Tyrosine sulphation is not required for microvillar expression of intestinal aminopeptidase N.

The effect of 2,6-dichloro-4-nitrophenol (DCNP), an inhibitor of phenol sulphotransferases (EC 2.8.2.-), on the biosynthesis of aminopeptidase N (EC 3.4.11.2) was studied in organ-cultured pig intestinal mucosal explants. At 50 microM DCNP did not affect protein synthesis but it decreased incorporation of [35S]sulphate into aminopeptidase N and other major microvillar hydrolases by 70-85% compared with controls, indicating an inhibition of their post-translational tyrosine sulphation. In labelling experiments with [35S]methionine from 0.5 to 5 h, DCNP was tested for its possible influence on synthesis, processing and microvillar expression of aminopeptidase N, but no effect on any of these parameters could be detected. It can therefore be concluded that tyrosine sulphation is not required (for instance as a sorting signal) for the targeting of newly synthesized enzymes to the microvillar membrane.     

Biosynthesis of intestinal microvillar proteins: Further characterization of the intracellular processing and transport

The effect of tunicamycin on synthesis and intracellular transport of pig small intestinal aminopeptidase N (EC 3.4.11.2), sucrase-isomaltase (EC 3.2.1.48–10) and maltase-glucoamylase (EC 3.2.1.20) was studied by labelling of mucosal explants with [³⁵S]methionine. The expression of the microvillar enzymes was greatly reduced by tunicamycin but could be partially restored by leupeptin, suggesting the existence of a mechanism whereby newly synthesized, malprocessed enzymes are recognized and degraded. In the presence of tunicamycin, polypeptides likely to represent non-glycosylated forms of the enzymes persisted in the Mg²⁺-precipitated membrane fraction, indicating that high mannose glycosylation is essential for transport to the microvillar membrane. Treatment of aminopeptidase N and sucrase-isomaltase with endo F reduced the size of the high mannose forms approximately to those seen in the presence of tunicamycin. The complex forms were also sensitive to endo F but did not coincide with the high mannose forms after treatment, indicating that the size difference cannot alone be ascribed to processing of N-linked carbohydrate.     

Biosynthesis of intestinal microvillar proteins. Pulse-chase labelling studies on aminopeptidase N and sucrase-isomaltase.

The biogenesis of two microvillar enzymes, aminopeptidase N (EC 3.4.11.2) and sucrase (EC 3.2.1.48)-isomaltase (EC 3.2.1.10), was studied by pulse-chase labelling of pig small-intestinal explants kept in organ culture. Both enzymes became inserted into the membrane during or immediately after polypeptide synthesis, indicating that translation takes place on ribosomes attached to the rough endoplasmic reticulum. The earliest detectable forms of aminopeptidase and sucrase-isomaltase were polypeptides of Mr 140 000 and 240 000 respectively. These polypeptides were susceptible to treatment with endo-beta-N-acetylglucosaminidiase H (EC 3.2.1.96), suggesting that the microvillar enzymes during or immediately after completion of protein synthesis become glycosylated with a 'high-mannose' oligosaccharide structure similarly to other plasma-membrane and secretory proteins. After 20--40 min or 60--90 min of chase, respectively, aminopeptidase N and sucrase-isomaltase were reglycosylated to give the polypeptides of Mr 166 000 (aminopeptidase N) and 265 000 (sucrase-isomaltase). These were expressed at the microvillar membrane after 60--90 min. During the entire process of synthesis and transport to the microvillar membrane the enzymes were bound to membranes, indicating that the biogenesis of aminopeptidase N and sucrase-isomaltase occurs in accordance with the membrane flow hypothesis.     

Biosynthesis of intestinal microvillar proteins. Role of the Golgi complex and microtubules.

