Protein oligomerization modulates raft partitioning and apical sorting of GPI-anchored proteins

An essential but insufficient step for apical sorting of glycosylphosphatidylinositol (GPI)-anchored proteins (GPI-APs) in epithelial cells is their association with detergent-resistant microdomains (DRMs) or rafts. In this paper, we show that in MDCK cells both apical and basolateral GPI-APs associate with DRMs during their biosynthesis. However, only apical and not basolateral GPI-APs are able to oligomerize into high molecular weight complexes. Protein oligomerization begins in the medial Golgi, concomitantly with DRM association, and is dependent on protein-protein interactions. Impairment of oligomerization leads to protein missorting. We propose that oligomerization stabilizes GPI-APs into rafts and that this additional step is required for apical sorting of GPI-APs. Two alternative apical sorting models are presented.     

Oligomerization is a specific requirement for apical sorting of glycosyl-phosphatidylinositol-anchored proteins but not for non-raft-associated apical proteins

Protein apical sorting in polarized epithelial cells is mediated by two different mechanisms, raft dependent and raft independent. In Madin-Darby canine kidney (MDCK) cells, an essential step for apical sorting of glycosyl-phosphatidylinositol (GPI)-anchored proteins (GPI-APs) is their coalescence into high-molecular-weight (HMW) oligomers. Here we show that this mechanism is also functional in Fischer rat thyroid cells, which possess a different sorting phenotype compared with MDCK cells. We demonstrate that, as in MDCK cells, both apical and basolateral GPI-APs associate with detergent-resistant microdomains, but that only the apical proteins are able to oligomerize into HMW complexes during their passage through the medial Golgi. We also show that oligomerization is a specific requirement for apical sorting of GPI-APs and is not used by transmembrane, non-raft-associated apical proteins.     

GPI-anchored proteins are directly targeted to the apical surface in fully polarized MDCK cells

The polarity of epithelial cells is dependent on their ability to target proteins and lipids in a directional fashion. The trans-Golgi network, the endosomal compartment, and the plasma membrane act as sorting stations for proteins and lipids. The site of intracellular sorting and pathways used for the apical delivery of glycosylphosphatidylinositol (GPI)-anchored proteins (GPI-APs) are largely unclear. Using biochemical assays and confocal and video microscopy in living cells, we show that newly synthesized GPI-APs are directly delivered to the apical surface of fully polarized Madin-Darby canine kidney cells. Impairment of basolateral membrane fusion by treatment with tannic acid does not affect the direct apical delivery of GPI-APs, but it does affect the organization of tight junctions and the integrity of the monolayer. Our data clearly demonstrate that GPI-APs are directly sorted to the apical surface without passing through the basolateral membrane. They also reinforce the hypothesis that apical sorting of GPI-APs occurs intracellularly before arrival at the plasma membrane.     

A Highlights from MBoC Selection: Apical sorting of lysoGPI-anchored proteins occurs independent of association with detergent-resistant membranes but dependent on their N-glycosylation

Most glycosylphosphatidylinositol-anchored proteins (GPI-APs) are located at the apical surface of epithelial cells. The apical delivery of GPI-APs is believed to result from their association with lipid rafts. We find that overexpression of C-terminally tagged PGAP3 caused predominant production of lysoGPI-APs, an intermediate precursor in the GPI lipid remodeling process in Madin–Darby canine kidney cells. In these cells, produced lysoGPI-APs are not incorporated into detergent-resistant membranes (DRMs) but still are delivered apically, suggesting that GPI-AP association with DRMs is not necessary for apical targeting. In contrast, apical transport of both fully remodeled and lyso forms of GPI-APs is dependent on N-glycosylation, confirming a general role of N-glycans in apical protein transport. We also find that depletion of cholesterol causes apical-to-basolateral retargeting not only of fully remodeled GPI-APs, but also of lysoGPI-APs, as well as endogenous soluble and transmembrane proteins that would normally be targeted to the apical membrane. These findings confirm the essential role for cholesterol in the apical protein targeting and further demonstrate that the mechanism of cholesterol-dependent apical sorting is not related to DRM association of GPI-APs.     

