N-glycans as apical targeting signals in polarized epithelial cells

MDCK (Madin-Darby canine kidney) cells represent a good model of polarized epithelium to investigate the signals involved in the apical targeting of proteins. As reported previously, GPI (glycosylphosphatidylinositol) anchors mediate the apical sorting of proteins in polarized epithelial cells through their interaction with lipid rafts. However, using a naturally N-glycosylated and GPI-anchored protein, we found that the GPI anchor does not influence the targeting of the protein. It is, in fact, the N-glycans that signal the protein to the apical surface. In the present review, the role of N-glycans and GPI anchors as apical signals is discussed along with the putative mechanisms involved.     

Rab14 regulates apical targeting in polarized epithelial cells

Epithelial cells display distinct apical and basolateral membrane domains, and maintenance of this asymmetry is essential to the function of epithelial tissues. Polarized delivery of apical and basolateral membrane proteins from the trans Golgi network (TGN) and/or endosomes to the correct domain requires specific cytoplasmic machinery to control the sorting, budding and fission of vesicles. However, the molecular machinery that regulates polarized delivery of apical proteins remains poorly understood. In this study, we show that the small guanosine triphosphatase Rab14 is involved in the apical targeting pathway. Using yeast two-hybrid analysis and glutathione S-transferase pull down, we show that Rab14 interacts with apical membrane proteins and localizes to the TGN and apical endosomes. Overexpression of the GDP mutant form of Rab14 (S25N) induces an enlargement of the TGN and vesicle accumulation around Golgi membranes. Moreover, expression of Rab14-S25N results in mislocalization of the apical raft-associated protein vasoactive intestinal peptide/MAL to the basolateral domain but does not disrupt basolateral targeting or recycling. These data suggest that Rab14 specifically regulates delivery of cargo from the TGN to the apical domain.     

Analysis of the signal for attachment of a glycophospholipid membrane anchor

The COOH terminus of decay accelerating factor (DAF) contains a signal that directs attachment of a glycophospholipid (GPI) membrane anchor. To define this signal we deleted portions of the DAF COOH terminus and expressed the mutant cDNAs it CV1 origin-deficient SV-40 cells. Our results show that the COOH-terminal hydrophobic domain (17 residues) is absolutely required for GPI anchor attachment. However, when fused to the COOH terminus of a secreted protein this hydrophobic domain is insufficient to direct attachment of a GPI anchor. Additional specific information located within the adjacent 20 residues appears to be necessary. We speculate that by analogy with signal sequences for membrane translocation, GPI anchor attachment requires both a COOH-terminal hydrophobic domain (the GPI signal) as well as a suitable cleavage/attachment site located NH2 terminal to the signal.     

Differential targeting of an epithelial plasma membrane glycoprotein in polarized Madin-Darby canine kidney cells

Using monoclonal antibodies directed against the plasma membrane of Madin-Darby canine kidney (MDCK) cells, we demonstrated previously that a glycoprotein with an Mr = 23,000 (gp23) had a non-polarized cell surface distribution and was observed on both the apical and basolateral membranes (Ojakian, G. K., Romain, R. E., and Herz, R. E. (1987) Am. J. Physiol. 253, C433-C443). However, in parallel studies on MDCK clonal lines (D11, D18) with high transepithelial electrical resistances and in kidney cells in vivo it was determined that gp23 had a polarized cell surface distribution, being localized only to the basolateral membrane. The cell surface distribution of other glycoproteins was identical in both MDCK and MDCK clonal lines, indicating that MDCK cells were not deficient in the ability to properly sort membrane glycoproteins. Metabolic labeling with radioactive substrates followed by immunopurification and gel electrophoresis demonstrated that gp23 from both MDCK and MDCK clone D11 had many biochemical similarities including electrophoretic mobility, glycosylation, and palmitate incorporation. However, proteolytic digestion of gp23 from MDCK and clone D11 cells produced unique peptide maps suggesting that these closely related glycoproteins may have different primary sequences. In this report, we present evidence that the differential targeting of gp23 may be due to differences between the primary sequences of the basolateral and non-targeted proteins. The possibility that the observed differences in gp23 targeting are due to the presence of a basolateral recognition signal in gp23 from clone D11 cells is discussed.     