The effect of monensin and colchicine on the biogenesis of aminopeptidase N (EC 3.4.11.2), aminopeptidase A (EC 3.4.11.7), dipeptidyl peptidase IV (EC 3.4.14.5), sucrase (EC 3.2.1.48)-isomaltase (EC 3.2.1.10) and maltase-glucoamylase (EC 3.2.1.20) was studied in organ-cultured pig small-intestinal explants. On the ultrastructural level, monensin (1 microM) caused an increasingly extensive dilation and vacuolization of the Golgi complex during 4h exposure of the explants. On the molecular level, the effect of monensin was twofold. (1) The processing from the initial high-mannose-glycosylated form to the mature complex-glycosylated form was arrested. For some of the enzymes studied, intermediate stages between the high-mannose and complex forms could be seen, probably corresponding to 'trimmed' or partially complex-glycosylated polypeptides. (2) Labelled microvillar enzymes failed to reach their final destination. These findings suggest the involvement of the Golgi complex in the post-translational processing and transport of microvillar enzymes. The presence in the growth medium of colchicine (50 micrograms/ml) caused a significant inhibition of the appearance of newly synthesized enzymes in the microvillar membrane during a 3 h labelling period. Since synthesis and post-translational modification of the microvillar enzymes were largely unaffected by colchicine, the results obtained suggest that microtubules play a role in the final transport of the enzymes from the Golgi complex to the microvillar membrane.     

Biosynthesis of intestinal microvillar proteins. The effect of swainsonine on post-translational processing of aminopeptidase N.

The post-translational processing of pig small-intestinal aminopeptidase N (EC 3.4.11.2) was studied in organ-cultured mucosal explants. Exposure of the explants to swainsonine, an inhibitor of Golgi mannosidase II, resulted in the formation of a Mr-160000 polypeptide, still sensitive to endo-beta-N-acetylglucosaminidase H. Swainsonine caused only a moderate inhibition of transport of the enzyme through the Golgi complex and the subsequent expression in the microvillar membrane. This may imply that the trimming of the high-mannose core and complex glycosylation of N-linked oligosaccharides is not essential for the transport of aminopeptidase N to its final destination. A different type of processing was observed to take place in the presence of swainsonine, resulting in a considerable increase in apparent Mr (from 140000 to 160000). This processing could not be ascribed to N-linked glycosylation, since treatment of the Mr-160000 polypeptide with endo-beta-N-acetylglucosaminidase H only decreased its apparent Mr by 15000. The susceptibility of the mature Mr-166000 polypeptide, but not the Mr-140000 polypeptide, to mild alkaline hydrolysis suggests that aminopeptidase N becomes glycosylated with O-linked oligosaccharides during its passage through the Golgi complex. Aminopeptidase N was not labelled by [3H]palmitic acid, indicating that the processing of the enzyme does not include acylation.     

Biosynthesis of intestinal microvillar proteins. Pulse-chase labelling studies on maltase-glucoamylase, aminopeptidase A and dipeptidyl peptidase IV

The biogenesis of three intestinal microvillar enzymes, maltase-glucoamylase (EC 3.2.1.20), aminopeptidase A (aspartate aminopeptidase, EC 3.4.11.7) and dipeptidyl peptidase IV (EC 3.4.14.5), was studied by pulse-chase labelling of pig small-intestinal explants kept in organ culture. The earliest detectable forms of the enzymes were polypeptides of Mr 225000, 140000 and 115000 respectively. These were found to represent the enzymes in a 'high-mannose' state of glycosylation, as judged by their susceptibility to treatment with endo-beta-N-acetylglucosaminidase H (EC 3.2.1.96). After about 40-60 min of chase, maltase-glucoamylase, aminopeptidase A and dipeptidyl peptidase IV were further modified to yield the mature polypeptides of Mr 245000, 170000 and 137000 respectively, which were expressed at the microvillar membrane after 60-90 min of chase. The fact that the enzymes before reaching the microvillar membrane were found in a Ca2+-precipitated membrane fraction (intracellular and basolateral membranes), but not in soluble form, indicates that during biogenesis maltase-glucoamylase, aminopeptidase A and dipeptidyl peptidase IV are transported and assembled in a membrane-bound state.     

Biosynthesis of intestinal microvillar proteins. The intracellular transport of aminopeptidase N and sucrase-isomaltase occurs at different rates pre-Golgi but at the same rate post-Golgi

The kinetics of processing and microvillar expression of aminopeptidase N (EC 3.4.11.2) and sucrose alpha-D-glucohydrolase-oligo-1,6-glucosidase (sucrase-isomaltase, EC 3.2.1.48 and EC 3.2.1.10) were compared by labelling of pig small intestinal mucosal explants with [35S]methionine. The conversion from transient (high mannose glycosylated) to mature (complex glycosylated) form was 1.7-times slower for sucrase-isomaltase than for aminopeptidase N, indicating a slower rate of migration from the rough endoplasmic reticulum to the Golgi complex. Likewise, sucrase-isomaltase appeared in the microvillar fraction at a slower rate than aminopeptidase N. The relative pool sizes of mature and transient forms of both enzymes in intracellular membranes (Mg2+-precipitated fraction) were determined to obtain information on the relative time, spent pre- and post-Golgi, respectively, prior to microvillar expression. This ratio was 0.24 +/- 0.06 (mean +/- SD) for sucrase-isomaltase as compared to 0.40 +/- 0.04 (mean +/- SD) for aminopeptidase N. Considering the slower rate of pre-Golgi transport for sucrase-isomaltase, this indicates that the two microvillar enzymes have rather similar if not identical rates of post-Golgi transport.     

