide in fractions enriched in Golgi elements suggests that processing occurred mainly in pregranular structures. Monensin, which exchanges protons for Na+, and the base chloroquine blocked processing probably by inhibition of transport through the Golgi apparatus. However, the lysosomotropic NH4+ cation did not inhibit processing or transport indicating that processing is not necessarily influenced by pH-dependent mechanisms. Results from digestion with endoglycosidase H, incubation with tunicamycin and metabolic labelling with [3H]mannose indicated that myeloperoxidase contained high mannose oligosaccharide side chains. Also [32P]phosphate incorporated into Mr 90 000 and Mr 62 000 myeloperoxidase was susceptible to endoglycosidase H indicating that oligosaccharide side chains are modified by phosphorylation as in lysosomal enzymes. Thus, even if myeloperoxidase contained mannose 6-phosphate residues, these may not necessarily be involved in directing transport to the azurophil granules.     

Biosynthesis of a lysosomal enzyme. Partial structure of two transient and functionally distinct NH2-terminal sequences in cathepsin D

Various biosynthetic forms of porcine spleen cathepsin D (Erickson, A. H. and Blobel, G. (1979) J. Biol. Chem. 254, 11771-11774), isolated by immunoprecipitation of in vivo- and in vitro-synthesized products, have been characterized by partial NH2-terminal sequence analysis. Two short lived and functionally distinct NH2-terminal sequence extensions, a "pre" sequence and a "pro" sequence, have been detected. Both sequence extensions are present in preprocathepsin D which is the primary translation product immunoprecipitated after translation of porcine spleen mRNA in a wheat germ cell-free system. Preprocathepsin D is not glycosylated and has an approximate Mr = 43,000. Its 20-residue pre sequence resembles the signal sequences of presecretory proteins in abundance of Leu residues (7 out of 20 residues). Addition of dog pancreatic microsomal vesicles to the translation system resulted in the cleavage of the pre sequence and yielded segregated and glycosylated procathepsin D (Mr = 46,000) that was indistinguishable from its in vivo-synthesized counterpart detected after pulse-labeling of cultured porcine kidney cells. Some of this in vivo-synthesized procathepsin D was secreted and persisted as such in the culture medium. The remainder was converted within a period of 15 min to 2 h to single chain cathepsin D (Mr = 44,000) by removal of a pro sequence which was estimated to be 44 residues. Its partial sequence showed considerable sequence homology to the 44-residue activation peptide of pepsinogen. It is possible, therefore, that the prosequence of procathepsin D serves as an activation peptide that keeps the enzyme inactive during intracellular transport to the lysosome. The enzymatically active single chain form of cathepsin D undergoes further cleavage into a light and a heavy chain (Mr = 15,000 and 30,000, respectively) over a period of 2-24 h after synthesis. The oligosaccharide moieties of procathepsin D and of the single chain and heavy chain forms of cathepsin D are cleaved by endoglycosidase H. Treatment of cells with tunicamycin arrests the biosynthetic pathway of cathepsin D at procathepsin D. The nonglycosylated procathepsin D is not proteolytically processed and its secretion is greatly inhibited.     

Synthesis and processing of cellulase from ripening avocado fruit

The biosynthesis and processing of cellulase from ripening avocado fruit was studied. The mature protein is a glycoprotein, as judged by concanavalin A binding, with a molecular weight of 54,200. Upon complete deglycosylation by treatment with trifluoromethane sulfonic acid the mature protein has a molecular weight of 52,800 whereas the immunoprecipitated in vitro translation product has a molecular weight of 54,000. This result indicates that cellulase is synthesized as a large molecular weight precursor, which presumably possesses a short-lived signal peptide. A membrane-associated and heavily glycosylated form of the protein was also identified. This putative secretory precursor was enzymically active and the carbohydrate side chains were sensitive to endoglycosidase H cleavage. Results of partial endoglycosidase H digestion suggest that this precursor form of the mature glycoprotein possesses two high-mannose oligosaccharide side chains. The oligosaccharide chains of the mature protein were insensitive to endoglycosidase H cleavage, indicating that transport of the membrane-associated cellulase to the cell wall was accompanied by modification of the oligosaccharide side chains. The presence of a large pool of endoglycosidase H-sensitive membrane-associated cellulase (relative to an endoglycosidase H-insensitive form) suggest that transit of this protein through the Golgi is rapid relative to transit through the endoplasmic reticulum.     