N-Glycosylation instead of cholesterol mediates oligomerization and apical sorting of GPI-APs in FRT cells

Sorting of glycosylphosphatidyl-inositol--anchored proteins (GPI-APs) in polarized epithelial cells is not fully understood. Oligomerization in the Golgi complex has emerged as the crucial event driving apical segregation of GPI-APs in two different kind of epithelial cells, Madin-Darby canine kidney (MDCK) and Fisher rat thyroid (FRT) cells, but whether the mechanism is conserved is unknown. In MDCK cells cholesterol promotes GPI-AP oligomerization, as well as apical sorting of GPI-APs. Here we show that FRT cells lack this cholesterol-driven oligomerization as apical sorting mechanism. In these cells both apical and basolateral GPI-APs display restricted diffusion in the Golgi likely due to a cholesterol-enriched membrane environment. It is striking that N-glycosylation is the critical event for oligomerization and apical sorting of GPI-APs in FRT cells but not in MDCK cells. Our data indicate that at least two mechanisms exist to determine oligomerization in the Golgi leading to apical sorting of GPI-APs. One depends on cholesterol, and the other depends on N-glycosylation and is insensitive to cholesterol addition or depletion.     

O-linked glycans mediate apical sorting of human intestinal sucrase-isomaltase through association with lipid rafts.

The plasma membrane of polarised epithelial cells is characterised by two structurally and functionally different domains, the apical and basolateral domains. These domains contain distinct protein and lipid constituents that are sorted by specific signals to the correct surface domain [1]. The best characterised apical sorting signal is that of glycophosphatidylinositol (GPI) membrane anchors [2], although N-linked glycans on some secreted proteins [3] and O-linked glycans [4] also function as apical sorting signals. In the latter cases, however, the underlying sorting mechanisms remain obscure. Here, we have analysed the role of O-glycosylation in the apical sorting of sucrase-isomaltase (SI), a highly polarised N- and O-glycosylated intestinal enzyme, and the mechanisms underlying this process. Inhibition of O-glycosylation by benzyl-N-acetyl-alpha-D-galactosaminide (benzyl-GalNAc) was accompanied by a dramatic shift in the sorting of SI from the apical membrane to both membranes. The sorting mechanism of SI involves its association with sphingolipid- and cholesterol-rich membrane rafts because this association was eliminated when O-glycosylation was inhibited by benzyl-GaINAc. The results demonstrate for the first time that O-linked glycans mediate apical sorting through association with lipid rafts.     

Apical transport and folding of prostate-specific membrane antigen occurs independent of glycan processing

Prostate-specific membrane antigen (PSMA) is an integral cell-surface membrane glycoprotein that is overexpressed in prostate carcinomas rendering it an appropriate target for antibody-based therapeutic strategies. The biosynthesis of PSMA in transfected COS-1 cells reveals a slow conversion of mannose-rich to complex glycosylated PSMA compatible with slow transport kinetics from the endoplasmic reticulum to the Golgi. Importantly, mannose-rich PSMA persists as a trypsin-sensitive protein throughout its entire life cycle, and only Golgi-located PSMA glycoforms acquire trypsin resistance. This resistance, used here as a tool to examine correct folding, does not depend on the type of glycosylation, because different PSMA glycoforms generated in the presence of inhibitors of carbohydrate processing in the Golgi are also trypsin resistant. The conformational transition of PSMA to a correctly folded molecule is likely to occur in the Golgi and does not implicate ER molecular chaperones, such as BiP. We show here that PSMA is not only heavily N-but also O-glycosylated. The question arising is whether glycans, which do not play a role in folding of PSMA, are implicated in its transport to the cell surface. Neither the cell-surface expression of PSMA nor its efficient apical sorting in polarized Madin-Darby canine kidney cells are influenced by modulators of N- and O-glycosylation. The acquisition of folding determinants in the Golgi, therefore, is an essential prerequisite for protein trafficking and sorting of PSMA and suggests that altered or aberrant glycosylation often occurring during tumorigenesis has no regulatory effect on the cell-surface expression of PSMA.     

Protein O-glycosylation in fungi: diverse structures and multiple functions

Protein glycosylation is essential for eukaryotic cells from yeasts to humans. When compared to N-glycosylation, O-glycosylation is variable in sugar components and the mode of linkages connecting the sugars. In fungi, secretory proteins are commonly mannosylated by protein O-mannosyltransferase (PMT) in the endoplasmic reticulum, and subsequently glycosylated by several glycosyltransferases in the Golgi apparatus to form glycoproteins with diverse O-glycan structures. Protein O-glycosylation has roles in modulating the function of secretory proteins by enhancing the stability and solubility of the proteins, by affording protection from protease degradation, and by acting as a sorting determinant in yeasts. In filamentous fungi, protein O-glycosylation contributes to proper maintenance of fungal morphology, hyphal development, and differentiation. This review describes recent studies of the structure and function of protein O-glycosylation in industrially and medically important fungi.     

