Mechanisms of apical protein sorting in polarized thyroid epithelial cells

Mechanisms of apical protein sorting in polarized thyroid epithelial cells

Biochimie 81 (1999) 347−353 © Société française de biochimie et biologie moléculaire / Elsevier, Paris Mechanisms of apical protein sorting in polari...

166KB Sizes 1 Downloads 22 Views

Biochimie 81 (1999) 347−353 © Société française de biochimie et biologie moléculaire / Elsevier, Paris

Mechanisms of apical protein sorting in polarized thyroid epithelial cells Concetta Lipardi, Lucio Nitsch, Chiara Zurzolo* Centro di Endocrinologia ed Oncologia Sperimentale del CNR-Dipartimento di Biologia e Patologia Cellulare e Molecolare, Università degli Studi di Napoli ‘Federico II’, Via Pansini 5, 80131 Naples, Italy (Received 6 July 1998; accepted 30 November 1998) Abstract — 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. © Société française de biochimie et biologie moléculaire / Elsevier, Paris thyroid polarity / lipid raft / p75hNTR / apical sorting / FRT cell line

1. Introduction The functional and morphological unit of the thyroid gland is the follicle. This is formed by a single layer of thyrocytes [1] which, like all epithelial cells, have their plasma membrane divided by tight junctions in two domains, the apical (AP) and basolateral (BL) domains, with distinct protein and lipid composition [1–3]. While the AP domain faces the follicular lumen into which the newly synthesized thyroglobulin is secreted and accumulated, thereby serving as the substrate for iodination and hormonogenesis [4, 5], the basolateral domain faces the interstitium/bloodstream from which thyrocytes take up iodide and release thyroid hormones [6–8]. The process leading to thyroid hormone secretion is vectorial and depends upon the polarized organization of the thyrocytes. The establishment and maintenance of this organization, and subsequently its functionality, is ensured by the continuous and regulated intracellular sorting of different lipids and proteins to the cell surface [9]. In the model system of MDCK cells, derived from dog kidney, it has been shown that plasma membrane proteins are exported to the Golgi complex where, at the trans-Golgi network (TGN), they are separated into AP and BL * Correspondence and reprints Abbreviations: AP, apical; BL, basolateral; TX-100, Triton X-100; GSLs, glycosphingolipids; GPI-anchored proteins, glycosylphosphatidylinositol-anchored proteins; TSH, thyrotropin; p75hNTR, p75 human neurotrophin receptor.

vesicles [10] through specific sorting information. While BL sorting signals have been identified as short amino acid sequences within the cytoplasmic tails of several transmembrane proteins [11, 12], sorting to the AP surface involves more than one determinant. The current model for AP sorting proposes that, within the Golgi bilayer, glycosphingolipids (GSLs) and cholesterol cluster into microdomains, which function as sorting platforms (socalled ‘rafts’) for proteins destined to the AP surface [13]. The rafts are insoluble in Triton X-100 (TX-100) at 4 °C and are able to float into lighter fractions on sucrose density gradients. Glycosylphosphatidyl inositol (GPI)anchored proteins possibly use their GPI-anchor to cluster with rafts and to be apically delivered [14, 15]. Association with rafts and AP sorting can also be mediated by the transmembrane domains, as shown in the cases of the influenza virus proteins neuraminidase [16] and hemagglutinin [17]. In addition, the mannose-rich core of N-glycans or a juxtamembrane region rich in O-glycosylated residues are involved in AP sorting of secretory and/or transmembrane proteins [18, 19]. AP and BL sorting signals are decoded by a specific molecular machinery. Defects in their recognition create abnormalities in protein transport and may specifically contribute, in the case of thyroid gland, to different pathologies [20]. For example, a defect in the targeting of thyroglobulin to the AP extracellular space may lead to metabolic disturbances, abnormal development, and goiter [21]. Furthermore, loss of epithelial cell polarity, and therefore alteration of protein sorting to the surface, is an

