Membrane domains and polarized trafficking of sphingolipids

Membrane domains and polarized trafficking of sphingolipids

seminars in CELL & DEVELOPMENTAL BIOLOGY, Vol. 12, 2001: pp. 149–161 doi:10.1006/scdb.2000.0232, available online at on Me...

1MB Sizes 10 Downloads 70 Views

seminars in CELL & DEVELOPMENTAL BIOLOGY, Vol. 12, 2001: pp. 149–161 doi:10.1006/scdb.2000.0232, available online at on

Membrane domains and polarized trafficking of sphingolipids Olaf Maier, Tounsia Aït Slimane and Dick Hoekstra∗

and signaling. In addition, a variety of disorders are known, that rely either on sphingolipid storage, which may originate from deficiencies of enzymes in their major biosynthetic pathways or from an accumulation of cholesterol as in Niemann–Pick disease, or on changes in sphingolipid expression, such as in tumor development. Furthermore, similar to proteins and the major class of lipids, phospholipids, sphingolipids also display an asymmetric or polarized distribution, marking the distinct identity of organelles, membranes and sub-domains within the lateral plane of the membrane. In particular the polarized distribution of glycosphingolipids (GSL) in polarized cells, such as epithelial cells and hepatocytes, has been well recognized. Thus, in these cells, distinct plasma membrane domains are distinguished, the apical and basolateral domains, which face the lumen of the organ and underlying cells and connective tissue, respectively. GSL appear especially enriched at the apical membrane surface. Also neuronal cells, such as neurons and oligodendrocytes reveal membrane polarity, and their axonal and myelin membrane domains are highly enriched in GSL. The ominous, yet distinctly polarized distribution of sphingolipids, including their localization in small sub-domains within a given membrane, the so-called ‘rafts’, in conjunction with their prominent cell biological properties have made these lipids attractive in studying: (i) how they acquire and maintain their polarized distribution; (ii) what the functional relevance is of this distribution; and (iii) how their distinct distribution may affect intraand intercellular communication. With regard to the latter, it is becoming apparent that sub-domains exist that are composed of different sets of sphingolipids, implying their potential importance as platforms that could express a multitude of yet unidentified functions. Clearly, the biogenesis, stability, of and communication between, sphingolipid domains are topics of major cell biological interest, which

The plasma membrane of polarized cells consists of distinct domains, the apical and basolateral membrane, that are characterized by a distinct lipid and protein content. Apical protein transport is largely mediated by (glyco)sphingolipid– cholesterol enriched membrane microdomains, so called rafts. In addition changes in the direction of polarized sphingolipid transport appear instrumental in cell polarity development. Knowledge is therefore required of the mechanisms that mediate sphingolipid sorting and the complexity of the trafficking pathways that are involved in polarized transport of both sphingolipids and proteins. Here we summarize specific biophysical properties that underly mechanisms relevant to sphingolipid sorting, cargo recruitment and polarized trafficking, and discuss the central role of a subapical compartment, SAC or common endosome (CE), as a major intracellular site involved in polarized sorting of sphingolipids, and in development and maintenance of membrane polarity. Key words: cell polarity / (glyco)sphingolipids / membrane transport / rafts / sorting c 2001 Academic Press

Introduction Although a relatively minor class of lipids, sphingolipids attract widespread attention because of their involvement in numerous functions relevant to cell development and biology, such as molecular sorting, intracellular transport, cell–cell contact,

From the Department of Membrane Cell Biology, University of Groningen, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands. *Corresponding author. E-mail: [email protected] c

2001 Academic Press 1084–9521 / 01 / 020149+ 13 / $35.00 / 0


O. Maier et al.

tively. 3 This process does not involve spontaneous transfer, 4 a feature often seen for short chain (C6) ceramide derivatives, 5 which are frequently used because such a derivative readily enters cells, where it is integrated into metabolic pathways and/or may trigger apoptosis. By contrast, natural ceramide does not readily leave the membrane, showing a half-time of interbilayer exchange in the order of days. 4 Accordingly, either a protein- and/or vesiclemediated transfer mechanism shuttles ceramide from ER to Golgi. Moreover, the occurrence of an ATP-dependent and -independent translocation step of ceramide, 6 resulting in a selective entry in the SM and GlcCer biosynthetic pathways, respectively, may imply that recruitment of the precursor takes place from different pools, involving an as yet unknown regulatory mechanism that signals or determines a need for the synthesis of either sphingolipid. GlcCer is at least in part translocated across the membrane into the Golgi lumen, thus allowing the biosynthesis of complex neutral and acidic glycolipids (i.e., sialic acid-containing glycolipids or gangliosides). 7 On reaching the trans-Golgi network (TGN), the lipids are packaged into vesicles, which mediate their transport to the plasma membrane. In polarized cells, pathways exist that may directly translocate newly synthesized sphingolipids from the TGN to their required domain, i.e., either the basolateral or the apical membrane. Alternatively, indirect pathways may also be operating, implying that prior to reaching the apical membrane, the lipid may be first transported to the basolateral membrane and is subsequently funneled into the transcytotic pathway to the apical surface. In addition, since the plasma membrane is subject of continuous exo- and endocytosis, sorting and redistribution of sphingolipids in the endosomal membrane system is also required to maintain the polarized structure. The endosomal trafficking pathways have been largely characterized by monitoring the flow of marker proteins after their endocytosis at either the apical or basolateral membrane (Figure 1, steps 2A and B). Internalized proteins first enter early sorting endosomes which are characterized by the small GTPase Rab 5 (Figure 1). The majority of the transferrin/transferrin receptor complex is recycling from basolateral early endosomes back to the plasma membrane (Figure 1, step 3B), whereas the LDL/LDL-receptor complex is sorted into the lysosome-directed degradation pathway (Figure 1, steps 4B and C). The polymeric immunoglobulin receptor (pIgR) as well as part of the transferrin receptor complex is transported

requires a basic understanding of their biology and developmentally regulated features in cells. Here, we will focus on recent progress made in this area, with a particular emphasis on sphingolipid flow and sorting during polarity development.

Occurrence of polarized trafficking A main feature of polarized epithelial cells is the formation of distinct plasma membrane domains, the apical and basolateral membrane, which are characterized by a specific protein and lipid content. Molecular randomization due to lateral movement of proteins and lipids between these membrane domains is prevented by the presence of tight junctions, acting as a fence between the domains. However, extensive transport of proteins and lipids between either plasma membrane domain and intracellular membranes, proceeding along the endocytic and biosynthetic pathways, occurs as well as transport between apical and basolateral membranes by means of transcytosis. These trafficking pathways are largely mediated by small transport vesicles. A detailed understanding at the molecular level as to how cells are able to maintain their polarized phenotype in spite of this constant membrane flux, including insight into the mechanisms of sorting and targeting of domain specific membrane compounds, is currently one of the major challenges in membrane cell biology. As noted, it has been well established that GSL are enriched in the apical domain of polarized epithelial cells. Indeed, newly synthesized glucosylceramide (GlcCer) and sphingomyelin (SM), the most simple sphingolipids that are synthesized from their common precursor ceramide to which either glucose or phosphorylcholine, respectively, is attached as head group, are found to be enriched in the exoplasmic leaflet of the apical and basolateral membrane, respectively, indicating that they are sorted prior to their delivery to the plasma membrane (Figure 1, steps 1A and B, see below). To properly appreciate polarized trafficking and underlying sorting mechanisms in sphingolipid flow, it is important to recognize the complexity of pathways that potentially mediate their transport. In recent years it has become apparent that ceramide is synthesized at the endoplasmic reticulum. 1,2 Subsequently, the precursor is translocated to early Golgi compartments for the biosynthesis of GlcCer and SM at the cytosolic site and within the lumen of the Golgi stack, respec150