Biosynthesis of intestinal microvillar proteins. Effect of castanospermine on cell-free synthesis of aminopeptidase N

Pig small intestinal mRNA was translated in a rabbit reticulocyte lysate system supplemented with microsomal membranes. Castanospermine, an inhibitor of glucosidase I, induced a high mannose-glycosylated form of microvillar aminopeptidase N (EC 3.4.11.2) of increased molecular mass, indicating the blocked removal of glucose residues. In contrast to its reduced expression in a mucosal explant system [(1986) Biochem. J. 240, 777-782], this molecular form of aminopeptidase N was at least as abundant in cell-free translation as its normal high mannose-glycosylated counterpart, ruling out degradation taking place in the rough endoplasmic reticulum. Degradation of newly produced, malprocessed enzyme must therefore occur at a later stage during intracellular transport, presumably in the sarcoplasmic reticulum or in transitional elements between this organelle and the Golgi complex.     

Biosynthesis of intestinal microvillar proteins. Dimerization of aminopeptidase N and lactase-phlorizin hydrolase

The pig intestinal brush border enzymes aminopeptidase N (EC 3.4.11.2) and lactase-phlorizin hydrolase (EC 3.2.1.23-62) are present in the microvillar membrane as homodimers. Dimethyl adipimidate was used to cross-link the two [35S]methionine-labeled brush border enzymes from cultured mucosal explants. For aminopeptidase N, dimerization did not begin until 5-10 min after synthesis, and maximal dimerization by cross-linking of the transient form of the enzyme required 1 h, whereas the mature form of aminopeptidase N cross-linked with unchanged efficiency from 45 min to 3 h of labeling. Formation of dimers of this enzyme therefore occurs prior to the Golgi-associated processing, and the slow rate of dimerization may be the rate-limiting step in the transport from the endoplasmic reticulum to the Golgi complex. For lactase-phlorizin hydrolase, the posttranslational processing includes a proteolytic cleavage of its high molecular weight precursor. Since only the mature form and not the precursor of this enzyme could be cross-linked, formation of tightly associated dimers only takes place after transport out of the endoplasmic reticulum. Dimerization of the two brush border enzymes therefore seems to occur in different organelles of the enterocyte.     

Biosynthesis of intestinal microvillar proteins. Low temperature arrests both processing and intracellular transport

The effect of culture at 20 degrees C on biosynthesis of microvillar enzymes was studied in pig small intestinal mucosal explants. At this temperature, aminopeptidase N (EC 3.4.11.2) and sucrase-isomaltase (EC 3.2.1.48-10) both accumulated intracellularly, predominantly in their transient, high mannose-glycosylated form characteristic of the newly synthesized enzymes prior to the molecular processing taking place in the Golgi complex. The general morphology of the enterocyte was unaffected by culture at low temperature except for the Golgi complex where the cisternae appeared condensed and surrounded by numerous vesicles of 50 to 55 nm. Both molecular processing and microvillar expression could be restored by shifting the temperature to 37 degrees C. Culture at low temperature did not induce any missorting of newly synthesized aminopeptidase N, but both molecular processing and microvillar expression only resumed at a slow rate after increasing the temperature, suggesting that reorganization of the Golgi complex is a time-requiring process.     