348 early event in tumor development; indeed MDCK cells transfected by the v-K-ras oncogene show an altered polarity [22]. At least two distinct pathways, the direct and the indirect route (or transcytosis), are used by different cell lines to target newly synthesized proteins to the plasma membrane [9, 23, 46]. MDCK cells mainly sort proteins from the TGN directly to the AP or BL surface [11]. The human intestinal epithelial cell line Caco-2 uses both direct and indirect pathways depending upon the protein [24, 25]. Hepatocytes do not have a direct AP pathway, therefore they deliver all AP proteins to the BL membrane from which they are endocytosed and resorted to the AP domain via transcytosis [26, 27]. Studies on isolated thyroid follicles have greatly contributed to the general concept of how cell polarity is established and maintained. Rat thyroid follicles, maintained in suspension culture in the normal [28] and inverted [29] configuration have been used to correlate differentiated functional properties of the thyroid (e.g., iodide trapping, thyroglobulin iodination, and response to thyrotropin (TSH)) with the polarized expression, respectively, of the iodide pump, the thyroperoxidase enzyme, and the TSH receptor on the surface [30]. More recently, a rat thyroid cell line (FRT) has been shown to have both morphological and functional polarity [31]. The targeting phenotype of these cells closely resembles MDCK cells. AP and BL proteins are delivered from the TGN to the surface mainly via a direct pathway, although a transcytotic route can also be used by the FRT cells, as evidenced during the maturation of the polarized monolayer in cultured cells [32]. Like other epithelial cells, in FRT cells the vesicular stomatitis virus buds from the BL surface while the influenza virus buds from the AP membrane [33]. On the contrary, significant differences with other polarized cells have been described in the case of polarized budding of the Semliki Forest virus [33] and for the polarity of transfected OKT8 lymphocyte antigen (Zurzolo and Nitsch, unpublished results). Furthermore, the surface distribution of endogenous BL transmembrane proteins such as Na+ K+ ATPase, integrins, and uvomorulin is conserved among FRT [34, 35] and other epithelial cells [36], and a BL sorting signal identified in the tail of the mutated p75 human neurotrophin receptor (p75hNTR) in MDCK cells is also functional in FRT cells (Lipardi et al., submitted) suggesting that a conserved mechanism for recognizing BL sorting determinants of transmembrane proteins is active in different epithelia. Interestingly, endogenous GPI-anchored proteins, which are mainly sorted to the AP surface in primary thyrocyte cultures [37] and in other epithelia [15], are preferentially delivered to the BL surface in FRT cells [38] indicating that the sorting machinery can vary considerably between different epithelial cell types. Since most of the AP sorting signals of transmembrane and secretory proteins had been identified in MDCK and

Lipardi et al. Caco-2 cells, we asked whether they were recognized by FRT cells and whether in these cells apically sorted proteins utilize lipid rafts to be targeted to the AP surface. In order to address these questions we have studied the plasma membrane distribution of wild type p75hNTR and two derived mutants (XI and SEC) in transfected FRT cells. The XI mutant has a deletion in the cytoplasmic tail which leaves only five amino acids, while the secretory mutant (SEC) contains only the ectodomain of the wild type protein. Both these proteins are apically localized in Caco-2 (Monlauzer et Le Bivic, in press) and MDCK cells [24] and in the latter they use an apical sorting signal recently identified in the juxtamembrane region of the ectodomain rich in O-glycosylated residues [19]. We found that the AP sorting signal present within the ectodomain of p75hNTR was also functional in FRT cells. This signal mediates direct delivery of both the wild type p75hNTR and the XI mutant to the AP membrane of FRT cells, confirming that FRT and MDCK, but not Caco-2 cells, use similar pathways to target transmembrane proteins to the AP surface. Furthermore, we analyzed the insolubility of this receptor in TX-100 at 4 °C and we found it to be completely soluble. These data indicate that AP sorting signals are conserved within different cell lines and that in FRT cells transmembrane proteins do not require lipid rafts to be delivered to the AP surface. Therefore a tissue-specific sorting machinery to deliver proteins to the AP surface might exist in thyroid cells. 2. Materials and methods 2.1. Reagents Cell culture reagents were purchased from Gibco Laboratories (Grand Island, NY, USA). Protein A-Sepharose was from Pharmacia Fine Chemicals (Uppsala, Sweden). Sulfosuccinimidyl-6-(biotinamide) hexanoate (NHS-LCbiotin), sulfo-N-hydroxyl-succinimido-biotin (NHS-SSbiotin), and streptavidin-agarose were purchased from Pierce Chemical Co. (Rockford, IL, USA). All other reagents were obtained from Sigma Chemical Co. (St. Louis, MO, USA). 2.2. Cells, antibodies and constructs FRT cells were grown in Coon’s modified Ham’s F12 media with 5% fetal bovine serum (Gibco, Paisley, UK), penicillin (50 U/mL) and streptomycin (32 µg/mL). 2 × 106 cells were seeded on Transwell chambers (Costar, Cambridge, USA) and cultured for 5 days. Monolayer integrity was confirmed by measuring transepithelial resistance (Millicell-ERS, Millipore, Bedford, USA). Antibodies were as follows: mouse monoclonal antibody ME 20.4 against the ectodomain of human neurotrophin receptor, hNTR (gift of A. Le Bivic), rabbit affinity purified

