Synthesis and intracellular processing of aminooligopeptidase by human intestine

GASTROENTEROLOGY

1988;94:1426-31

Synthesis and Intracellular Processing of Aminooligopeptidase by Human Intestine TIMOTHY WARD

BURKE,

MARK

LLOYD,

VILJA

LORENZSONN,

and

OLSEN

Gastroenterology Research Laboratory, Middleton Veterans Medicine, University of Wisconsin, Madison, Wisconsin

Aminooligopeptidase is an intrinsic glycoprotein of the brush border membrane important for hydrolysis of the oligopeptide products of intraluminal protein digestion. To study its synthesis and intracellular processing, we performed pulse-chase experiments using [35S]methionine to label proteins of cultured human intestinal explants obtained by endoscopic biopsy. Aminooligopeptidase was isolated by immune precipitation with a monoclonal antibody and its molecular size was assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and fluorography. A precursor of relative molecular weight (Mr) 127,000 appeared within 10 min of chase and appeared’ to begin conversion to an &f, 150,000 form (the size of brush border membrane aminooligopeptidase) within 60 min. To determine if the change in molecular size was the consequence of alterations in glycosylation, we studied the susceptibility of the two forms to endo/3-N-acetylglucosaminidase H, which cleaves immature high-mannose iv-linked carbohydrate chains, and to peptide: iV4-(N-acetyl-P-glucosaminyl)asparagine amidase, which cleaves both the highmannose and complex iv-linked carbohydrate aminooligochains. Only the early M, 127,000 peptidase was sensitive to endo+iV-acetylglucosaminidase H, suggesting that the larger form results from trimming of high-mannose cores and adding terminal sugars in the Golgi complex. Both forms were sensitive to peptide:N4-(N-acetyl-P-glucosaminyl)asparagine amidase, generating an M, 114,000 species. The kinetics of the synthesis and processing of aminooligopeptidase and sucraseisomaltase were compared by immunoprecipitation of both proteins from the same tissue after separating the microvillous membrane from the remainder of the cellular membranes. Labeled aminooligopeptidase was present intracellularly in its mature form within 60 min and was detected exclusively in the brush border membrane by 90 min. Most of the

Hospital,

and Department

of

labeled sucrase-isomaltase pool had not yet undergone complex glycosylation during the same period. These data demonstrate that although human intestinal aminooligopeptidase undergoes N-linked glycosylation like sucrase-isomaltase, the synthesis of aminooligopeptidase differs from that of sucraseisomaltase in respect to the absence of a highmolecular-weight precursor and more rapid preGolgi processing.

A

minooligopeptidase, one of the major integral glycoproteins of the intestinal brush border membrane, is important in the final hydrolysis of products of intraluminal protein digestion. As the most abundant brush border peptidase, aminooligopeptidase has been extensively studied in nonhuman species (l-7). These studies have shown species differences in respect to the subunit form of the enzyme in the microvillous membrane [monomer in the case of rabbit (8) and homodimer in rat (9) and pig (lo)]. Despite these differences, in all species studied and additionally in cultured Caco-2 cells (1 l), aminooligopeptidase is initially synthesized as a molecule that approximates the size of the final subunit form in the brush border membrane. No large pro-forms have been identified with double active sites as in the precursor complexes of the intestinal disaccharidases, sucrase-isomaltase, and lactase-phlorizin hydrolase. Animal studies have shown that the intestinal brush border membrane hydrolases share common mechanisms in their synthesis and posttranslational modification (12,13). Despite this, the kinetics of posttranslational processing and transport vary, and it appears that indi-

Abbreviations used in this paper: endo H, endo-P-Nacetylglucosaminidase H; SDS, sodium dodecyl sulfate. 0 1988 by the American Gastroenterological Association 6616~5005/88/$3.56