Polarized sphingolipid trafficking

Figure 1. Schematic representation of transport pathways in hepatocytes. Newly synthesized lipids and proteins can use two pathways to reach the correct surface of polarized cells, a direct or an indirect one. In the direct route, lipids and proteins are sorted in the TGN and delivered directly to the apical or basolateral surface. In the indirect route, they are first sent to one surface and may then reach the opposite surface by transcytosis. This figure focuses on transport pathways for polarized trafficking of sphingolipids. Newly synthesized glucosylceramide (GlcCer), galactosylceramide (GalCer) and sphingomyelin (SM) are delivered directly from the TGN to the apical (bile canalicular, BC) and the basolateral plasma membrane respectively (steps 1A and B). Upon internalization, as visualized by monitoring the flow of fluorescent derivatives, sphingolipids are delivered to distinct AEE (step 2A) or BEE (step 2B) and transported to the SAC/CE (steps 5A and B). In this compartment, the sphingolipids are sorted, followed by vesicular transport of GlcCer to the BC (step 6A), while vesicular transport similarly carries SM and GalCer preferentially to the basolateral plasma membrane (step 6B). When cells have been treated with dbcAMP, SM is redirected from the SAC to the BC (steps 6C and C0 ). This translocation step includes trafficking via the rab-11 positive compartment (ARE/SIC), which is considered to represent a subcompartment of SAC (see Reference 16). Note that, in contrast to GlcCer (step 6A), delivery of SM to the BC requires intact microtubules (step 6C0 ). Some endocytosed proteins (see text for details) can recycle to their original plasma membrane domain from the AEE (step 3A) and BEE (step 3B), while others are delivered to the LE (steps 4A and B) and ultimately to the Lys (step 4C). In liver cells, several resident apical proteins have been shown to follow the transcytotic pathway, via SAC (steps 2B, 5B and 6C), but a direct pathway also exists (step 1A; see text). The dotted line indicates the possible direct connection between the TGN and the SAC. TJ: tight junction, AEE: apical early endosome, BEE: basolateral early endosome, SAC: sub-apical compartment, CE: common endosome, ARE: apical recycling endosome, SIC: sub-apical intermediate compartment, LE: late endosome, Lys: lysosome, the gray lines indicate the microtubule-dependent pathways, identified thus far.

step 6C). 10 Also basolaterally-localized SM and GlcCer can enter the endocytic pathway and reach the SAC via basolateral early endosomes (van IJzendoorn SCD, van der Wouden JM, Jonker M, and Hoekstra D, unpublished observations). This compartment is highly enriched in SM 11 and is characterized by the localization of various rab proteins that may be involved in the regulation of polarity-dependent transport. Moreover, the exclusive localization of rab 11, 17 and 25 (Figure 1) in specific subcompartments

to a compartment that is referred to as subapical compartment (SAC 8 ) or common endosome (CE), 9 representing a tubulovesicular compartment where endocytic transport originating from apical and basolateral membrane merge (Figure 1, step 5A and B). While the transferrin receptors are recycled to the basolateral membrane (Figure 1, step 6B), the pIgR/ligand complex enters the transcytotic pathway and is transported via the apical recycling endosomes (ARE) to the apical membrane (Figure 1, 151

O. Maier et al.

of the SAC, 12–14 in conjunction with observations on pH differences in the CE, 15 suggests that this compartment consists of functionally different subcompartments. 16 Evidence has been obtained which demonstrates that biosynthetic transport also, leaving the TGN, and directed towards either the apical or basolateral membrane can pass through SAC or CE 17 (Figure 1, ‘?’). The central role of this compartment in polarized trafficking is further emphasized by the notion that apical-to-basolateral transcytosis of SM and galactosylceramide (GalCer), as well as apical recycling of GlcCer takes place via SAC or CE, after transport through apical early endosomes (represented in Figure 1 as steps 2A, 5A, 6A and B). 18 Trafficking from or through compartments and the enrichment of particular lipids in distinct membrane domains or subdomains evidently requires sorting and at least a transient stability of these domains for functioning. Our current knowledge as to how this is accomplished for sphingolipids will be further described.

and cholesterol. These rafts constitute the core domains for the transport of apical proteins that are embedded in these microdomains, which eventually bud as transport vesicles from the TGN. 24,25 In yeast, the formation of sphingolipid enriched rafts mediating the transport of GPI-linked proteins, a class of proteins that shows a strong preference for partitioning in such microdomains, may occur already in the ER. 26 Nevertheless it is likely that in mammalian cells rafts functioning in the polarized sorting of proteins arise only in the TGN, because the Golgi shows a cis to trans gradient of cholesterol 27 and a critical sphingolipid/cholesterol ratio is required for stabilization of the liquid ordered domain (see below). Importantly, at low temperature, rafts are insoluble in detergents, a feature often used to demonstrate their presence, whereas they destabilize upon cholesterol depletion. 28 Although the existence of sphingolipid/cholesterolenriched microdomains in cellular membranes has been a matter of debate, 29 it is now well established that such domains not only exist in model membranes, but also in biological membranes (reviewed in Reference 30). In biological membranes, such domains presumably arise from a dynamic interplay between lipids and proteins, implying a transient stability. Thus, depending on the tightness of packing, molecules may or may not engage in dynamic interactions with the direct environment of the patch, allowing partitioning in and out of such domains. 31 There are a variety of parameters that will determine the nature of these domains, and in this regard the physico–chemical properties of the lipids should particularly be taken into account. Head group, acyl chains, and their overall shape will determine the core structure of these domains. For example, given its carbohydrate nature, GlcCer will have a greater tendency to engage in head group-mediated hydrogen bonding than SM. Moreover, since the latter has a more strongly hydrated head group, inter head group repulsion will dictate a less tight packing than in the case of GlcCer. Accordingly, lipids like cholesterol and membrane proteins may partition and/or be recruited to different degrees into such domains, which may have consequences for vesiculation and trafficking. Indeed, as has been revealed in both various sphingolipid storage disorders and Niemann–Pick C (NPC) disease, which relies on a defective functioning of the sterol-sensing NPC1 protein, perturbation of cholesterol homeostasis leads to altered sorting and/or accumulation of a variety of lipids and proteins. 32,33 In these cases an

Mechanisms of sphingolipid sorting Although early work suggested that newly-synthesized SM and GlcCer reach their respective domains by vesicle-mediated transport, following sorting in the TGN, later studies revealed that a considerable fraction of GlcCer can reach the surface by a vesicle-independent pathway after its synthesis at the cytoplasmic leaflet of the Golgi. 19–21 After delivery to the inner leaflet of the apical membrane, implying that sorting also occurs in this pathway, GlcCer gains access to the outer leaflet by monomeric lipid translocation via multidrug resistance proteins. 21 Sorting of GlcCer and SM might, therefore, be partly mediated by their distinct sites of synthesis in the Golgi. However, since, for example, newly synthesized SM and the ganglioside GM1 are transported predominantly to the basolateral and apical surface in a renal intercalated cell line, respectively, 22 it is evident that sphingolipids in the same membrane leaflet can be sorted laterally into distinct membrane patches prior to their delivery to the plasma membrane. While it is unclear how the enrichment of SM in basolateral-directed transport vesicles is accomplished, apical sorting relies on a widely held model that involves the assembly of so-called ‘rafts’, which are in essence microdomains of approximately 70–80 nm in size 23 consisting of (glyco)sphingolipids 152