Lipid rafts and apical sorting of p75hNTR


antibody to whole molecule mouse IgG (Cappel, Organon Teknika Corp., West Chester, PA, USA). Full length hNTR (WT), XI mutant, SEC mutant were gift of A. Le Bivic and were obtained as previously described [39]. FRT cells were transfected as described [33]. Neomycin resistant clones were screened for hNTR expression by indirect immunofluorescence. 2.3. Biotinylation and targeting assay Cells grown on filters for 5 days were incubated for 30 min in DME medium without cysteine and pulsed in the same medium containing 1 mCi/mL [35S] cysteine (Amersham, Buckinghamshire, UK; 1000 Ci/mmol) overnight (at steady state), or for 20 min (targeting assay, followed by a chase for various lengths of time). Proteins on either the AP or BL plasma membrane domains were biotinylated at 4 °C using NHS-LC-biotin as described [43]. After cell lysis, biotinylated proteins were immunoprecipitated with ME 20.4 antibody (1:200) conjugated to protein A-Sepharose. Immunoprecipitates were washed as previously described and subsequently precipitated with streptavidin agarose-beads [41]. Immunoprecipitates were then run on 10% SDS-polyacrylamide gels (PAGE) and revealed by fluorography. 2.4. Pulse-chase and TX-100 extraction TX-100 extractability during pulse-chase experiments was performed as previously described [42]. Cells, after a pulse in DME medium without cysteine containing 1 mCi/mL of [35S] cysteine for 5 min, were incubated in chase medium (DME medium containing 10% FBS and 100 × met and cys) for different times. After each time point, cells were lysed in buffer containing 1% TX-100, collected and centrifuged at 140 000 rpm for 2 min at 4 °C. Supernatants, representing the soluble material, were removed and pellets were solubilized in buffer containing 0.1% SDS. Both soluble and insoluble materials were adjusted to 0.1% SDS before immunoprecipitation. 2.5. Sucrose density gradients Sucrose gradient analysis of TX-100 extracts was performed as previously described [42]. Cell were pulsed in DME medium containing 1 mCi/mL [35S] cysteine for 30 min, chased for 3 h and lysed for 10 min in buffer containing 1% TX-100 on ice. Cell lysates, brought to 40% sucrose, were placed at the bottom of a centrifuge tube and covered by a step sucrose gradient (5–35% in TNE). After centrifugation in a SW41 rotor (Beckman) at 39 000 rpm for 18–20 h, 1 mL fractions were collected from the top. Proteins were immunoprecipitated with anti-p75hNTR antibody and resolved on SDS-PAGE.

Figure 1. Schematic representation of WT and p75hNTR mutants. The stippled box represents the ectodomain. The black box represents the transmembrane domain. The empty box represents the cytoplasmic tail. WT hNTR is a full length cDNA, whereas the XI mutant has a deletion of 150 amino acids on the COOH-terminal side. The SEC mutant contains only the ectodomain.

3. Results 3.1. FRT cells recognize the AP sorting signal present in the ectodomain of p75hNTR p75hNTR, a type I transmembrane glycoprotein, is apically localized in transfected MDCK cells [24] using the juxtamembrane region of the ectodomain rich in O-glycosylated residues [19]. To determine whether this region was also recognized by FRT cells as an AP signal we stably transfected this cell line with cDNAs encoding the full length p75hNTR and the XI mutant generated by a large deletion that leaves only five amino acids of the cytoplasmic tail (figure 1). Different clones expressing the two constructs were selected by indirect immunofluorescence (data not shown) and their surface distribution was determined by selective biotinylation from the AP or BL surface of cells grown on filter [40, 42]. Biotinylated proteins were precipitated with a specific monoclonal antibody and with streptavidin beads and detected by SDS-PAGE. As shown in figure 2A, both the WT and the XI mutant were localized predominantly to the AP domain. Quantitative scanning densitometry of two independent experiments indicated that more than 80% of the WT and 90% of the XI mutant were apically localized in the different clones. To determine whether the p75hNTR ectodomain was directly responsible for its AP distribution we expressed this domain (SEC mutant) (figure 1) in FRT cells and analyzed its distribution in the AP and BL media of labeled FRT clones grown on filters. We found that this mutant was secreted almost exclusively into the AP medium (figure 2B), indicating that the information for AP sorting was contained within the ectodomain of the protein, as previously shown in MDCK cells. 3.2. Surface delivery of newly synthesized WT and XI mutant of p75hNTR To test whether the AP localization of p75hNTR in FRT cells was due to direct or transcytotic delivery, the arrival


Lipardi et al.