An internally positioned signal can direct attachment of a glycophospholipid membrane anchor

All known glycophosphatidylinositol (GPI)-anchored membrane proteins contain a COOH-terminal hydrophobic domain necessary for signalling anchor attachment. To examine the requirement that this signal be at the COOH terminus of the protein, we constructed a chimeric protein, DAFhGH, in which human growth hormone (hGH) was fused to the COOH terminus of decay accelerating factor (DAF) (a GPI-anchored protein), thereby placing the GPI signal in the middle of the chimeric protein. We show that the fusion protein appears to be processed at the normal DAF processing site in COS cells, producing GPI-anchored DAF on the cell surface. This result indicates that the GPI signal does not have to be at the COOH terminus to direct anchor addition, suggesting that the absence of a hydrophilic COOH-terminal extension (beyond the hydrophobic domain) is not a necessary requirement for GPI anchoring. A similar DAFhGH fusion, containing an internal GPI signal in which the DAF hydrophobic domain was replaced with the signal peptide of hGH, also produced GPI-anchored cell surface DAF. The signal for GPI attachment thus exhibits neither position specificity nor sequence specificity. In addition, mutant DAF or DAFhGH constructs lacking an NH2-terminal signal peptide failed to produce GPI-anchored protein, suggesting that membrane translocation is necessary for anchor addition.     

Human respiratory syncytial virus glycoproteins are not required for apical targeting and release from polarized epithelial cells

Human respiratory syncytial virus (HRSV) is released from the apical membrane of polarized epithelial cells. However, little is known about the processes of assembly and release of HRSV and which viral gene products are involved in the directional maturation of the virus. Based on previous studies showing that the fusion (F) glycoprotein contained an intrinsic apical sorting signal and that N- and O-linked glycans can act as apical targeting signals, we investigated whether the glycoproteins of HRSV were involved in its directional targeting and release. We generated recombinant viruses with each of the three glycoprotein genes deleted individually or in groups. Each deleted gene was replaced with a reporter gene to maintain wild-type levels of gene expression. The effects of deleting the glycoprotein genes on apical maturation and on targeting of individual proteins in polarized epithelial cells were examined by using biological, biochemical, and microscopic assays. The results of these studies showed that the HRSV glycoproteins are not required for apical maturation or release of the virus. Further, deletion of one or more of the glycoprotein genes did not affect the intracellular targeting of the remaining viral glycoproteins or the nucleocapsid protein to the apical membrane.     

Regulation of membrane trafficking in polarized epithelial cells

Polarized epithelial cells continuously sort transmembrane proteins to either apical or basolateral plasma membrane domains. Research in recent years has made tremendous progress in understanding the molecular mechanisms of the major pathways to either basolateral or apical domain. This understanding will help us elucidating how these pathways are interconnected in ensuring maintenance of cell polarity and integrity of epithelial monolayers.     

Annexin II is required for apical transport in polarized epithelial cells

The sorting of apical proteins comprises an initial recognition step in the trans Golgi network and a final partitioning of the apical pool of proteins into at least two different types of vesicular carriers. One criteria of these carriers is the association or non-association of the protein content with lipid rafts. We have previously characterized a population containing the raft-associated sucrase-isomaltase-carrying vesicles (SAVs) and another one, the non-raft-associated lactase-phlorizin hydrolase-carrrying vesicles (LAVs) that are targeted separately to the apical membrane. Here, we demonstrate biochemically and by employing confocal laser microscopy that the annexin II-S100A10 complex is a component of SAVs and is absent from LAVs. The unequivocal role of annexin II in the apical targeting of SI is clearly demonstrated when down-regulation of this protein by annexin II-specific small interfering RNA drastically decreases the apical delivery of SI in the epithelial cell line Madin-Darby canine kidney. The annexin II-S100A10 complex plays therefore a crucial role in routing SAVs to the apical membrane of epithelial cells.     

Mechanisms of apical protein sorting in polarized thyroid epithelial cells.