Analysis of detergent-resistant membranes associated with apical and basolateral GPI-anchored proteins in polarized epithelial cells

Detergent-resistant membranes (DRMs) represent specialized membrane domains resistant to detergent extraction, which may serve to segregate proteins in a specific environment in order to improve their function. Segregation of glycosylphosphatidylinositol-anchored proteins (GPI-APs) in DRMs has been shown to be involved in their sorting to the apical membrane in polarized epithelial cells. Nonetheless, we have shown that both apical and basolateral GPI-APs associate with DRMs. In this report we investigated the lipid composition of DRMs associated with an apical and a basolateral GPI-AP. We found that apical and basolateral DRMs contain the same lipid species although in different ratios. This specific lipid ratio is maintained after mixing the cells before lysis indicating that DRMs maintain their identity after Triton extraction.     

Intestinal dipeptidyl peptidase IV is efficiently sorted to the apical membrane through the concerted action of N- and O-glycans as well as association with lipid microdomains.

The apical sorting of human intestinal dipeptidyl peptidase IV (DPPIV) occurs through complex N-linked and O-linked carbohydrates. Inhibition of O-linked glycosylation by benzyl-N-acetyl-alpha-d-galactosaminide affects significantly the sorting behavior of DPPIV in intestinal Caco-2 and HT-29 cells. However, random delivery to the apical and basolateral membranes and hence a more drastic effect on the sorting of DPPIV in both cell types is only observed when, in addition to O-glycans, the processing of N-glycans is affected by swainsonine, an inhibitor of mannosidase II. Together the data indicate that both types of glycosylation are critical components of the apical sorting signal of DPPIV. The sorting mechanism of DPPIV implicates its association with detergent-insoluble membrane microdomains containing cholesterol and sphingolipids, whereas an efficient association largely depends on the presence of a fully complex N- and O-linked glycosylated DPPIV. Interestingly, cholesterol is a more critical component in this context than sphingolipids, because cholesterol depletion by beta-cyclodextrin affects the detergent solubility and the sorting behavior of DPPIV more strongly than fumonisin, an inhibitor of sphingolipid synthesis.     

CD39 reveals novel insights into the role of transmembrane domains in protein processing, apical targeting and activity

Cargo proteins of the biosynthetic secretory pathway are folded in the endoplasmic reticulum (ER) and proceed to the trans Golgi network for sorting and targeting to the apical or basolateral sides of the membrane, where they exert their function. These processes depend on diverse protein domains. Here, we used CD39 (NTPdase1), a modulator of thrombosis and inflammation, which contains an extracellular and two transmembrane domains (TMDs), as a model protein to address comprehensively the role of native TMDs in folding, polarized transport and biological activity. In MDCK cells, CD39 exits Golgi dynamin-dependently and is targeted to the apical side of the membrane. Although the N-terminal TMD possesses an apical targeting signal, the N- and C-terminal TMDs are not required for apical targeting of CD39. Folding and transport to the plasma membrane relies only on the C-terminal TMD, while the N-terminal one is redundant. Nevertheless, both N- and C-terminal anchoring as well as genuine TMDs are critical for optimal enzymatic activity and activation by cholesterol. We conclude therefore that TMDs are not just mechanical linkers between proteins and membranes but are also able to control folding and sorting, as well as biological activity via sensing components of lipid bilayers.     

N-Linked oligosaccharides on the meprin A metalloprotease are important for secretion and enzymatic activity, but not for apical targeting

The alpha and beta subunits of meprins, mammalian zinc metalloendopeptidases, are extensively glycosylated; approximately 25% of the total molecular mass of the subunits is carbohydrate. The aim of this study was to investigate the roles of the N-linked oligosaccharides on the secreted form of mouse meprin A. Recombinant meprin alpha and mutants in which one of the 10 potential Asn glycosylation sites was mutated to Gln were all secreted and sorted exclusively into the apical medium of polarized Madin-Darby canine kidney cells, indicating that no specific N-linked oligosaccharide acts as a determinant for apical targeting of meprin alpha. Several of the mutant proteins had decreased enzymatic activity using a bradykinin analog as substrate, and deglycosylation of the wild-type protein resulted in loss of 75-100% activity. Some of the mutants were also more sensitive to heat inactivation. In studies with agents that inhibit glycosylation processes in vivo, tunicamycin markedly decreased secretion of meprin, whereas castanospermine and swainsonine had little effect on secretion, sorting, or enzymatic properties of meprin. When all the potential glycosylation sites on a truncated form of meprin alpha (alpha-(1-445)) were mutated, the protein was not secreted into the medium, but was retained within the cells even after 10 h. These results indicate that there is no one specific glycosylation site or type of oligosaccharide (high mannose- or complex-type) that determines apical sorting, but that core N-linked carbohydrates are required for optimal enzymatic activity and for secretion of meprin alpha.     