Polarized sphingolipid trafficking

into the more solid GSL domains, driven by van der Waals interactions and hydrogen-bonding, which provides a basis for the liquid-ordered GSL-enriched microdomain formation, such as rafts. 36 Interestingly, for raft assembly per se, GSL are not required, but rather, SM and cholesterol, as stabilizers of raft structure, suffice, as revealed in a GSL-deficient melanoma cell line. 37 Furthermore, in developing neurons, upregulation of SM biosynthesis, while leaving GlcCer and cholesterol levels unaltered, induces the axonal sorting of a GPI-linked protein, Thy-1. 38 In the latter study, it was also shown that the ganglioside GM1, which randomly distributes in early-developing neurons, accumulates in the rafts arising from SM biosynthesis, thereby leading to enrichment of the ganglioside in the axonal membrane, and suggesting that rafts are capable of sorting sphingolipids. In contrast, to stimulate axonal growth ongoing GlcCer synthesis is required, which can be stimulated by basic fibroblast growth factor without affecting SM synthesis. 39 These data point to differences in regulation and hence differences in functioning of both sphingolipids in axonal development. Next to SM, cholesterol plays a prominent role in raft-mediated transport as its removal by extraction with cyclodextrin dissipates the raft and hence, apical delivery. Yet, although such treatment has been shown to impair surface directed transport of influenza virus hemagglutinin in MDCK cells, 28 this has not been seen in other studies where transport of a GPI-linked protein to the cell surface in MEB4 melanoma cells or its GSL-deficient derivative, GM95, were not affected upon cholesterol extraction, 37 implying that transport per se is not necessarily impaired after such treatments. This may indicate a raft-preference for the packing of less bulky proteins such as those linked to lipid (like)-anchors like GPI and fatty acids, which will more readily partition in a strongly liquid ordered environment than bulky transmembrane proteins. 40 One possibility of how microdomain formation may facilitate polarized sorting is that, depending on the properties of the molecules involved, rafts of different compositions are formed, implying structural and functional heterogeneity of distinct membrane domains. Thus, translocation of certain receptor proteins into lipid rafts has been reported to occur upon their oligomerization, their partitioning being favored by the decreased lateral diffusion upon cross-linking. 41,42 Alternatively, it is possible that glycosphingolipids may display a structure-dependent

accumulation of cholesterol is seen, which causes lysosomal storage of cholesterol and sphingolipids. Specifically, in fibroblasts obtained from patients suffering from a variety of sphingolipid storage diseases, massive sphingolipid trafficking into the late endosomal/lysosomal pathway could be precluded, apparently independent of the chemical nature of the sphingolipid, when cholesterol was depleted from these cells. Instead, in the cholesterol-depleted cells, the sphingolipids recycled via recycling endosomes, as observed in control fibroblasts. In the latter, aberrant trafficking into the lysosomal pathway was accomplished by supplementing the cells with excess cholesterol. Taken together, these data strongly support the view that cholesterol, possibly because of altering membrane rigidity (fluidity) and the degree of motional freedom by causing clustering of molecules in raft-like microdomains, modulates molecular sorting. In the case of NPC disease, the molecular defects and consequences are more complex and beyond the scope of this work (see Reference 34 for details), but in case of the sphingolipid storage diseases, the defect appears to be functionally expressed at the level of the early sorting endosome, where trafficking diverts into either the degradation pathway or recycling via recycling endosomes (cf. Figure 1 and References 16 and 34). In this context it is finally of interest to note that in differentiated HT29 colon carcinoma cells, plasma membrane-derived GlcCer and SM largely flow along the late-endosomal/lysosomal pathway, whereas in the undifferentiated counterpart, the lipids rapidly recycle and return to the plasma membrane after internalization by endocytosis. 35 Whether a difference in cellular levels of cholesterol underlies these differentiation-dependent differences in sorting and trafficking of the sphingolipids remains to be determined. To appreciate the significance of sphingolipids in raft formation, the following observations are particularly instructive. In general, sphingolipids are characterized by the presence of long chain fatty acids (more than 18–20 carbons in length) and a long alkyl chain sphingoid base, which implies that their liquid–crystalline phase transition temperatures are usually located well above physiologically relevant temperatures. In contrast, phospholipids contain a considerable fraction of unsaturated fatty acids which imparts membrane fluidity. When mixed, the solid GSL display a tendency to phase separate (‘cluster’) from the fluid phospholipids. Cholesterol, as a fluidizer of solid membranes, prefers to partition 153

O. Maier et al.

Polarized trafficking: orchestration and mediators

preference for association with certain proteins (or vice-versa), as exemplified by the differential association of GM1 and GD1a with detergentinsoluble caveolae-like membrane domains. 43 In model systems it has been shown that sphingolipid segregation may increase the local concentration to a density where their clustering becomes stabilized, even in the absence of cholesterol. 44 Thus at high concentrations sphingolipids can maintain an ordered state in the absence of cholesterol, the latter largely facilitating raft formation at low sphingolipid concentrations. Accordingly, rafts may exist even in the absence of cholesterol 45 and sphingolipidrich/cholesterol-poor microdomains, enriched in SM and the ganglioside GM3, have been detected on the surface of melanoma B16 cells. Here they appear to function in GM3-dependent cell adhesion and signaling, which is not affected by cholesterol depletion 46 indicating that, depending on local sphingolipid concentrations, cholesterol may or may not be involved in sorting of sphingolipids. Current evidence largely supports a facilitating role of cholesterol (see Reference 34) since, like cholesterol, the ganglioside GM3 accumulates in endosomal compartments of NPC cells, which are defective in NPC1, the sterol-sensing protein that plays a role in the egress of cholesterol from late-endosomal and lysosomal compartments. 47 Nevertheless the possibility of separating GM3 and SM from a cholesterol-enriched domain, which contains GlcCer and caveolin, the major protein of caveolae, 46 indicates that the stability of microdomains could be governed by different factors, involving self-aggregation of sphingolipid, association with cholesterol and clustering with different sets of proteins. In this context it may be noted that oligomerization of caveolin-1 leads to detergent insolubility of the protein. 48 Microdomain composition seems, therefore, to be an important parameter in determining the assembly of a given sphingolipid in such a domain. For example, fluorescent C6-NBD-SM participates in raft formation in developing neurons, 38 segregates into distinct domains in SAC in HepG2 cells, 49 but acts like a bulk membrane marker in the endocytic recycling pathway of non-polarized CHO cells. 50 Sorting thus includes a spatial segregation of lipids into microdomains, with or without proteins, which are recognized by specific transport machinery.