Perturbation of intestinal microvillar enzyme biosynthesis by amino acid analogs. Evidence that dimerization is required for the transport of aminopeptidase N out of the endoplasmic reticulum

The amino acid analogs canavanine, 3-hydroxynorvaline, thialysine, 6-fluorotryptophan, m-fluorotyrosine, and 2-fluorophenylalanine were incorporated into proteins, synthesized in pig intestinal mucosal explants, and their effect on molecular processing and intracellular transport of microvillar enzymes studied. Unless they were used in combination, none of the analogs drastically reduced the expression of aminopeptidase N (EC 3.4.11.2) or sucrase-isomaltase (EC 3.2.1.48, EC 3.2.1.10), but to a varying extent, they all slowed the rate of transport to the apical surface. In contrast, the cellular export of a secretory protein, apolipoprotein A-1, was largely unaffected. For the microvillar enzymes, all six analogs caused an accumulation of the transient, high mannose-glycosylated form, indicating an analog-sensitive stage prior to the Golgi-associated processing. For aminopeptidase N, this arrest was shown to correlate with a reduced ability of its transient high mannose-glycosylated form to form homodimers as judged from cross-linking experiments, suggesting dimerization to be obligatory for transport out of the endoplasmic reticulum.     

Tyrosine sulfation, a post-translational modification of microvillar enzymes in the small intestinal enterocyte.

Protein sulfation in small intestinal epithelial cells was studied by labelling of organ cultured mucosal explants with [35S]-sulfate. Six bands in SDS-PAGE became selectively labelled; four, of 250, 200, 166 and 130 kd, were membrane-bound and two, of 75 and 60 kd, were soluble. The sulfated membrane-bound components were all enriched in the microvillar fraction but either absent or barely detectable in intracellular or basolateral membranes. Immunopurification of sucrase-isomaltase, maltase-glucoamylase, aminopeptidase N and aminopeptidase A showed that these microvillar enzymes become sulfated. Most if not all the sulfate was bound to tyrosine residues rather than to the carbohydrate of the microvillar enzymes, showing that this type of modification can occur on plasma membrane proteins as well as on secretory proteins.     

Biosynthesis of intestinal microvillar proteins. Forskolin reduces surface expression of aminopeptidase N

The effect of forskolin on the biosynthesis and intracellular transport of pig intestinal aminopeptidase N (EC 3.4.11.2) was studied in organ cultured mucosal explants. The drug which activates adenylate cyclase and hence the cAMP-dependent glycogenolytic pathway did not affect the explant content nor microvillar enrichment of the enzyme. Forskolin, however, caused a decrease in the microvillar expression of aminopeptidase N which developed in a time-dependent manner from about 40% by 80 min to 80% by 4 h of labeling. The intracellular pool size of the transient, high mannose glycosylated form of aminopeptidase N was unaffected by forskolin, indicating a normal synthesis in the rough endoplasmic reticulum. The decrease in surface expression is therefore caused by an induced posttranslational degradation of the enzyme, most likely taking place in the Golgi complex. The degradatory effect on newly synthesized aminopeptidase N was not accompanied by any morphological alterations of the enterocyte; in particular, the microvillar membrane appeared entirely unaffected by forskolin. The results obtained provide evidence for the existence of a posttranslational mechanism, whereby a polarized cell is capable of regulating its expression of apical proteins.     

Proteins of the kidney microvillar membrane. Biosynthesis of endopeptidase-24.11, dipeptidylpeptidase IV and aminopeptidases N and A in pig kidney slices.

An organ culture employing slices of renal-cortex tissue from piglets of the Yucatan strain was used to study the biogenesis of four microvillar peptidases: endopeptidase-24.11 (EC 3.4.24.11), dipeptidyl peptidase IV (EC 3.4.14.5), aminopeptidase N (EC 3.4.11.2) and aminopeptidase A (EC 3.4.11.7). The viability of the culture system was confirmed by the preservation of ultrastructural integrity and by an unchanged uptake of [3H]alanine into cells during the period of the experiments. After labelling with [35S]methionine, treatment with Mg2+ yielded two fractions, one containing microvilli and another, the Mg2+ pellet, containing intracellular and basolateral membranes. The labelled forms of the peptidases, isolated by immunoprecipitation, were analysed by sodium dodecyl sulphate/polyacrylamide-gel electrophoresis and fluorography. The Mg2+ pellet contained the earliest detectable forms of the enzymes. In each case, a polypeptide of lower Mr than the mature form and sensitive to treatment with endo-beta-N-acetylglucosaminidase H was the first form to be detected. These high-mannose forms were followed, about 30 min after the pulse, by a complex glycosylated form of higher Mr. Only the latter form was observed in microvilli and then only after 90 min of the chase period. A quantitative study of dipeptidyl peptidase IV showed that the forms observed in the Mg2+ pellet were precursors of those in the microvillar fraction. No labelled forms were observed in the cytosol. All four peptidases were thus synthesized within membrane compartments and glycosylated in two steps before assembly in microvilli.     