Post-translational glycosylation of coronavirus glycoprotein E1: inhibition by monensin.

The intracellular sites of biosynthesis of the structural proteins of murine hepatitis virus A59 have been analyzed using cell fractionation techniques. The nucleocapsid protein N is synthesized on free polysomes, whereas the envelope glycoproteins E1 and E2 are translated on the rough endoplasmic reticulum (RER). Glycoprotein E2 present in the RER contains N-glycosidically linked oligosaccharides of the mannose-rich type, supporting the concept that glycosylation of this protein is initiated at the co-translational level. In contrast, O-glycosylation of E1 occurs after transfer of the protein to smooth intracellular membranes. Monensin does not interfere with virus budding from the membranes of the endoplasmic reticulum, but it inhibits virus release and fusion of infected cells. The oligosaccharide side chains of E2 obtained under these conditions are resistant to endoglycosidase H and lack fucose suggesting that transport of this glycoprotein is inhibited between the trans Golgi cisternae and the cell surface. Glycoprotein E1 synthesized in the presence of monensin is completely carbohydrate-free. This observation suggests that the intracellular transport of this glycoprotein is also blocked by monensin.     

Biosynthesis of high density lipoprotein by chicken liver: intracellular transport and proteolytic processing of nascent apolipoprotein A-1

To study the in vivo processing and secretion of Apolipoprotein A-I (Apo A-I), young chickens were administered individual L-[3H]amino acids intravenously and the time of intracellular transport of nascent Apo A-I from rough endoplasmic reticulum (RER) to the Golgi apparatus was measured. Within 3 to 9 min there was maximal incorporation of radioactivity into Apo A-I in both the RER and the Golgi cell fractions. By contrast, the majority of radioactive albumin was also present in the RER by 3 to 9 min, but did not reach peak amounts in the Golgi fraction until 9 to 25 min. Both radioactive Apo A-I and albumin appeared in the blood at about the same time (between 20 and 30 min). NH2-terminal amino acid sequence analysis of nascent intracellular Apo A-I showed that it contains a pro-hexapeptide extension identical to that of human Apo A-I. After 30 min of administration of radioactive amino acids radioactive Apo A-I was isolated by immunoprecipitation from the liver and serum. NH2-terminal sequence analysis of 20 amino acids indicated that chicken liver contained an equal mixture of nascent pro-Apo A-I and fully processed Apo A-I, whereas the serum only contained processed Apo A-I. Further studies showed that the RER only contained pro-Apo A-I, whereas a mixture of pro-Apo A-I and processed Apo A-I was found in the Golgi complex. These results indicate that, in chicken hepatocytes, there is a more rapid transport of Apo A-I than of albumin from the RER to the Golgi cell fractions, and that Apo A-I remains in the Golgi apparatus for a longer period of time before it is secreted into the blood. In addition these studies show that the in vivo proteolytic processing of chicken pro-Apo A-I to Apo A-I occurs in the Golgi cell fractions.     

Processing of the propeptide form of rat renal gamma-glutamyltranspeptidase

The biosynthesis of rat renal gamma-glutamyltranspeptidase (EC 2.3.2.2) was studied by sodium dodecyl sulfate gel electrophoresis and fluorography of specific immunoprecipitates obtained at varying times' postinjection with [35S]methionine. At 20 min postinjection 3 endo-beta-N-acetylglucosaminidase H-sensitive bands were observed representing the propeptide (Mr 75 000) large subunit (Mr 49 500) and small subunit (Mr 29 000) of transpeptidase. The alterations in Mr are consistent with removal of 6 N-linked coreoligosaccharides from the propeptide; 4 from the large subunit and 2 from the small subunit. All 3 bands became more diffuse and less endoglycosidase H-sensitive by 40 min and completely resistant by 60 min postinjection. At 20 h postinjection no propeptide remained. Thus, the primary propeptide cleavage reaction occurs prior to the loss of endoglycosidase H sensitivity while about 30% of the propeptide is processed along with the heterodimer and cleaved at a later time.     