The first evidence showing that epithelial cells are capable of polarized sorting of newly synthesized sphingolipids, came from monitoring the trafficking of fluorescently labeled (C6-NBD) sphingolipid analogues. After labeling with C6-NBD-ceramide, and allowing biosynthesis of fluorescently-labeled SM and GlcCer (which are the main products formed), MDCK and Caco-2 cells transport C6-NBD-SM predominantly to the basolateral and C6-NBD-GlcCer to the apical surface 51,52 (indicated in Figure 1 as steps 1A and B). Consistent with such a sorting, is the observation that distinct transport vesicle populations enriched in their lumenal leaflet in either newly synthesized SM or GlcCer could be isolated from HT29 cells, thus providing further evidence that cells are indeed able to laterally separate these sphingolipids from each other followed by selective packaging into distinct transport vesicles, 53 a prerequisite for their polarized transport. In addition, most polarized epithelial cells transport the majority or even all of their apical proteins directly to their final location and sphingolipid-enriched rafts constitute the platform for apical transport of GPI-linked proteins as well as some transmembrane proteins, like influenza virus hemagglutinin. The mechanism of how rafts are involved in apical targeting is as yet largely unknown, but it is likely that other signals like N -glycosylation of proteins are also important. 54 In addition, proteins have been described that may mediate the vesiculation and/or targeting of the rafts to the apical membrane. One such protein that might be involved in apical targeting is annexin XIIIb, since apically directed vesicle trafficking from the TGN can be inhibited by annexin XIIIb antibodies, and is stimulated by raft-associated myristoylated recombinant annexin XIIIb. 55 MAL/VIP17 has also been implicated in vesicle formation, since the protein was found to be associated with Golgi-derived transport vesicles and the apical membrane surface. Its overexpression has been reported to stimulate apical transport, whereas downregulation of MAL/VIP17 expression inhibited apical transport without affecting basolateral transport. 56,57 Although MAL/VIP17 can be found associated with rafts, it has been implicated in overall apical transport in MDCK and Fisher rat thyroid (FRT) cells, i.e., also in non-raft mediated apical trafficking. 58 FRT cells are intriguing in that most endogenous GPI-linked proteins are soluble 154

Polarized sphingolipid trafficking

express sphingolipid sorting capacity at the plasma membrane, in that lactosylceramide is selectively retrieved at the plasma membrane, while sulfatide enters the endocytic pathway. 64 It will be of interest to examine whether and how these events relate to myelin assembly and to determine the potential role of MAL/VIP17 in sphingolipid transport upon biogenesis of the myelin sheath. Nevertheless, the OLG system emphasizes the complexity of the orchestration underlying raft-mediated trafficking. Similar to OLG, neurons have no tight junctions, but a barrier for the movement of lipids between the axon and the cell body has been found in polarized hippocampal neurons. 65 Cognate polarized trafficking pathways have also been revealed in neuronal cells for viral and GPI-linked proteins, 66,67 the axonal membrane corresponding to the apical membrane of epithelial cells. Similar to FRT cells detergent-soluble GPI-linked proteins, such as the prion protein, are localized in the cell body whereas detergent-insoluble GPI-linked proteins like the Thy-1 protein are expressed predominantly in the axon, 68 suggesting that in neurons ‘apical’ transport is raft-mediated. This view is confirmed by the observation that the localization of Thy-1 changes from a random to a primarily axonal one during axon development, which is induced upon upregulation of SM biosynthesis (see above). The latter appears instrumental in assembly of detergent-insoluble microdomains, which next to Thy-1 includes the ganglioside GM1. Although polarized trafficking pathways thus appear to be operative in neuronal cells as well, their traffic itinerary in terms of passage through and sorting in distinct intracellular compartments, as currently unraveled in polarized epithelial cells, remains to be clarified.

in non-ionic detergents for an unknown reason and accordingly are transported to the basolateral surface. 59 Nevertheless FRT cells can form detergent insoluble domains and some GPI-linked, as well as transmembrane proteins, are directly transported to the apical membrane by a mechanism that does involve the sphingolipid-enriched rafts. 60 Although they might, therefore, not be sufficient per se to mediate apical transport, rafts may be important for the recruitment of components of the apical sorting machinery; for example MAL/VIP17. MAL/VIP17 is also found in oligodendrocytes (OLG), which are localized in the central nervous system, where they are responsible for myelinating axons of neurons, which is essential for nerve conduction. Myelin membranes are continuous with the plasma membrane of the OLG, but the composition of the two membranes differs dramatically, the former being strongly enriched in GalCer and its sulfated derivative sulfatide, representing some 50 % of the total lipid pool in the outer leaflet of the myelin membrane. Given these distinct domains, an analogy with the apical (myelin) and the basolateral (plasma membrane of cell body) membrane domains in polarized cells is tempting. Indeed, although tight junctions have not been detected in OLG, distinct polarity properties could be revealed and the action of a raft-mediated mechanism in delivering (some) GPI-linked proteins to the myelin membrane was proposed, based upon detergent-insolubility. 61 However, in OLG another ‘typical’ apical membrane marker and raft-associated protein, influenza hemagglutinin, is delivered to the plasma membrane of the cell body in a raft-mediated mechanism, while a basolateral marker, VSV glycoprotein is transported to the sphingolipid enriched myelin domain. 62 Moreover, none of the major myelin-specific proteins (PLP, MAG, CNPase, MBP) are delivered to myelin according to a raft-mediated mechanism. 63 Hence cognate polarized pathways are operating in these cells, but the target membranes are partly reversed compared to typical polarized trafficking. A role of MAL/VIP17 in these cells remains to be defined, although a similarity in action to that seen in FRT cells is tempting to suggest. However, the protein’s expression is particularly apparent during OLG differentiation, concomitant with an increase in sphingolipid transport and myelin assembly. It is possible, however, that MAL/VIP17 is functioning in plasma membrane-directed transport, but indirectly promotes myelin assembly in OLG. This speculation is derived from the interesting observation that OLG

The subapical compartment and polarized sphingolipid sorting Although apical proteins are generally considered as the ones to be exclusively located and transported in rafts, raft directed trafficking to basolateral membranes cannot be excluded, 69 although existing evidence is scanty. 70 Rather, as described and discussed elsewhere, basolateral trafficking prominently relies on the involvement of heterotetrameric adaptor proteins/clathrin interactions. 9,16 It is interesting, however, that in liver cells, most apical proteins, identified thus far, have been shown to be 155