The process leading to thyroid hormone synthesis is vectorial and depends upon the polarized organization of the thyrocytes into the follicular unit. Thyrocyte membrane proteins are delivered to two distinct domains of the plasma membrane using apical (AP) and basolateral (BL) sorting signals. A recent hypothesis for AP sorting proposes that apically destined proteins cluster with glycosphingolipids (GSLs) and cholesterol, into microdomains (or rafts) of the Golgi membrane from which AP vesicles originate. In MDCK cells the human neurotrophin receptor, p75hNTR, is delivered to the AP surface through a sorting signal, rich in O-glycosylated sugars, identified in its ectodomain. We have investigated whether this signal is functional in the thyroid-derived FRT cell line and whether p75hNTR clusters into lipid rafts to be sorted to the AP membrane. We found that p75hNTR is apically delivered via a direct pathway and does not associate with rafts during its transport to the surface of FRT cells. Therefore, although the same signal could be recognized by different cell types thyroid cells may possess a tissue-specific sorting machinery.     

Peeling as a novel, simple, and effective method for isolation of apical membrane from intact polarized epithelial cells

Apical membrane of polarized epithelial cells is generally isolated by physicochemical methods, that is, precipitation with polyethylene glycol (PEG) or MgCl₂ followed by differential centrifugation or sucrose density gradient centrifugation. However, these protocols are considerably sophisticated and frequently accompanied by impurities (e.g., contaminations of basolateral membrane and intracellular organelles), particularly by inexperienced investigators. We have developed a simple and effective method for isolation of apical membrane from intact polarized renal tubular epithelial cells. On the basis of hydrous affinity and/or ionic interaction, the apical membrane could be efficiently peeled from the cells by four different materials—Whatman filter paper, nitrocellulose membrane, cellophane, and glass coverslip—all of which are available in most research laboratories. Phase-contrast and laser-scanning confocal microscopic examinations using anti-ZO-1 antibody showed that other parts of the cells, particularly tight junction complex, remained intact after peeling by all four of these surfaces. Western blot analyses of gp135 (apical membrane marker) and of Na⁺/K⁺-ATPase, LAMP-2, COX-4, and calpain-1 (markers of basolateral membrane, lysosome, mitochondria, and cytosolic compartment, respectively) revealed that peeling with Whatman filter paper and glass coverslip was most and second-most effective, respectively, without any contaminations from basolateral membrane and other intracellular organelles that could be detected in the samples isolated by peeling with nitrocellulose membrane and cellophane and by conventional methods (i.e., precipitation with PEG or MgCl₂ followed by differential centrifugation or sucrose density gradient centrifugation). Our physical method is very simple, easy to follow (even by inexperienced investigators), time-saving, and cost-effective with a higher efficiency (as compared with conventional methods) for isolation of apical membrane from polarized epithelial cells.     

Vectorial targeting of an endogenous apical membrane sialoglycoprotein and uvomorulin in MDCK cells

We studied the cell-surface delivery pathways of newly synthesized membrane glycoproteins in MDCK cells and for this purpose we characterized an endogenous apical integral membrane glycoprotein. By combining a pulse-chase protocol with domain-selective cell-surface biotinylation, immune precipitation, and streptavidin-agarose precipitation (Le Bivic et al. 1989. Proc. Natl. Acad. Sci USA. 86:9313-9317), we followed the appearance at the cell surface of a major apical sialoglycoprotein, gp114, a basolateral protein, uvomorulin, and a transcytosing protein, the polyimmunoglobulin receptor (pIg-R). We determined that both gp114 and uvomorulin appeared to be delivered directly to their respective surface, with mistargeting levels of 8 and 2%, respectively. Using the same technique, the pIg-R was first detected on the basolateral domain and then on the apical domain, to be finally released into the apical medium, as described (Mostov, K. E., and D. L. Deitcher. 1986. Cell. 46:613-621). To directly determine whether the gp114 pool present on the basolateral surface was a precursor of the apical gp114, we compared it with the equivalent pIg-R pool, by labeling with sulfo-NHS-SS-biotin, a cleavable, tight junction-impermeable probe, and by following the fraction of this probe that became resistant to basal glutathione and accessible to apical glutathione during incubation at 37 degrees C. We found that, contrary to pIg-R, basolateral gp114 was poorly endocytosed and was not transcytosed to the apical side. These results demonstrate that an endogenous apical integral membrane glycoprotein of Madin-Darby canine kidney cells is sorted intracellularly and is vectorially targeted to the apical surface.     