SYNTHESIS OF HUMAN AMINOOLIGOPEPTIDASE

June 1988

1427

vidual intestinal hydrolases migrate to the brush border membrane at characteristic rates (11,14). Limited data are available describing the biosynthesis of human aminooligopeptidase (15). In the present study intestinal organ culture is used to characterize the synthesis and processing of the enzyme in normal human intestine. Additionally, the kinetics of posttranslational processing of human aminooligopeptidase and sucrase-isomaltase are compared. Materials

and Methods

Subjects Multiple duodenal biopsy specimens were obtained with a SF biopsy forceps during diagnostic endoscopy from 6 patients who had no evidence of small bowel disease. All patients had given written informed consent for participation in the study, approved by the University of Wisconsin Human Subjects Committee in December 1986. Specimens were transported to the laboratory in ice-cold oxygenated Krebs-Ringer bicarbonate buffer and were prepared for organ culture within 15 min.

Figure 1. Composite fluorogram of two pulse-chase experiments (left) compared with a protein-stained gel of immunoprecipitated aminooligopeptidase from purified human microvillous membrane (right). Aminooligopeptidase was initially recognized as an M, 127,000 molecule and within 60 min began conversion to an M,.150,000 form, similar in size to the protein in the brush border membrane.

Organ Culture Experiments Biopsy specimens were cut in half and mounted on stainless steel mesh according to the method of Browning and Trier (16). The specimens were placed in sterile organ culture dishes (Falcon, Oxnard, Calif.) and incubated in sterile culture media of Trowell’s T-8 (Gibco, Grand Island, N.Y.) with 10% heat-inactivated fetal calf serum and 10,000 U of penicillin and 0.01 g of streptomycin per 100 ml of media. Dishes were placed in stainless steel Torbal jars (Torsion Balance, Clifton, N.J.), gassed with 95% 025% COZ, and incubated at 37°C. Pulse-chase experiments were performed as described below. After incubating for a 1-h equilibration period, the grids were transferred to dishes containing 200-500 &i of [35S]methionine (New England Nuclear, Boston, Mass.) in culture media, returned to the Torbal jars, regassed, and allowed to incubate for 15 min. After the pulse period, the grids were rinsed, blotted, and then placed in culture dishes to incubate with nonradioactive media containing 2.5 mM methionine (Sigma, St. Louis, MO.). The final incubation was terminated at selected time points by rinsing the grids with iced saline and transferring the explants to -80°C storage. Time-course experiments as depicted in Figure 1 were performed three times, each with tissue from a different subject. Results were similar in each subject. Tissue was obtained from 3 other subjects and used in experiments terminated at specific time points as presented in Results.

Immunoprecipitation The explants were thawed and homogenized in 1.0 ml of phosphate-buffered saline containing 1% Triton X-100, pH 7.5, using 20 strokes of the Potter-Elvehjem tissue homogenizer, and allowed to stand on ice for 2 h,

then centrifuged at 27,000 g for 1 h. Aminooligopeptidase was immunoprecipitated from the supernatant by incubating overnight at 4°C with an excess of monoclonal antibody (H34FllF6). Collection of the precipitate was enhanced by exposure to Pansorbin (Calbiochem, La Jolla, Calif.) at 4°C for 30 min immediately before centrifugation. The immunoprecipitate was recovered by centrifugation for 5 min in a Microfuge 12 (Beckman, Fullerton, Calif.) and washed once with 0.3 M NaCl containing 0.1% sodium dodecyl sulfate (SDS), 0.05% Triton X-100, and 10 mM Tris-HCl, pH 6.8, then three times with 62.5 mM Tris, pH 6.8. Aminooligopeptidase was extracted with electrophoresis buffer containing 2% SDS and 5% 2-mercaptoethanol (BioRad, Richmond, Calif.) by boiling for 3 min, then applied to a 5% polyacrylamide gel for electrophoresis using the buffer system of Laemmli (17). Gels were developed with silver stain using the technique outlined by Heukeshoven and Dernick (18). Subsequently gels were treated with EN3HANCE (New England Nuclear), dried on filter paper, and exposed to Kodak X-OMAT AR film (Eastman Kodak, Rochester, N.Y.) for 3-7 days at -80°C to yield the completed fluorograms. Homogenization and immunoisolation in the presence of a cocktail of protease inhibitors containing antipain (1 pgiml final concentration), benzamidine (17.5 pgiml), pepstatin (1 pg/ml), aprotinin (10 pg/ml), phenylmethylsulfonyl fluoride (1 mM), and leupeptin (2 pg/ml) did not alter the relative molecular weight of either form of aminooligopeptidase as determined on SDS gels; thus, protease inhibitors were not usually included. For purposes of comparison, aminooligopeptidase was also immunoprecipitated from Tritonsolubilized brush border membranes. The membranes were a gift from Dr. K. Ramaswammy, Medical College of