O. Maier et al.

transported to their resident domain in an indirect manner. Thus GPI-linked and transmembrane apical proteins are first delivered from the TGN to the basolateral membrane, followed by transcytosis to the apical membrane. 71 In contrast, newly synthesized sphingolipids can be transported directly from the TGN to the canalicular membrane of HepG2 cells, 5,72 although the sorting of SM and GlcCer on the level of the TGN has not yet been proven in these cells. Sorting of sphingolipids could, however, be demonstrated in the reverse transcytotic pathway, i.e., from the apical to the basolateral membrane, which occurs in a subapical compartment, SAC, of fully polarized HepG2 cells (Figure 2). Within the luminal leaflet of the SAC compartment, sorting devices operate that cause GlcCer to recycle back to the apical membrane, while SM and GalCer are transported predominantly to the basolateral membrane 18,73 (Figure 1, steps 6A and B). Importantly, it has been shown that these SAC-derived transport pathways for these various sphingolipids are vesicle-mediated. 18 The trafficking pathways of sphingolipids in the endosomal pathways of MDCK cells have been less well-characterized. But the recent observation that SM and cholesterol are highly enriched in the SAC/CE of these cells 11 would be consistent with the possibility that membrane domains can be generated in this compartment, in line with analogous observations in enterocytes (see below), 74 emphasizing the important role of rafts for lipid and protein sorting in polarized epithelial cells. In spite of the long lasting dogma that apical proteins in liver cells reach the apical membrane exclusively by an indirect transport pathway, evidence has recently been obtained that direct protein transport between TGN and apical (canalicular) membrane does occur, as demonstrated for the transfer of canalicular ABC transporters 75 and ATP7B, a protein that mediates copper excretion at the canalicular membrane. 76 Interestingly, apical transport of both these proteins may be regulated by physiological demand, triggered by enhanced levels of bile acids and copper, respectively. Furthermore, canalicular transport of ABC transporters is stimulated by dibutyryl cAMP (dbcAMP) and depends on intact microtubules. 77,78 Similarly, apically directed transport of sphingolipids is strongly promoted by dbcAMP in the biosynthetic and transcytotic pathway, promoting polarity development and leading to a fast expansion of the apical membrane. 5 Moreover, the direct Golgi-to-apical route of sphingolipid transport is, like direct protein transport, strongly

Figure 2. Sphingolipid sorting in the SAC of HepG2 cells during apical to basolateral transport. The bile canalicular (‘apical’) membrane of HepG2 cells was labeled (for experimental procedure see References 72 and 18) with either C6-NBD-SM (A, C) or C6-NBD-GlcCer (B, D) and the fate of both lipids is monitored by raising the temperature to traffic-permissive conditions (37 ◦ C). SM and GlcCer are sorted during transport from the bile canalicular (arrows) to the basolateral membrane. SM is transported to the basolateral membrane (A) while GlcCer is recycled back to the apical membrane (B). During this transport both sphingolipids can be trapped at 18 ◦ C in the SAC (C, D), as revealed by quenching the fluorescent label, present in the canalicular membrane, with sodium dithionite. Sorting devices that direct vesicle-mediated recycling (GlcCer) and vesicle-mediated basolateral-directed transport (SM) reside at the luminal site of the SAC compartment.

inhibited when the microtubules are disrupted, following nocodazole treatment. 79 At the level of SAC, activation of protein kinase A by dbcAMP causes the redirection of SM transport to the apical membrane in a microtubule-dependent pathway that is clearly distinct from the apical recycling pathway of GlcCer 80 (Figure 1, steps 6C and C0 ), the latter being essentially unaffected by PKA activation. PKA-activated SM-transport, thus reflecting ‘the apical domain assembly pathway’, includes passage 156

Polarized sphingolipid trafficking

SPGP with pIgR in distinct transport vesicles, 81 pIgR also being processed through SAC during basolateral to apical transcytosis and co-localizing with the SM, but not GlcCer-marked transport pathway between SAC and apical membrane, 16 is a further indication that a direct pathway exists between TGN and SAC. In support of this notion, in small intestinal enterocytes aminopeptidase N colocalizes with IgA, a ligand of the pIgR, in compartments near the apical, but not basolateral surface. 74 Intriguingly, the protein is present in rafts only after a 20–40 min chase and reaches the apical surface after about 60 min, in contrast to a 30 min time interval previously noted to be required for direct TGN-apical membrane transport. 75 Finally, both newly synthesized transferrin receptor 82 and asialoglycoprotein receptor 83 have been detected in endosomal compartments that may represent CE or SAC, prior to their delivery to the basolateral surface. Taken together, these data further support: (i) the view of a direct pathway between TGN and SAC/CE; (ii) that proteins can be transported to SAC/CE via a raft-independent mechanism; and (iii) that for apical targeting (SPGP; aminopeptidase N) they may become recruited into rafts that are assembled in the SAC rather than the TGN. With regard to the latter, cholesterol-depletion caused a missorting of apical proteins to the basolateral membrane in enterocytes, 84 a phenomenon noted before in other cell types, 28,85 but this study does not exclude that the missorting step may have originated from missorting at the level of SAC. In terms of sphingolipid sorting this would imply that raft assembly is the driving force for carrying cargo to the apical membrane in a direct pathway, either originating from TGN or SAC. Issues that remain to be resolved are the identity of factors that regulate which proteins are embedded in rafts and where (TGN versus SAC), and which are excluded, since apical proteins such as the GPI-anchored 50 nucleotidase, and transmembrane proteins such as dipeptidylpeptidase IV and the canalicular cell adhesion molecule cCAM 105, 71,86 reach the bile canaliculus only by transcytosis, i.e., delivery to the basolateral membrane prior to transport to the apical membrane. In this context it should be mentioned that the MAL/VIP17 protein that may be a key molecule for direct apical targeting in polarized epithelial cells (see above), is not expressed in liver cells. 87 In WIF-B hepatocytes, some of these apical membrane proteins have been shown to reach the apical membrane via a subapical compartment, 88 which is very likely reminiscent of the SAC, as defined

through the ARE, also called subapical intermediate compartment or SIC, and transport from the SIC to the canalicular membrane (Figure 1, step 6C0 ), which depends on intact microtubules (van IJzendoorn SCD, van der Wouden JM, Jonker M, Hoekstra D, unpublished observations). Interestingly, the evidence revealed that this PKA-stimulated apical trafficking of SM occurs under natural conditions in early polarity development of HepG2 cells, 80 during which an enhanced, but transient expression of intrinsic PKA, is triggered. Once polarized, the trafficking of SM largely proceeds again from SAC to the basolateral membrane, as observed in fully polarized cell cultures. These remarkable differences in PKA activation-dependent trafficking of SM on the one hand and GlcCer on the other clearly demonstrates the sorting capacity of the SAC compartment and emphasizes that, given that the ceramide backbone structure is identical for both lipids, head group differences are likely to be relevant to their sorting. It is also of particular interest to mention the remarkable similarity and potential correlation that seems to exist (and which merits further study) between a stimulation and/or apical transport of SM per se and polarity development, as seen in such diverse cell types as neurons (see above) and liver cells. As noted above, a direct sphingolipid transport pathway exists in liver cells that carries de novo synthesized SM and GlcCer to the apical membrane, without prior transport to the basolateral membrane followed by transcytosis. An intriguing question is to what extent this direct transport pathway involves trafficking via SAC. Put differently, is there a direct connection between TGN and SAC in polarized trafficking? In this context, the observations that the direct apical protein transport pathways in hepatocytes departing from the Golgi may differ, as revealed by differences in the kinetics by which the ABC transporters MDR1 and MDR 2 (30 min), and SPGP (2 h) reached the canalicular membrane, 75 are most intriguing. A transient sequestering of SPGP in subapically located compartments was suggested and it is tempting to suggest that these compartments represent SAC (indicated in Figure 1, ‘?’). Accordingly, the intriguing possibility may exist that the differences in kinetics of protein arrival at the apical membrane may be due to the assembly of distinct sphingolipidcontaining domains in the lateral plane of the Golgi membrane, which recruit different proteins, giving rise to different transport vesicles that transfer either directly to the apical membrane or via SAC. In this respect, the previously observed colocalization of 157

O. Maier et al.

in HepG2 cells. It will be of interest, therefore, to determine whether these proteins may become integrated in this compartment in sphingolipid domains for apical delivery.

of sphingolipids with lipids like cholesterol and their potential affinity for (distinct?) proteins. However, the notion that the SAC plays an important role in sphingolipid trafficking related to polarity development, may provide an opportunity to isolate such organelles for biochemical characterization. By employing appropriate techniques, including photoaffinity labeled sphingolipids, and by exploiting experimental systems that show an impairment in lipid flow, as induced under several pathological conditions connected with mutations in distinct proteins, it will become possible to obtain insight into molecular devices that regulate the trafficking of sphingolipids.