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.     

Polarised secretion of cytokines in primary human microvascular endothelial cells is not dependent on N-linked glycosylation

Human microvascular endothelial cells (HMVEC) grow in monolayers on Transwell filters and restrict permeability between the apical and basolateral media. We show that these cell monolayers are capable of sorting labelled endogenous proteins, including chemokines, growth factors and cytokines, to either the apical or basolateral media. IL-8 and GMCSF were secreted predominantly into the apical medium, whereas MIC-1 was secreted into the basolateral medium. This polarity did not correlate with glycosylation, as IL-8 and MIC-1 are both N-glycosylated, but were sorted to opposite sides of the cell. IL-6 is not glycosylated and did not display significant polarity in secretion. Similarly, the polarity of secretion of endogenous glycoproteins was not related to their glycosylation.     

N-聚糖和O-聚糖在极性化细胞蛋白质分拣中的作用

极性化上皮细胞的质膜因其所含蛋白质、脂质等组分不同,可以分为细胞膜顶端和细胞膜基底侧端两个区域,而新合成的蛋白质向这两个区域的有效分拣是上皮细胞维持其自身极性及正常功能所必需的。细胞膜基底侧端蛋白质的分拣主要由位于该蛋白质胞质区的信号肽所介导,关于这方面的研究是比较深入的;而细胞膜顶端蛋白质的分拣机制目前尚未阐明,因而显得比较复杂。近年来,糖类分子作为生物体内细胞识别和调控过程的信息分子日益受到关注,人们通过干扰聚糖合成、基因突变以及构建糖基化缺陷细胞株等实验方法,逐渐地认识到糖类分子在极性化上皮细胞的蛋白质分拣调节中起重要作用。由于糖分子本身结构非常复杂,而且目前缺乏研究糖类分子的有效手段,使得糖生物学的研究远远落后于蛋白质和核酸的研究。从而导致探讨糖类分子在蛋白质分拣过程的具体机制相对来说比较困难。本综述拟简要概括糖类分子中N-聚糖和O-聚糖在极性化上皮细胞的蛋白质分拣过程中的作用,以及两种聚糖在此过程中行使分拣信号功能的可能机制。     

Detergent-insoluble GPI-anchored proteins are apically sorted in fischer rat thyroid cells, but interference with cholesterol or sphingolipids differentially affects detergent insolubility and apical sorting

In contrast to Madin-Darby canine kidney cells, Fischer rat thyroid cells deliver the majority of endogenous glycosylphosphatidyl inositol (GPI)-anchored proteins to the basolateral surface. However, we report here that the GPI proteins Placental Alkaline Phosphatase (PLAP) and Neurotrophin Receptor-Placental Alkaline Phosphatase (NTR-PLAP) are apically localized in transfected Fischer rat thyroid cells. In agreement with the "raft hypothesis," which postulates the incorporation of GPI proteins into glycosphingolipids and cholesterol-enriched rafts, we found that both of these proteins were insoluble in Triton X-100 and floated into the lighter fractions of sucrose density gradients. However, disruption of lipid rafts by removal of cholesterol did not cause surface missorting of PLAP and NTR-PLAP, and the altered surface sorting of these proteins after Fumonisin B1 treatment did not correlate with reduced levels in Triton X-100 -insoluble fractions. Furthermore, in contrast to the GPI-anchored forms of both of these proteins, the secretory and transmembrane forms (in the absence of a basolateral cytoplasmic signal) were sorted to the apical surface without association with lipid microdomains. Together, these data demonstrate that the GPI anchor is required to mediate raft association but is not sufficient to determine apical sorting. They also suggest that signals present in the ectodomain of the proteins play a major role and that lipid rafts may facilitate the recognition of these signals in the trans-Golgi network, even though they are not required for apical sorting.     