Identification of the C3b receptor-binding domain in third component of complement

We report here that complement receptor type one (CR1) binds to a region of C3b that is contained within the NH2 terminus of the alpha' chain. In an enzyme-linked immunosorbent assay, CR1 bound to C3b, iC3b, and C3c but not to C3d, and this binding was inhibited by soluble C3b and C3c. Further attempts to generate a small C3 fragment capable of binding CR1 were unsuccessful. However, elastase degradation of C3 generated four species of C3c (C3c I-IV), two of which bound CR1. NH2-terminal sequence analysis and sodium dodecyl sulfate-gel electrophoresis of the C3cs indicated that the beta chains and the 40,000-dalton COOH-terminal alpha' chain fragments were identical; the NH2-terminal alpha' chain fragments of C3c I-IV varied from 21,000 to 27,000 daltons and accounted for the differential binding to CR1. C3c-I and II, which do not bind CR1, were missing 8 and 9 residues from the NH2 terminus of the alpha' chain when compared with the intact alpha' chain of C3b. C3c-III and IV, which bind CR1, had NH2 termini identical to the intact NH2-terminal alpha' chain of C3b. Using iodinated concanavalin A and endoglycosidase H, we showed that the NH2-terminal alpha' chains of C3c-I and III were glycosylated, while C3c-II and IV were not. Therefore, these data indicated that the amino terminus of the NH2-terminal alpha' chain fragment of C3c was responsible for binding CR1 while the COOH terminus of this fragment was not involved since the presence or absence of this region in C3c did not affect CR1 binding to C3c. Subsequently, two peptides were synthesized from the NH2-terminal alpha' chain fragment of C3c: X42, 42 residues in length from the NH2 terminus and C30, 30 residues in length from the COOH terminus. X42 inhibited binding of CR1 to C3b, and this effect was also observed with antipeptide antibodies against the X42 peptide. The C30 and other C3-derived peptides and antipeptide antibodies had no effect on the binding of CR1 to C3b.     

Loss of one asparagine-linked oligosaccharide from human transferrin receptors results in specific cleavage and association with the endoplasmic reticulum.

The coding sequence for the human transferrin receptor gene has been mutated in order to abolish the attachment of the oligosaccharide closest to the transmembrane sequence. Expression of the mutant receptor in Chinese hamster ovary cells resulted, after analysis under nonreducing conditions, in an 85- and a 73-kDa receptor species. After reduction, the 85-kDa receptors were mostly converted to 73 kDa, although with short labeling periods some 85-kDa receptor remained suggesting that the mutated gene was capable of coding for the entire polypeptide which was then proteolytically processed. This was supported by in vitro translation of mRNA from either wild type or mutant cells which in both cases yielded an 80-kDa protein, the full size of the nonglycosylated protein. The solubility characteristics of the mutant receptor suggest that it contains the COOH-terminal extracellular domain but not the transmembrane sequence. This is supported by the endoglycosidase H sensitivity of the 73-kDa protein which is compatible with the retention of two of the original three high mannose oligosaccharides. The receptor that lacks one oligosaccharide is unable to form intermonomer disulfide bridges or to migrate to the cell surface and is located in the endoplasmic reticulum where it is further degraded after a lag period of about 30 min.     

Resensitization of lutropin-desensitized tumour Leydig-cell adenylate cyclase with human erythrocyte membranes.

Purified rat tumour Leydig cells were pretreated with or without lutropin (1 h at 32 degrees C). The plasma membranes were then isolated and the adenylate cyclase activity measured in the presence of freshly prepared or heat-inactivated (1 h at 60 degrees C) human erythrocyte membranes. In plasma membranes from control cells in the presence of heat-inactivated human erythrocyte membranes both guanosine 5'-[beta, gamma-imido]triphosphate (p[NH]ppG) plus lutropin and NaF caused a 45--50-fold increase in cyclic AMP production over 30 min compared with 12--13 fold p[NH[ppG and 2--3-fold with lutropin alone. In plasma membranes isolated from lutropin-pretreated cells the NaF- and the p[NH]ppG-stimulated cyclic AMP production rates were unchanged, but no effect of lutropin could be demonstrated with or without added p[NH]ppG. However, after mixing lutropin-desensitized Leydig tumour-cell plasma membranes with freshly prepared human erythrocyte plasma membranes, the adenylate cyclase activity in the presence of lutropin, p[NH]ppG, lutropin plus p[NH]ppG and NaF were similar to those of control cell plasma membranes treated in the same manner. The possible mechanisms of this reversal of lutropin-induced desensitization by human erythrocytes are discussed.     