Immunomicroscopic localization of aminopeptidase N in the pig enterocyte. Implications for the route of intracellular transport

The subcellular localization of aminopeptidase N (EC 3.4.11.2) in the pig enterocyte was investigated by immunofluorescence and immunoelectron microscopy (immunogold staining). By indirect immunofluorescence on either frozen or paraffin-embedded sections, a very intense staining in the microvillar membrane and a weak intracellular staining was demonstrated. No staining was detected in the basolateral membrane. Likewise, the immunogold labelling on Epon-embedded sections was concentrated in the microvillar membrane, whereas the basolateral membrane did not contain significant amounts of labelling. Labelling was demonstrated in the Golgi apparatus and in a minor fraction of the intracellular smooth vesicles positioned between the Golgi apparatus and the microvillar membrane. These observations are compatible with the view that newly synthesized aminopeptidase N is delivered directly to the microvillar membrane by smooth vesicles having a diameter about 70 to 100 nm and does not pass the basolateral membrane on its way to the brush border membrane.     

Biosynthesis of intestinal microvillar proteins. Evidence for an intracellular sorting taking place in, or shortly after, exit from the Golgi complex

Pig small intestinal mucosal explants, labelled with [35S]-methionine, were fractionated into Mg2+-precipitated (intracellular and basolateral) and microvillar membranes, and the orientation of newly synthesized aminopeptidase N (EC 3.4.11.2) in vesicles from the two fractions was studied by its accessibility to proteolytic cleavage. The mature polypeptide of Mr 166 000 from the latter fraction was cleaved by trypsin, proteinase K and papain, consistent with an extracellular location of the enzyme at its site of function. In contrast, both the mature form and the transient form of Mr 140 000 from the Mg2+-precipitated fraction were equally well protected from proteolytic cleavage (in the absence of Triton X-100). This indicates that the basolateral plasma membrane is unlikely to be involved in the post-Golgi transport of newly synthesized aminopeptidase N and suggests instead a direct delivery of the enzyme to the apical plasma membrane. A crude membrane preparation from labelled explants was used in immunoelectrophoretic purification of membranes to determine at what stage during intracellular transport newly synthesized microvillar enzymes are sorted, i.e., accumulated in areas of the membrane from where other proteins are excluded. The transient form of aminopeptidase N was only moderately enriched by immunopurification, using antibodies against different microvillar enzymes, but the mature form was enriched approximately 30-fold from explants, labelled for 30 min. This suggests that for microvillar enzymes, the aspects of sorting studied take place in, or shortly after exit from, the Golgi complex.     

Folding of intestinal brush border enzymes. Evidence that high-mannose glycosylation is an essential early event.

A polyvalent antiserum which precipitates the native, folded, but not the denatured molecular forms of pig intestinal aminopeptidase N (EC 3.4.11.2) and sucrase-isomaltase (EC 3.2.1.48, EC 3.2.1.10) was used to determine the kinetics of polypeptide folding of the two newly synthesized brush border enzymes. In pulse-labeled mucosal explants, complete synthesis of the polypeptide chains of aminopeptidase N and sucrase-isomaltase required about 2 and 4 min, respectively, whereas maximal antiserum precipitation was acquired with half-times of 4-5 and 8 min, respectively. Fructose, which induces a defective cotranslational high-mannose glycosylation, increased the half-time of polypeptide folding to about 12 min for aminopeptidase N as well as for sucrase-isomaltase. Short-pulse experiments suggested that fructose exerts its effect by slowing the rate of glycosylation, making this partially a posttranslational process. In the presence of fructose, not only the malglycosylated forms but also the electrophoretically normal, high-mannose-glycosylated form of the brush border enzymes were retained in the endoplasmic reticulum and proteolytically degraded. The results obtained demonstrate an intimate interrelationship between glycosylation and polypeptide folding in the synthesis of membrane glycoproteins and, more specifically, indicate that the timing of these two early biosynthetic events is essential for correct polypeptide folding.     

Proteins of the kidney microvillar membrane. Effects of monensin, vinblastine, swainsonine and glucosamine on the processing and assembly of endopeptidase-24.11 and dipeptidyl peptidase IV in pig kidney slices.