Role of N- and O-glycans in polarized biosynthetic sorting

The maintenance of proper epithelial function requires efficient sorting of newly synthesized and recycling proteins to the apical and basolateral surfaces of differentiated cells. Whereas basolateral protein sorting signals are generally confined to their cytoplasmic regions, apical targeting signals have been identified that localize to luminal, transmembrane, and cytoplasmic aspects of proteins. In the past few years, both N- and O-linked glycans have been identified as apical sorting determinants. Glycan structures are extraordinarily diverse and have tremendous information potential. Moreover, because the oligosaccharides added to a given protein can change depending on cell type and developmental stage, the potential exists for altering sorting pathways by modulation of the expression pattern of enzymes involved in glycan synthesis. In this review, we discuss the evidence for glycan-mediated apical sorting along the biosynthetic pathway and present possible mechanisms by which these common and heterogeneous posttranslational modifications might function as specific sorting signals. glycosylation; epithelia; polarity; kidney; intestine     

The yeast SEC20 gene is required for N- and O-glycosylation in the Golgi. Evidence that impaired glycosylation does not correlate with the secretory defect

The Golgi plays a fundamental role in posttranslational glycosylation, transport, and sorting of proteins. The mechanism of protein transport through the Golgi has been seen as controversial in recent years. During the characterization of N-glycosylation-defective mutants (ngd) previously isolated by this laboratory, it was found that ngd20 is allelic to sec20. SEC20 was reported to be required for transport from endoplasmic reticulum to Golgi, but its precise function remains to be determined. We show now that SEC20 is also required for N- and O-glycosylation in the Golgi but not in the ER, in a cargo-specific manner, and that the glycosylation defect does not correlate with the secretory defect. By pulse-chase labeling experiments in combination with mannose linkage-specific antibodies, invertase and carboxypeptidase were found to be efficiently secreted to their final compartment, even upon shift to the nonpermissive temperature, while glycosylation in the Golgi was severely impaired. Using microsomal membranes isolated from ngd20, we found that mannosyl transfer from GDP-Man to various mannose-oligosaccharides, indicative for Golgi mannosylation, was strongly diminished. Analysis of the carbohydrate component of chitinase, an exclusively O-mannosylated protein, or of the bulk mannoprotein indicates that O-mannosylation is also reduced. The results demonstrate that in addition to secretion SEC20 also affects glycosylation in the Golgi, presumably because it exerts a more general role in maintenance and function of the Golgi compartments.     

Theoretical and experimental characterization of the scope of protein O-glycosylation in Bacteroides fragilis

Among bacterial species demonstrated to have protein O-glycosylation systems, that of Bacteroides fragilis and related species is unique in that extracytoplasmic proteins are glycosylated at serine or threonine residues within the specific three-amino acid motif D(S/T)(A/I/L/M/T/V). This feature allows for computational analysis of the proteome to identify candidate glycoproteins. With the criteria of a signal peptidase I or II cleavage site or a predicted transmembrane-spanning region and the presence of at least one glycosylation motif, we identified 1021 candidate glycoproteins of B. fragilis. In addition to the eight glycoproteins identified previously, we confirmed that another 12 candidate glycoproteins are in fact glycosylated. These included four glycoproteins that are predicted to localize to the inner membrane, a compartment not previously shown to include glycosylated proteins. In addition, we show that four proteins involved in cell division and chromosomal segregation, two of which are encoded by candidate essential genes, are glycosylated. To date, we have not identified any extracytoplasmic proteins containing a glycosylation motif that are not glycosylated. Therefore, based on the list of 1021 candidate glycoproteins, it is likely that hundreds of proteins, comprising more than half of the extracytoplasmic proteins of B. fragilis, are glycosylated. Site-directed mutagenesis of several glycoproteins demonstrated that all are glycosylated at the identified glycosylation motif. By engineering glycosylation motifs into a naturally unglycosylated protein, we are able to bring about site-specific glycosylation at the engineered sites, suggesting that this glycosylation system may have applications for glycoengineering.     

O-glycosylation as a sorting determinant for cell surface delivery in yeast

Little is known about the mechanisms that determine localization of proteins to the plasma membrane in Saccharomyces cerevisiae. The length of the transmembrane domains and association of proteins with lipid rafts have been proposed to play a role in sorting to the cell surface. Here, we report that Fus1p, an O-glycosylated integral membrane protein involved in cell fusion during yeast mating, requires O-glycosylation for cell surface delivery. In cells lacking PMT4, encoding a mannosyltransferase involved in the initial step of O-glycosylation, Fus1p was not glycosylated and accumulated in late Golgi structures. A chimeric protein lacking O-glycosylation motif was missorted to the vacuole and accumulated in late Golgi in wild-type cells. Exocytosis of this protein could be restored by addition of a 33-amino acid portion of an O-glycosylated sequence from Fus1p. Our data suggest that O-glycosylation functions as a sorting determinant for cell surface delivery of Fus1p.