1428 BURKE ET AL.

GASTROENTEROLOGYVol. 94, No. 6

Wisconsin, who prepared them from an organ transplant donor, using the method of Hopfer et al. (19). Aminooligopeptidase activity was assayed by the method of Porteous and Clark (20) and sucrase by the method of Dahlqvist (21). Separation of microvillous membranes from basolateral and intracellular membranes was achieved by magnesium chloride precipitation, a modification of the method by Kessler et al. (22). Immunoprecipitation of sucrase-isomaltase was performed as described above using specific monoclonal antibody (H40D4C5). Glycosidase

Experiments

Enzymatic probes that cleave oligosaccharide side chains from glycoproteins were used to investigate the glycosylation of aminooligopeptidase. In experiments using endo-p-N-acetylglucosaminidase H (endo H) (Miles Laboratories, Naperville, Ill.), immune pellets were boiled in 0.05 M sodium citrate containing 2% SDS, pH 5.5, for 5 min and centrifuged, and the supernatants were diluted in citrate buffer and mixed with endo H at a concentration of 80 mu/ml. The incubation mixture was kept at 37°C overnight. After digestion the protein was precipitated with 10% trichloroacetic acid, washed with acetone, and finally solubilized in sample buffer before SDS-polyacrylamide gel electrophoresis. Control specimens were treated identically except for deletion of the endo H. Experiments using peptide:N4-(N-acetyl-P-glucosaminyl)asparagine amidase (Genzyme, Boston, Mass.) were conducted by suspending the immune pellet in a solution of 9.5% SDS and 0.1 M 2-mercaptoethanol, boiling for 3 min, and then diluting the sample with 9.55 M sodium phosphate buffer, pH 8.6, and 7.5% NP-46 as described by the manufacturer. Peptide:N4-(N-acetyl+glucosaminyl)asparagine amidase was added to the sample at a concentration of 10 U/ml, and the mixture was incubated overnight at 37°C. After digestion the volume of the sample was brought up to 50 ~1 with electrophoresis sample buffer and the sample was analyzed by SDS-polyacrylamide gel electrophoresis and fluorography. Controls were treated similarly except for the deletion of the glycosidic enzyme. Monoclonal

Antibodies

Monoclonal antibodies were prepared according to the protocol (1985) of the Hybridoma Facility of the University of Wisconsin using a modification of the method of Galfre and Milstein (23). Briefly, a brush border membrane fraction was prepared from small intestinal mucosal scrapings of autopsy specimens (22). BALB/c mice were initially injected intraperitoneally with 10 mg of microvillar protein in 0.5 ml of saline mixed 1:l with Freund’s complete adjuvant (Calbiochem). Four booster immunizations of decreasing amounts of protein without adjuvant were performed .over 3 mo with an additional three injections during the week before fusion. On the day of fusion the mouse was killed by cervical dislocation and 10’ spleen cells were mixed with lo7 NS-1 mouse myeloma cells (P3-NSl-Ag4/1). Fusion was accomplished by adding 2 ml of 40% polyethylene glycol 1000 in Dulbecco’s modified Eagle’s medium (Gibco) at 37°C over a l-min period. Cells were incubated for 1 min and then