Concluding remarks Recent progress in the field of polarized sphingolipid trafficking reveals that, for a detailed understanding of its mechanism and regulation, an integrated approach that includes the flow of membrane proteins, is imperative. Particularly in polarized cells, but no less so in non-polarized cells where often cognate polarized trafficking pathways can be recognized, sphingolipid and protein trafficking cannot be considered as separate events. Sphingolipids, in conjunction with cholesterol, clearly can provide a sorting matrix for proteins, and their distinct structural and functional identity is gradually emerging, as for example in the case of a switch in sphingomyelin-specific basolateral versus apical trafficking, triggered by a transient activation of protein kinase A during polarity development, and a sphingomyelin-synthesis driven apical sorting process, as observed in developing neurons. Moreover, evidence is also accumulating that domains of different sphingolipid composition may arise, within the same lateral plane of the bilayer. Such domains may express different biological functions, which include processes like signaling and cell–cell contact, and a major challenge will be to reveal the molecular mechanism that drives the assembly, dynamics and likely transient life time of these different domains, including the selective recruitment of the distinct sphingolipids, and the potential cross-talk between such domains. In this regard, it is also relevant to consider the possibility that, by means of interdigitation, arising from their long alkyl and acyl chains, sphingolipids can mobilize specific phospholipids in the opposed leaflet, potentially giving rise to equally important cell biological consequences that rely on lipid-mediated activities. Unfortunately, our understanding of the molecular sorting mechanism of sphingolipids is still very scanty, although cholesterol appears to be an important co-factor in governing their flow (and fate), which presumably relates to primary effects on biophysical properties like fluidity, which are subsequently recognized by some cellular machinery. Further insight into these biophysical properties will be required, including the interaction

Acknowledgements O. M. was supported by grant BMBF-LPD 980121 from the Deutsche Akademie der Naturforscher Leopoldina.

References 1. Hirschberg K, Rodger J, Futerman AH (1993) The long-chain sphingoid base of sphingolipids is acylated at the cytosolic surface of the endoplasmic reticulum in rat liver. Biochem J 290:751–757 2. Bose R, Verheij M, Haimovitz-Friedman A, Scotto K, Fuks Z, Kolesnick R (1995) Ceramide synthase mediates daunorubicin-induced apoptosis: an alternative mechanism for generating death signals. Cell 82:405–414 3. Ichikawa S, Hirabayashi Y (1998) Glucosylceramide synthase and glycosphingolipid synthesis. Trends Cell Biol 8:198–202 4. Simon CG, Jr, Holloway PW, Gear AR (1999) Exchange of C(16)-ceramide between phospholipid vesicles. Biochemistry 38:14676–14682 5. Zegers MMP, Hoekstra D (1997) Sphingolipid transport to the apical plasma membrane domain in human hepatoma cells is controlled by PKC and PKA activity: a correlation with cell polarity in HepG2 cells. J Cell Biol 138:307–321 6. Fukasawa M, Nishijima M, Hanada K (1999) Genetic evidence for ATP-dependent endoplasmic reticulum-to-Golgi apparatus trafficking of ceramide for sphingomyelin synthesis in Chinese hamster ovary cells. J Cell Biol 144:673–685 7. Lannert H, Bunning C, Jeckel D, Wieland FT (1994) Lactosylceramide is synthesized in the lumen of the Golgi apparatus. FEBS Lett 342:91–96 8. Van IJzendoorn SCD, Hoekstra D (1999) The subapical compartment: a novel sorting centre? Trends Cell Biol 9:144– 149 9. Futter CE, Gibson A, Allchin EH, Maxwell S, Ruddock LJ, Odorizzi G, Domingo D, Trowbridge IS, Hopkins CR (1998) In polarized MDCK cells basolateral vesicles arise from clathrin-γ -adaptin-coated domains on endosomal tubules. J Cell Biol 141:611–623 10. Mostov KE, Verges M, Altschuler Y (2000) Membrane traffic


Polarized sphingolipid trafficking

29. Hooper NM (1998) Membrane biology: do glycolipid microdomains really exist? Curr Biol 8:R114–R116 30. Mukherjee S, Maxfield FR (2000) Role of membrane organization and membrane domains in endocytic lipid trafficking. Traffic 1:203–211 31. Vaz WL, Almeida PF (1993) Phase topology and percolation in multiphase bilayers: is the biological membrane a domain mosaic? Curr Opin Struct Biol 3:482–488 32. Neufeld EB et al. (1999) The Niemann–Pick C1 protein resides in a vesicular compartment linked to retrograde transport of multiple lysosomal cargo. J Biol Chem 274:9627– 9635 33. Puri V, Watanabe R, Dominguez M, Sun X, Wheatley CL, Marks DL, Pagano RE (1999) Cholesterol modulates membrane traffic along the endocytic pathway in sphingolipidstorage diseases. Nat Cell Biol 1:386–388 34. Hoekstra D, van IJzendoorn SCD (2000) Lipid trafficking and sorting: how cholesterol is filling gaps. Curr Opin Cell Biol 12:496–502 35. Kok JW, Babia T, Hoekstra D (1991) Sorting of sphingolipids in the endocytic pathway of HT29 cells. J Cell Biol 114:231– 239 36. Brown DA, London E (2000) Structure and function of sphingolipid- and cholesterol-rich membrane rafts. J Biol Chem 275:17221–17224 37. Ostermeyer AG, Beckrich BT, Ivarson KA, Grove KE, Brown DA (1999) Glycosphingolipids are not essential for formation of detergent-resistant membrane rafts in melanoma cells. Methyl-β-cyclodextrin does not affect cell surface transport of a GPI-anchored protein. J Biol Chem 274:34459– 34466 38. Ledesma MD, Brugger B, Bunning C, Wieland FT, Dotti CG (1999) Maturation of the axonal plasma membrane requires upregulation of sphingomyelin synthesis and formation of protein-lipid complexes. EMBO J 18:1761–1771 39. Boldin SA, Futerman AH (2000) Up-regulation of glucosylceramide synthesis upon stimulation of axonal growth by basic fibroblast growth factor. Evidence for post-translational modification of glucosylceramide synthase. J Biol Chem 275:9905– 9909 40. Melkonian KA, Ostermeyer AG, Chen JZ, Roth MG, Brown DA (1999) Role of lipid modifications in targeting proteins to detergent-resistant membrane rafts. Many raft proteins are acylated, while few are prenylated. J Biol Chem 274:3910– 3917 41. Cheng PC, Dykstra ML, Mitchell RN, Pierce SK (1999) A role for lipid rafts in B cell antigen receptor signaling and antigen targeting. J Exp Med 190:1549–1560 42. Montixi C, Langlet C, Bernard AM, Thimonier J, Dubois C, Wurbel MA, Chauvin JP, Pierres M, He HT (1998) Engagement of T cell receptor triggers its recruitment to low-density detergent-insoluble membrane domains. EMBO J 17:5334–5348 43. Wolf AA, Jobling MG, Wimer-Mackin S, Ferguson-Maltzman M, Madara JL, Holmes RK, Lencer WI (1998) Ganglioside structure dictates signal transduction by cholera toxin and association with caveolae-like membrane domains in polarized epithelia. J Cell Biol 141:917–927 44. Rock P, Allietta M, Young WW, Jr, Thompson TE, Tillack TW (1991) Ganglioside GM1 and asialo-GM1 at low concentration are preferentially incorporated into the gel phase in two-component, two-phase phosphatidylcholine bilayers. Biochemistry 30:19–25 45. Brown RE (1998) Sphingolipid organization in biomembranes: what physical studies of model membranes reveal. J