The effects of various inhibitors were studied on the biogenesis of endopeptidase-24.11 (EC 3.4.24.11) and dipeptidyl peptidase IV (EC 3.4.14.5) in slices of renal cortex, from piglets of the Yucatan strain, maintained in organ culture. These microvillar peptidases were synthesized within membrane compartments and underwent glycosylation to yield high-mannose and complex forms [the preceding paper, Stewart & Kenny (1984) Biochem. J. 224, 549-558]. Monensin caused very gross ultrastructural changes in the proximal-tubular cells, resulting from distension of the Golgi sacs. It blocked the processing of the high-mannose to the complex glycosylated forms of the peptidases and prevented their assembly in the microvillar membrane. Swainsonine, an inhibitor of alpha-mannosidase II, generated new 'hybrid' forms of the proteins, intermediate in Mr between the high-mannose and the complex forms, but did not prevent assembly of the hybrid forms in microvilli. Vinblastine, an agent that affects microtubules, delayed, but did not abolish, either the processing or the transport to microvilli. Glucosamine interfered with the initial glycosylation reactions and generated heterogeneous sets of partially glycosylated polypeptides of lower Mr than the high-mannose forms. These results are discussed in relation to the site and mechanism of glycosylation and the involvement of the Golgi complex and microtubules in the biogenesis of these membrane peptidases.     

Investigation of the physical properties of dog intestinal microvillar membrane proteins by polyacrylamide gel electrophoresis: a comparison between normal dogs and dogs with exocrine pancreatic insufficiency

Procedures have been validated for the investigation of the physical properties of canine microvillar membrane proteins by SDS-polyacrylamide gel electrophoresis. These have been used to examine mucosal samples from eight control dogs and from five dogs with naturally occurring exocrine pancreatic insufficiency (EPI) in order to evaluate the potential role of the pancreas in the normal turnover of microvillar membrane proteins in the dog. Gel scanning showed that the proportion of total membrane protein in bands corresponding to a molecular mass greater than 200 kDa was up to 20-times higher in dogs with EPI than in control dogs. In particular, a band of apparent molecular mass 218 kDa represented between 8 and 28% of membrane protein in all affected dogs, compared with only 0.5 to 1.8% in controls, and is most likely to contain single chains of both pro-maltase-glucoamylase and pro-sucrase-isomaltase. Incubation of microvillar membranes in vitro with either trypsin or canine pancreatic juice resulted in degradation of this high molecular mass band and a corresponding increase in the amount of protein in three bands representing molecular masses of 150, 133 and 106 kDa. In samples from control dogs aminopeptidase N was identified in the 133 kDa band by Western blotting and incubation with monospecific antiserum. These findings suggest that pancreatic enzymes play a major role in the normal post-translational processing of intestinal microvillar membrane proteins in the dog.     

Temporal association of the N- and O-linked glycosylation events and their implication in the polarized sorting of intestinal brush border sucrase-isomaltase, aminopeptidase N, and dipeptidyl peptidase IV.

The temporal association between O-glycosylation and processing of N-linked glycans in the Golgi apparatus as well as the implication of these events in the polarized sorting of three brush border proteins has been the subject of the current investigation. O-Glycosylation of pro-sucrase-isomaltase (pro-SI), aminopeptidase N (ApN), and dipeptidyl peptidase IV (DPPIV) is drastically reduced when processing of the mannose-rich N-linked glycans is blocked by deoxymannojirimycin, an inhibitor of the Golgi-located mannosidase I. By contrast, O-glycosylation is not affected in the presence of swainsonine, an inhibitor of Golgi mannosidase II. The results indicate that removal of the outermost mannose residues by mannosidase I from the mannose-rich N-linked glycans is required before O-glycosylation can ensue. On the other hand, subsequent mannose residues in the core chain impose no sterical constraints on the progression of O-glycosylation. Reduction or modification of N- and O-glycosylation do not affect the transport of pro-SI, ApN, or DPPIV to the cell surface per se. However, the polarized sorting of two of these proteins, pro-SI and DPPIV, to the apical membrane is substantially altered when O-glycans are not completely processed, while the sorting of ApN is not affected. The processing of N-linked glycans, on the other hand, has no influence on sorting of all three proteins. The results indicate that O-linked carbohydrates are at least a part of the sorting mechanism of pro-SI and DPPIV. The sorting of ApN implicates neither O-linked nor N-linked glycans and is driven most likely by carbohydrate-independent mechanisms.