suspended in 30 ml of Dulbecco’s modified Eagle’s medium supplemented with hypoxanthine, thymidine, penicillin, streptomycin, and 29% fetal bovine serum (Gibco). The cells were sedimented over 5 min at 200 g, resuspended in 15 ml of supplemented Dulbecco’s modified Eagle’s medium, and placed in a 7% CO2 atmosphere at 37°C for 2 h. Subsequently, the cells were sedimented over 5 min at 200 g and resuspended in media with amethopterin and 0.5% mouse red blood cells. Hybrids were cloned and maintained with 6.5% mouse red blood cells added as necessary. Hybridomas were screened by enzyme-linked immunosorbent assay on microvillous membrane-coated microtiter plates with peroxidase-labeled goat-antimouse immunoglobulin G (heavy and light chain specific). Positive colonies were screened for antiaminooligopeptidase and antisucrase activities using a protein A-Sepharose (Sigma) immunoassay as described by Hauri (11). Colonies positive for more than one enzyme were excluded. Ascites prepared from positive valid clones was purified on MAPS II column (Biorad). Thirty micrograms of protein A-Sepharose-bound monoclonal immunoglobulin G (H34FllF6) removed 99% of aminooligopeptidase activity from a Triton X-199 solubilized microvillar membrane preparation with no detectable binding of sucrase, isomaltase, lactase, trehalase, or alkaline phosphatase activity.

Results The fluorograms demonstrated the appearance of an M, 127,000 aminooligopeptidase by 10 min of chase, with conversion to an M, 150,000 form beginning at 60 min (Figure 1). With the emergence of the latter form, there was a gradual disappearance of the smaller molecule, suggesting a productprecursor relationship. Sodium dodecyl sulfatepolyacrylamide gel electrophoresis of an immunoprecipitate from highly purified human intestinal microvillous membranes confirmed a major band at M, 150,000. To assess the state of glycosylation of the two molecular forms, susceptibility both to endo H, which cleaves immature, high-mannose N-linked carbohydrate chains from polypeptides (24,25), and to peptide:N4-(N-acetyl-P-glucosaminyl)asparagine amidase, which cleaves both the high-mannose and complex N-linked carbohydrate chains (261, was determined. Only the early M, 127,000 aminooligopeptidase showed enhanced migration (Mr 108,000) when digested with endo H (Figure 2), suggesting that the larger M, 150,000 form results from trimming of high-mannose cores and the addition of terminal sugars in the Golgi complex. Both forms were peptide:N4-(N-acetyl-P-glucosaminyl)asparagine amidase-sensitive, generating an Mr 114,000 species. To compare the kinetics of the synthesis and processing of aminooligopeptidase with sucrase-

June

Figure

1988

2. Fluorograms of aminooligopeptidase digested with glycosidic enzymes. The immunoprecipitate from 30 min of chase (A) shows accelerated migration after endo H treatment, whereas the immunoprecipitate from 80 min of chase (B) does not, findings characteristic of high-mannose and complex forms of glycoproteins, respectively. The mature form of the enzyme is sensitive to hydrolysis by peptide:N“-(N-acetyl-Pglucosaminyl)asparagine amidase (PNGase F) (C), which cleaves both high-mannose and complex Nlinked oligosaccharides.

isomaltase, we separated labeled tissue homogenates into an Mg2+-precipitable nonmicrovillar membrane fraction (combined endoplasmic reticulum, Golgi, and basosolateral membranes] and a microvillar membrane fraction. The two enzymes were both immunoprecipitated from each fraction at three separate time points. Mature aminooligopeptidase, which appeared intracellularly within 60 min of chase and in the brush border fraction within 90 min, was consistently detected earlier than the mature sucrase-isomaltase precursor. Figure 3 demonstrates complete conversion of aminooligopeptidase to complex form and transport of the enzyme to the brush border membrane before significant incorporation of label into the mature precursor form of sucrase-isomaltase.