in polarized epithelial cells. Curr Opin Cell Biol 12:483–490 11. Gagescu R, Demaurex N, Parton RG, Hunziker W, Huber LA, Gruenberg J (2000) The recycling endosome of Madin–Darby canine kidney cells is a mildly acidic compartment rich in raft components. Mol Biol Cell 11:2775–2791 12. Hunziker W, Peters PJ (1998) Rab17 localizes to recycling endosomes and regulates receptor-mediated transcytosis in epithelial cells. J Biol Chem 273:15734–15741 13. Zacchi P, Stenmark H, Parton RG, Orioli D, Lim F, Giner A, Mellman I, Zerial M, Murphy C (1998) Rab17 regulates membrane trafficking through apical recycling endosomes in polarized epithelial cells. J Cell Biol 140:1039–1053 14. Casanova JE, Wang X, Kumar R, Bhartur SG, Navarre J, Woodrum JE, Altschuler Y, Ray GS, Goldenring JR (1999) Association of Rab25 and Rab11a with the apical recycling system of polarized Madin–Darby canine kidney cells. Mol Biol Cell 10:47–61 15. Wang E, Brown PS, Aroeti B, Chapin SJ, Mostov KE, Dunn KW (2000) Apical and basolateral endocytic pathways of MDCK cells meet in acidic common endosomes distinct from a nearly-neutral apical recycling endosome. Traffic 1:480–493 16. Van IJzendoorn SCD, Maier O, van der Wouden JM, Hoekstra D (2000) The subapical compartment and its role in intracellular trafficking and cell polarity. J Cell Physiol 184:151–160 17. Orzech E, Cohen S, Weiss A, Aroeti B (2000) Interactions between the exocytic and endocytic pathways in polarized Madin–Darby canine kidney cells. J Biol Chem 275:15207– 15219 18. Van IJzendoorn SCD, Hoekstra D (1998) (Glyco)sphingolipids are sorted in sub-apical compartments in HepG2 cells: a role for non-Golgi-related intracellular sites in the polarized distribution of (glyco)sphingolipids. J Cell Biol 142:683–696 19. Warnock DE, Lutz MS, Blackburn WA, Young WW, Jr, Baenziger JU (1994) Transport of newly synthesized glucosylceramide to the plasma membrane by a non-Golgi pathway. Proc Natl Acad Sci USA 91:2708–2712 20. Zegers MM, Kok JW, Hoekstra D (1997) Use of photoactivatable sphingolipid analogues to monitor lipid transport in mammalian cells. Biochem J 328:489–498 21. Raggers RJ, van Helvoort A, Evers R, van Meer G (1999) The human multidrug resistance protein MRP1 translocates sphingolipid analogs across the plasma membrane. J Cell Sci 112:415–422 22. van’t Hof W, Malik A, Vijayakumar S, Qiao J, van Adelsberg J, Al-Awqati Q (1997) The effect of apical and basolateral lipids on the function of the band 3 anion exchange protein. J Cell Biol 139:941–949 23. Varma R, Mayor S (1998) GPI-anchored proteins are organized in submicron domains at the cell surface. Nature 394:798–801 24. Ikonen E, Simons K (1998) Protein and lipid sorting from the trans-Golgi network to the plasma membrane in polarized cells. Semin Cell Dev Biol 9:503–509 25. Jacobson K, Dietrich C (1999) Looking at lipid rafts? Trends Cell Biol 9:87–91 26. Muniz M, Riezman H (2000) Intracellular transport of GPIanchored proteins. EMBO J 19:10–15 27. Orci L, Montesano R, Meda P, Malaisse-Lagae F, Brown D, Perrelet A, Vassalli P (1981) Heterogeneous distribution of filipin–cholesterol complexes across the cisternae of the Golgi apparatus. Proc Natl Acad Sci USA 78:293–297 28. Keller P, Simons K (1998) Cholesterol is required for surface transport of influenza virus hemagglutinin. J Cell Biol 140:1357–1367


O. Maier et al.

Cell Sci 111:1–9 46. Iwabuchi K, Handa K, Hakomori S (1998) Separation of “glycosphingolipid signaling domain” from caveolin-containing membrane fraction in mouse melanoma B16 cells and its role in cell adhesion coupled with signaling. J Biol Chem 273:33766–33773 47. Sato M, Akaboshi S, Katsumoto T, Taniguchi M, Higaki K, Tai T, Sakuraba H, Ohno K (1998) Accumulation of cholesterol and GM2 ganglioside in cells cultured in the presence of progesterone: an implication for the basic defect in Niemann–Pick disease type C. Brain Dev 20:50–52 48. Scheiffele P, Verkade P, Fra AM, Virta H, Simons K, Ikonen E (1998) Caveolin-1 and -2 in the exocytic pathway of MDCK cells. J Cell Biol 140:795–806 49. Van Ijzendoorn SCD, Hoekstra D (1999) Polarized sphingolipid transport from the subapical compartment: evidence for distinct sphingolipid domains. Mol Biol Cell 10:3449–3461 50. Mayor S, Presley JF, Maxfield FR (1993) Sorting of membrane components from endosomes and subsequent recycling to the cell surface occurs by a bulk flow process. J Cell Biol 121:1257– 1269 51. Van Meer G, Stelzer EH, Wijnaendts van Resandt RW, Simons K (1987) Sorting of sphingolipids in epithelial (MadinDarby canine kidney) cells. J Cell Biol 105:1623–1635 52. Van’t Hof W, van Meer G (1990) Generation of lipid polarity in intestinal epithelial (Caco-2) cells: sphingolipid synthesis in the Golgi complex and sorting before vesicular traffic to the plasma membrane. J Cell Biol 111:977–986 53. Babia T, Kok JW, van der Haar M, Kalicharan R, Hoekstra D (1994) Transport of biosynthetic sphingolipids from Golgi to plasma membrane in HT29 cells: involvement of different carrier vesicle populations. Eur J Cell Biol 63:172–181 54. Benting JH, Rietveld AG, Simons K (1999) N -Glycans mediate the apical sorting of a GPI-anchored, raft-associated protein in Madin–Darby canine kidney cells. J Cell Biol 146:313–320 55. Lafont F, Lecat S, Verkade P, Simons K (1998) Annexin XIIIb associates with lipid microdomains to function in apical delivery. J Cell Biol 142:1413–1427 56. Cheong KH, Zacchetti D, Schneeberger EE, Simons K (1999) VIP17/MAL, a lipid raft-associated protein, is involved in apical transport in MDCK cells. Proc Natl Acad Sci USA 96:6241–6248 57. Puertollano R, Alonso MA (1999) MAL, an integral element of the apical sorting machinery, is an itinerant protein that cycles between the trans-Golgi network and the plasma membrane. Mol Biol Cell 10:3435–3447 58. Martin-Belmonte F, Puertollano R, Millan J, Alonso MA (2000) The MAL proteolipid is necessary for the overall apical delivery of membrane proteins in the polarized epithelial Madin–Darby canine kidney and Fischer rat thyroid cell lines. Mol Biol Cell 11:2033–2045 59. Zurzolo C, Lisanti MP, Caras IW, Nitsch L, RodriguezBoulan E (1993) Glycosylphosphatidylinositol-anchored proteins are preferentially targeted to the basolateral surface in Fischer rat thyroid epithelial cells. J Cell Biol 121:1031–1039 60. Lipardi C, Nitsch L, Zurzolo C (2000) 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. Mol Biol Cell 11:531–542 61. Kramer EM, Koch T, Niehaus A, Trotter J (1997) Oligodendrocytes direct glycosyl phosphatidylinositol-anchored proteins to the myelin sheath in glycosphingolipid-rich complexes. J Biol Chem 272:8937–8945 62. De Vries H, Schrage C, Hoekstra D (1998) An apical-type



