Microtubule perturbation inhibits intracellular transport of an apical membrane glycoprotein in a substrate-dependent manner in polarized Madin-Darby canine kidney epithelial cells.

The effects of microtubule perturbation on the transport of two different viral glycoproteins were examined in infected Madin-Darby canine kidney (MDCK) cells grown on both permeable and solid substrata. Quantitative biochemical analysis showed that the microtubule-depolymerizing drug nocodazole inhibited arrival of influenza hemagglutinin on the apical plasma membrane in MDCK cells grown on both substrata. In contrast, the microtubule-stabilizing drug taxol inhibited apical appearance of hemagglutinin only when MDCK cells were grown on permeable substrata. On the basis of hemagglutinin mobility on sodium dodecyl sulfate gels and its sensitivity to endo H, it was evident that nocodazole and taxol arrested hemagglutinin at different intracellular sites. Neither drug caused a significant increase in the amount of hemagglutinin detected on the basolateral plasma membrane domain. In addition, neither drug had any noticeable effect on the transport of the vesicular stomatitis virus (VSV)-G protein to the basolateral surface. These results shed light on previous conflicting reports using this model system and support the hypothesis that microtubules play a role in the delivery of membrane glycoproteins to the apical, but not the basolateral, domain of epithelial cells.     

The ammonium transporter RhBG: requirement of a tyrosine-based signal and ankyrin-G for basolateral targeting and membrane anchorage in polarized kidney epithelial cells

RhBG is a nonerythroid member of the Rhesus (Rh) protein family, mainly expressed in the kidney and belonging to the Amt/Mep/Rh superfamily of ammonium transporters. The epithelial expression of renal RhBG is restricted to the basolateral membrane of the connecting tubule and collecting duct cells. We report here that sorting and anchoring of RhBG to the basolateral plasma membrane require a cis-tyrosine-based signal and an association with ankyrin-G, respectively. First, we show by using a model of polarized epithelial Madin-Darby canine kidney cells that the targeting of transfected RhBG depends on a YED motif localized in the cytoplasmic C terminus of the protein. Second, we reveal by yeast two-hybrid analysis a direct interaction between an FLD determinant in the cytoplasmic C-terminal tail of RhBG and the third and fourth repeat domains of ankyrin-G. The biological relevance of this interaction is supported by two observations. (i) RhBG and ankyrin-G were colocalized in vivo in the basolateral domain of epithelial cells from the distal nephron by immunohistochemistry on kidney sections. (ii) The disruption of the FLD-binding motif impaired the membrane expression of RhBG leading to retention on cytoplasmic structures in transfected Madin-Darby canine kidney cells. Mutation of both targeting signal and ankyrin-G-binding site resulted in the same cell surface but nonpolarized expression pattern as observed for the protein mutated on the targeting signal alone, suggesting the existence of a close relationship between sorting and anchoring of RhBG to the basolateral domain of epithelial cells.     

Single-channel recordings of apical membrane chloride conductance in A6 epithelial cells

Summary The apical membrane of epithelial cells from the A6 cell line grown on impermeable substrata was studied using the patch-clamp technique. We defined the apical membrane as that membrane in contact with the growth medium. In about 50% of the patches, channels with single-unit conductances of 360±45 pS in symmetrical 105mm NaCl solutions, and characteristic voltage-dependent inactivation were observed. Using excised membrane patches and varying the ionic composition of the bathing medium, we determined that the channels were anion selective, with a permeability ratio for Cl^– over Na⁺ of about 91, calculated from the reversal potential using the constantfield equation. The channel was most active at membrane potentials between ±20 mV and inactivated, usually within a few seconds, at higher potentials of either polarity. Reactivation from this inactivation was slow, sometimes requiring minutes. In addition to its fully open state, the channel could also enter a flickering state, which appeared to involve rapid transitions to one or more submaximal conductance levels. The channel was inhibited by the disulfonic stilbene SITS in a manner characteristic of reversible open-channel blockers.