SYNTHESIS

OF HUMAN

AMINOOLIGOPEPTIDASE

1429

cytoplasmic surface of the endoplasmic reticulum, where they undergo immediate membrane translocation and early (probably cotranslational) N-linked glycosylation to the high-mannose form of the molecule. Through a series of trimming reactions glucose and some mannose residues are removed and replaced by terminal sugars in the Golgi apparatus to form carbohydrate chains that characterize the complex forms of these enzymes. Intracellular and extracellular peptide cleavage has also been shown to play a role in the generation of the mature brush border membrane form of some glycoprotein hydrolases. Sucrase-isomaltase is cleaved after insertion into the brush border membrane (31,32) and lactasephlorizin hydrolase is believed to be cleaved intracellularly (28). Animal studies have shown that aminooligopeptidase is also a glycoprotein that undergoes both high mannose and complex N-linked glycosylation as part of its processing (1,3,33). Our studies show that the processing of human aminooligopeptidase parallels that seen in nonhuman species. The earliest identifiable aminooligopeptidase of M, 127,000 has already undergone high mannose glycosylation as evidenced by its sensitivity to glycosidic digestion with endo H, consistent with cotranslational processing of the molecule. The M, 108,000 molecule following endo H digestion probably approximates the size of the primary translation product. With subsequent modifications of the oligosaccharide side chains, the complex form of the

Discussion In most species studied, aminooligopeptidase is composed of identical subunits associated with one another in the brush border membrane as homodimers (13). Each subunit is inserted into the microvillous membrane by its own anchoring segment in contrast to the single hydrophobic attachment of the heterodimeric sucrase-isomaltase complex. Our data show that the subunits of human aminooligopeptidase are synthesized as individual glycoproteins. This is in agreement with animal studies and contrasts with the mechanism of synthesis of sucrase-isomaltase and lactase-phlorizin hydrolase, where subunits arise from the cleavage of large precursor molecules (27,28). The synthesis and processing of brush border membrane proteins have certain elements in common (29,30). All are thought to be synthesized on the

Figure

3. Fluorograms from the simultaneous immunoprecipitation of sucrase-isomaltase (SI, top) and aminooligopeptidase (AOP, bottom) obtained from Mg’+-precipitable intracellular membranes (left) and microvillar membranes (right) over increasing periods of chase. Mature aminooligopeptidase was detected earlier than the mature sucrase-isomaltase precursor in both the intracellular and brush border membrane fractions.

1430

BURKE ET AL.

molecule is generated and endo H sensitivity is lost. This development indirectly marks the time required for transport from the endoplasmic reticulum to the trans side of the Golgi region (34), -50 min for aminooligopeptidase synthesized in organ culture. There is no evidence that proteolysis plays a role in the processing of human aminooligopeptidase as the mature enzyme identified in the intracellular membrane compartment is of similar size to that seen in the brush border membrane. Despite qualitative similarities in the processing of all studied membrane glycoproteins, there appear to be major differences in the kinetics of those processes for individual glycosidases and peptidases. Danielsen and Cowell (14) compared the intracellular transport of labeled aminooligopeptidase and sucrase-isomaltase in cultured pig intestine. On the basis of appearance of the two enzymes in the high mannose and complex forms (as markers of their presence in the endoplasmic reticulum and Golgi apparatus, respectively), they concluded that aminooligopeptidase was transported to the Golgi 1.7 times faster than sucrase-isomaltase. Interestingly, the two enzymes had similar rates of post-Golgi transport. More recently, Hauri et al. (11)studied the processing of a variety of microvillar membrane hydrolases expressed in the Caco-2 human adenocarcinoma cell line. Aminooligopeptidase as well as dipeptidylpeptidase IV, angiotensin-converting enzyme, and p-aminobenzoic acid peptide hydrolase appeared to be transported to the Golgi region more rapidly than sucrase-isomaltase or lactase-phlorizin hydrolase. Their results suggested that there may be a fast transport mechanism for peptidases and a slower one for disaccharidases. They postulated that the membrane orientation of the particular hydrolase may dictate the efficiency of the transport process. Our results with simultaneous precipitation of human aminooligopeptidase and sucrase-isomaltase from Mg’+-precipitable and microvillous membrane fractions indicate a slower rate for pre-Golgi transport of the latter enzyme in human intestine also. Our experience with human lactase-phlorizin hydrolase (unpublished data) is similar, lending further credence to the concept that brush border membrane peptidases and disaccharidases have either dissimilar transport receptor systems or different affinities for a shared system. In conclusion, human intestine synthesizes aminooligopeptidase as an M, 108,000 (approximately) peptide that undergoes early (probably cotranslational) N-linked glycosylation to a transient M, 127,000form, which is converted to a mature M, 150,000glycoprotein by further processing of carbohydrate chains. The synthesis of human aminooligopeptidase differs from that of sucrase-isomaltase in