trafficking pathway is present in cultured oligodendrocytes but the sphingolipid-enriched myelin membrane is the target of a basolateral-type pathway. Mol Biol Cell 9:599–609 De Vries H, Hoekstra D (2000) On the biogenesis of the myelin sheath: cognate polarized trafficking pathways in oligodendrocytes. Glycoconjugate J 17:1–9 Watanabe R, Asakura K, Rodriguez M, Pagano RE (1999) Internalization and sorting of plasma membrane sphingolipid analogues in differentiating oligodendrocytes. J Neurochem 73:1375–1383 Kobayashi T, Storrie B, Simons K, Dotti CG (1992) A functional barrier to movement of lipids in polarized neurons. Nature 359:647–650 Dotti CG, Simons K (1990) Polarized sorting of viral glycoproteins to the axon and dendrites of hippocampal neurons in culture. Cell 62:63–72 Dotti CG, Parton RG, Simons K (1991) Polarized sorting of glypiated proteins in hippocampal neurons. Nature 349:158– 161 Madore N, Smith KL, Graham CH, Jen A, Brady K, Hall S, Morris R (1999) Functionally different GPI proteins are organized in different domains on the neuronal surface. EMBO J 18:6917–6926 Zegers MMP, Hoekstra D (1998) Mechanisms and functional features of polarized membrane traffic in epithelial and hepatic cells. Biochem J 336:257–269 Oliferenko S, Paiha K, Harder T, Gerke V, Schwarzler C, Schwarz H, Beug H, Gunthert U, Huber LA (1999) Analysis of CD44-containing lipid rafts: Recruitment of annexin II and stabilization by the actin cytoskeleton. J Cell Biol 146:843–854 Schell MJ, Maurice M, Stieger B, Hubbard AL (1992) 50 nucleotidase is sorted to the apical domain of hepatocytes via an indirect route. J Cell Biol 119:1173–1182 Zaal KJ, Kok JW, Sormunen R, Eskelinen S, Hoekstra D (1994) Intracellular sites involved in the biogenesis of bile canaliculi in hepatic cells. Eur J Cell Biol 63:10–19 Van IJzendoorn SCD, Zegers MM, Kok JW, Hoekstra D (1997) Segregation of glucosylceramide and sphingomyelin occurs in the apical to basolateral transcytotic route in HepG2 cells. J Cell Biol 137:347–357 Hansen GH, Niels-Christiansen LL, Immerdal L, Hunziker W, Kenny AJ, Danielsen EM (1999) Transcytosis of immunoglobulin A in the mouse enterocyte occurs through glycolipid raft- and rab17-containing compartments. Gastroenterology 116:610–622 Kipp H, Arias IM (2000) Newly synthesized canalicular ABC transporters are directly targeted from the Golgi to the hepatocyte apical domain in rat liver. J Biol Chem 275:15917– 15925 Roelofsen H, Wolters H, van Luyn MJA, Miura N, Kuipers F, Vonk RJ (2000) Copper-induced apical trafficking of ATP7B in polarized hepatoma cells provides a mechanism for biliary copper excretion. Gastroenterology 119:782–793 Roelofsen H, Soroka CJ, Keppler D, Boyer JL (1998) Cyclic AMP stimulates sorting of the canalicular organic anion transporter (Mrp2/cMoat) to the apical domain in hepatocyte couplets. J Cell Sci 111:1137–1145 Gatmaitan ZC, Nies AT, Arias IM (1997) Regulation and translocation of ATP-dependent apical membrane proteins in rat liver. Am J Physiol 272:G1041–G1049 Zegers MM, Zaal KJ, van IJzendoorn SCD, Klappe K, Hoekstra D (1998) Actin filaments and microtubules are involved in different membrane traffic pathways that transport sphingolipids to the apical surface of polarized HepG2 cells. Mol Biol Cell 9:1939–1949

Polarized sphingolipid trafficking

85. Ledesma MD, Simons K, Dotti CG (1998) Neuronal polarity: essential role of protein–lipid complexes in axonal sorting. Proc Natl Acad Sci USA 95:3966–3971 86. Bartles JR, Feracci HM, Stieger B, Hubbard AL (1987) Biogenesis of the rat hepatocyte plasma membrane in vivo: comparison of the pathways taken by apical and basolateral proteins using subcellular fractionation. J Cell Biol 105:1241– 1251 87. Millan J, Alonso MA (1998) MAL, a novel integral membrane protein of human T lymphocytes, associates with glycosylphosphatidylinositol-anchored proteins and Srclike tyrosine kinases. Eur J Immunol 28:3675–3684 88. Ihrke G, Martin GV, Shanks MR, Schrader M, Schroer TA, Hubbard AL (1998) Apical plasma membrane proteins and endolyn-78 travel through a subapical compartment in polarized WIF-B hepatocytes. J Cell Biol 141:115–133

80. Van IJzendoorn SCD, Hoekstra D (2000) Polarized sphingolipid transport from the subapical compartment changes during cell polarity development. Mol Biol Cell 11:1093–1101 81. Soroka CJ, Pate MK, Boyer JL (1999) Canalicular export pumps traffic with polymeric immunoglobulin A receptor on the same microtubule-associated vesicle in rat liver. J Biol Chem 274:26416–26424 82. Futter CE, Connolly CN, Cutler DF, Hopkins CR (1995) Newly synthesized transferrin receptors can be detected in the endosome before they appear on the cell surface. J Biol Chem 270:10999–11003 83. Leitinger B, Hille-Rehfeld A, Spiess M (1995) Biosynthetic transport of the asialoglycoprotein receptor H1 to the cell surface occurs via endosomes. Proc Natl Acad Sci USA 92:10109–10113 84. Hansen GH, Niels-Christiansen LL, Thorsen E, Immerdal L, Danielsen EM (2000) Cholesterol depletion of enterocytes. Effect on the Golgi complex and apical membrane trafficking. J Biol Chem 275:5136–5142