Detection and localization of a cytoplasmic domain on the beta-subunit of Na+,K+-ATPase. A monoclonal antibody study

A monoclonal antibody directed against the beta-subunit of dog kidney Na+,K+-ATPase was generated. Immunoblots demonstrate that monoclonal antibody III 18A binds exclusively to the denaturated beta-subunit. Binding experiments with membranes and whole cells reveal that III 18A binds to membranes but not to whole cells, indicating that the antibody binds to a cytoplasmic domain on the native beta-subunit. To localize the antibody-binding epitope, purified membrane-bound enzyme was fragmented by protease treatment. Tryptic digestion yields a 30-kDa fragment of the beta-subunit, which still retains the binding capacity for the antibody. Thus III 18A probably does not bind to the NH2-terminal segment of the protein. On the other hand, fragmentation of the beta-subunit with low concentrations of papain, which is known to yield a 40-kDa NH2-terminal and a 16-kDa COOH-terminal fragment, results in a complete loss of III 18A binding. These results suggest that the antibody-binding epitope is localized at or near a papain cleavage site on the COOH-terminal part of the beta-subunit. This is inconsistent with a structure model of the beta-subunit containing only a single transmembrane hydrophobic segment with a cytoplasmic NH2-terminal portion, but agrees quite well with a hypothetical structure with four intramembrane segments.     

Heparan sulphate-degrading endoglycosidase in liver plasma membranes.

An endoglycosidase is described in isolated liver plasma membranes that brings about a rapid and selective degradation of membrane-associated heparan sulphate, pre-labelled biosynthetically with Na2(35)SO4. The enzyme attacked mainly the polysaccharide chains of a hydrophobic membrane proteoglycan and it had little effect on a proteoglycan that could be displaced from the membranes with 1.0 M-NaCl. The highest activity was measured in the pH range 7.5-8.0, and the enzyme was almost completely inhibited below pH 5.5. Breakdown of susceptible polysaccharide chains was fast, being complete in 20-30 min. The major oligosaccharide fraction (Mr approx. 6000) produced by the enzyme was considerably smaller than the intact heparan sulphate chains. Enzyme activity was retained in membranes solubilized in 1% (v/v) Triton X-100. The high pH optimum and plasma-membrane association distinguish this enzyme from other heparan sulphate-degrading endoglycosidases that have acid pH optima and may be of lysosomal origin. A plasma-membrane endoglycosidase could modulate cellular interactions mediated by heparan sulphate, and/or release biologically active fragments of the polysaccharide from the cell periphery.     

Biosynthesis of the adenosine deaminase-binding protein in human fibroblasts and hepatoma cells

The adenosine deaminase-binding protein has previously been localized to the cell surface of human fibroblasts (Andy, R. J., and Kornfeld, R. (1982) J. Biol. Chem. 257, 7922-7925). In this study we examine the biosynthesis of binding protein in human fibroblasts, human hepatoma HepG2 cells, and a human kidney tumor cell line. Binding protein immunoprecipitated from radioiodinated detergent-extracted fibroblast membranes has a molecular weight of 120,000 when analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. An additional band of Mr 100,000 is also present which we believe is a result of proteolysis of the 120,000 band. Purified soluble kidney binding protein has an Mr of 112,000. Binding protein from fibroblasts pulse-labeled with [35S]methionine for 15 min migrates as a 110-kDa band on sodium dodecyl sulfate-polyacrylamide gels. Within 30-60 min of chase, the intensity of the 110-kDa band is diminished, and a 120-kDa band has appeared. Binding protein reaches the cell surface of fibroblasts within 30-60 min of chase. The same results are obtained with the other cell lines studied. Thus, binding protein is initially synthesized as a precursor of 110 kDa which chases into a 120-kDa mature form. The shift of 10 kDa is probably due to processing of its oligosaccharide chains since soluble kidney-binding protein contains 7-9 complex N-linked chains. Upon endoglycosidase H treatment, the 110,000 precursor shifts to a Mr of 89,000 while the 120,000 mature band shifts to 115,000, consistent with the presence of 7-9 high mannose chains on the precursor and 1-2 high mannose chains on the mature form. These results and the presence of complex N-linked chains on binding protein were confirmed by lectin affinity chromatography of glycopeptides derived from [2-3H]mannose-labeled binding protein. Analysis of [6-3H]glucosamine-labeled binding protein indicates the presence of 1 sialic acid residue per chain.