GASTROENTEROLOGY

Vol. 94, No. 6

that there is no high-molecular-weight precursor its pre-Golgi processing occurs more rapidly.

and

References 1. Danielsen EM, Skovberg H, Noren 0, Sjostrom H. Biosynthesis of intestinal microvillar proteins: nature of precursor forms of microvillar enzymes from Ca2+-precipitated enterocyte membranes. FEBS Lett 1981;132:197-200. 2. Danielsen EM, Sjostrom H, Noren 0, Bro B, Dabelsteen E. Biosynthesis of intestinal microvillar proteins: characterization of intestinal explants in organ culture and evidence for the existence of pro-forms of the microvillar enzymes. Biochem J 1982;202:647-54. 3. Danielsen EM. Biosynthesis of intestinal microvillar proteins: pulse-chase labeling of aminopeptidase N and sucraseisomaltase. Biochem J 1982;204:639-45. 4. Danielsen E, Noren 0, Sjostrom H. Biosynthesis of intestinal microvillar proteins: processing of aminopeptidase N by microsomal membranes. Biochem J 1983;212:161-5. 5. Danielsen E, Cowell G. Biosynthesis of intestinal microvillar membranes: further characterization of the intracellular processing and transport. FEBS Lett 1984;166:28-32. 6. Ahnen D, Santiago N, Cezard J-P, Gray G. Intestinal aminooligopeptidase: in vivo synthesis on intracellular membranes of rat jejunum. J Biol Chem 1982;257:12129-35. 7. Ahnen D, Mircheff A, Santiago N, Yoshioka C, Gray G. Intestinal surface aminooligopeptidase: distinct molecular forms during assembly on intracellular membranes in vivo. J Biol Chem 1983;258:5960-6. a. Feracci H, Maroux S. Rabbit intestinal aminopeptidase N: purification and molecular properties. Biochim Biophys Acta 1980;599:448-63. 9. Kim YS, Brophy EJ. Rat intestinal brush border membrane peptidases: solubilization, purification, and physicochemical properties of five different forms of enzyme. J Biol Chem 1976;251:3199-205. 10. Svensson B. Covalent cross-linking of porcine small intestinal aminopeptidase: subunit structure of the membrane-bound and the solubilized enzyme. Carlsberg Res Commun 1979;44:417-30. 11. Hauri H-P, Sterchi E, Bienz D, Fransen J, Marxer A. Expression and intracellular transport of microvillus membrane hydrolases in human intestinal epithelial cells. J Cell Biol 1985;101:838-51. E, Cowell G, Noren 0, Sjostrom H. Review article: 12. Danielsen biosynthesis of microvillar proteins. Biochem J 1984;221: 1-14. G. Anchoring and biosynthesis of stalked brush 13. Semenza border membrane proteins: glycosidases and peptidases of enterocytes and renal tubuli. Ann Rev Cell Biol 1986;2: 255-313. 14. Danielsen E, Cowell G. The intracellular transport of aminopeptidase N and sucrase-isomaltase occurs at different rates pre-Golgi but at the same rate post-Golgi. FEBS Lett 1985;190:69-72. H, Danielsen EM, Noren 0, Sjostrom H. Evidence 15. Skovberg for the biosynthesis of lactase-phlorizin hydrolase as a singlechain high-molecular weight precursor. Biochim Biophys Acta 1984;798:247-51. TH, Trier JS. Organ culture of mucosal biopsies of 16. Browning human small intestine. J Clin Invest 1969;48:1423-32. 17. Laemmli UK. Cleavage of structural proteins during the assembly of the head of the bacteriophage T4. Nature 1970;227:680-5. J, Dernick R. Simplified method for silver la. Heukeshoven

SYNTHESIS OF HUMAN AMINOOLIGOPEPTIDASE

June 1988

gels and the mechastaining of proteins in polyacrylamide nism of silver staining. Electrophoresis 1985;6:103-12. 19. Hopfer U, Nelson K, Perrotto J, Isselbacher KJ. Glucose transport in isolated brush border membrane from rat small intestine. J Biol Chem 1973;248:25-32. 20. Porteous JW, Clark B. The isolation and characterization of subcellular components of the epithelial cells of rabbit small intestine. Biochem J 1965;96:159-71. 21. Dahlqvist

A.

Assay

of

intestinal

G. A modified branes.

procedure vesicles

Biochim

23. Galfre

C, Milstein

strategies

for rapid

J Biol Chem

of efficiently

brush

border

mem34.

Enzymol and

antibodies:

1981;73:346. properties

of an

from Streptomyces

from

acetylglucosaminidase

TH, Maley specific

JH, Alexander

of intact

by endo-P-N-

1974;249:818-24.

S, Phelan

AW, Tarentino

of peptide:N-glycosidase

endo-P-N-acetylglucosaminidase

F activity

F preparations.

in

J Biol Chem

1984;259:107004. ML, Olsen

WA. A study

sucrase-isomaltase processing

deficiency:

of the enzyme.

28. Naim HY, Sterchi

29. Kenny

cells.

hydrolase Biochem

AJ, Maroux

of the molecular

pathology

a defect

intracellular

in the

N Engl J Med 1987;316:43842.

EE, Lentze

of lactase-phlorizin epithelial

and

intestine.

Physiol

Rev

1982;62:

griseus.

F. The release

glycoproteins

H. J Biol Chem

TH, Elder

Demonstration

27. Lloyd

of kidney

91-128. Lennarz W. Overview: role of intracellular membrane systems in glycosylation of proteins. Methods Enzymol 1983;98:91-7. Skovbjerg H. High molecular weight pro-sucrase-isomaltase in the human fetal intestine. Pediatr Res 1982;16:948-9. Hauri HP, Quaroni A, Isselbacher KJ. Biogenesis of intestinal plasma membrane: posttranslational route and cleavage of sucrase-isomaltase. Proc Nat1 Acad Sci USA 1979;76:5183-6. Danielsen E, Cowell G, Noren 0, Sjostrom H. Biosynthesis of intestinal microvillar proteins: effect of swainsonine on posttranslational processing of aminopeptidase N. Biochem J 1983;216:325-31. Griffiths G, Quinn P, Warren G. Dissection of the Golgi complex: monensin inhibits the transport of viral membrane proteins from medial to trans Golgi cisterni in baby hamster kidney cells infected with Semliki Forest virus. J Cell Biol 1983;96:835-50.

1974;249:811-7.

oligosaccharides

AL.

32.

M, Semenza

of monoclonal

F. Purification

AL, Plummer

26. Plummer

preparation

Methods

endo-P-N-acetyl-glucosaminid,ase 25. Tarentino

H, Muller

intestinal

C. Preparation

AL, Maley

31.

Anal

Acta 1978;506:136-54.

and procedures.

24. Tarentino

C, Murer

from small

Biophys

30.

33.

Biochem 1968;22:99-107. 22. Kessler M, Acute 0, Storelli transporting

disaccharidases.

hydrolases

1431

MJ. Biosynthesis in the human

and maturation small

intestinal

J 1987;241:427-34.

S. Topology

of microvillar

membrane

of

Received September 28, 1987. Accepted January 18, 1988. Address requests for reprints to: Ward A. Olsen, M.D., Medical Service, William S. Middleton Memorial Veterans Hospital, 2500 Overlook Terrace, Madison, Wisconsin 53705. This work was supported by grant AM-13927 from the National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases and by Veterans Administration Research Funds. Dr. Lloyd was supported by the University of Wisconsin Clinical Nutrition Center and National Institutes of Health grant 5P30-AM-26659 and by National Institutes of Health Clinical Investigator Award K08 DKol789-01. The authors thank Helen Korsmo and Rita Malinowski for technical assistance. They also thank Dr. K. Ramaswammy, Medical College of Wisconsin, Milwaukee, Wisconsin, for supplying highly purified human intestinal brush border membranes.