Figure 3. Surface delivery of newly synthesized WT and XI mutant of p75hNTR in transfected FRT cells. Cells grown on filters were pulsed for 20 min with [35S]cysteine and then chased for the times indicated. Newly synthesized proteins at the cell surface were detected by selective biotinylation from the apical (AP) or basolateral (BL) sides. After cell lysis biotinylated proteins were immunoprecipitated using the mAb anti p75hNTR, reprecipitated with streptavidin beads and analyzed by SDS-PAGE and fluorography.

Figure 2. Distribution of WT and p75hNTR mutants in transfected FRT cells. FRT cells grown on filters for 5 days were pulsed with [35S]cysteine at 37 °C overnight. A. Surface proteins were biotinylated from the apical (A) or basolateral (B) side. After cell lysis, proteins were immunoprecipitated with a p75hNTR monoclonal antibody (1:200). Biotinylated proteins were then reprecipitated with streptavidin beads and analyzed by 10% SDSPAGE and autoradiography. B. Apical (A) and basolateral (B) media were collected and immunoprecipitated with anti p75hNTR mAb and analyzed by SDS-PAGE and fluorography.

of newly synthesized proteins to the surface was detected by a biotin targeting assay [40]. Cells grown on filters were pulsed briefly with [35S] cysteine, chased for 0, 30, and 100 min, and subsequently biotinylated from the AP or the BL side. After cell lysis, biotinylated proteins were immunoprecipitated with the specific antibody, reprecipitated with streptavidin beads and analyzed by SDS-PAGE fluorography (figure 3). We found that both the WT and the XI mutant were directly delivered to the AP surface with similar kinetics, in fact they were expressed on the AP membrane after 30 min of chase without passing through the BL domain (figure 3). 3.3. p75hNTR is totally excluded from glycosphingolipid and cholesterol rafts To test whether glycosphingolipid-cholesterol rafts could be involved in the AP sorting of p75hNTR in FRT

Lipid rafts and apical sorting of p75hNTR


Figure 4. Pulse chase analysis of WT of p75hNTR solubility in TX-100 in FRT cells. FRT stably expressing WT of p75hNTR were grown to confluence and then pulsed for 5 min with [35S]cysteine, followed by incubation in chase medium for the indicated times. After extraction in TX-100 at 4 °C, both soluble (S) and insoluble (I) fractions were collected after centrifugation, immunoprecipitated with the mAb anti p75hNTR and analyzed by SDS-PAGE and fluorography.

cells, we performed a pulse-chase experiment and determined whether p75hNTR was incorporated in TX-100 insoluble fractions. We found that this receptor was mainly soluble in this detergent at the different times of chase, as shown in figure 4. We then analyzed the TX-100 insoluble fractions on sucrose density gradients as described in Materials and methods. We found that the wild type p75hNTR was enriched in fractions 8–12 (35–40% sucrose) at the bottom of the gradient, as expected for a soluble protein (figure 5). These data indicated that in FRT cells p75hNTR does not need to be incorporated into lipid rafts to be sorted to the AP domain. 4. Discussion Some of the early experiments using viral chimeric proteins suggested that an AP sorting signal was contained within the ectodomain of apically sorted proteins (probably within the three-dimensional patches of highly folded proteins) [43]. Because of the inability to identify specific AP determinants for many years, another hypothesis has been that AP sorting of proteins could occur by default [11]. More recently, it has been found that different types of proteins use different motifs and specific mechanisms to be delivered to the AP surface. N-glycans [18, 44] or O-glycosylated sugars [19] are required for the AP delivery of secretory and/or transmembrane proteins. Furthermore, the GPI anchor has been suggested to act as an

AP sorting signal in different cell lines [15, 45] via clustering with GSLs and cholesterol in membrane microdomains so called rafts [2, 13]. In MDCK cells the neuramidase [16] and hemagglutinin of the influenza virus [17] are apically sorted using their transmembrane domain to cluster with rafts, and the majority of AP proteins, independently of their putative signal (e.g., oligosaccarides, GPI-anchor, TM domain) are insoluble in TX-100 [13, 17]. Therefore the current hypothesis for AP sorting suggests that inclusion of proteins in lipid rafts is a prerequisite for their AP delivery. We show here that p75hNTR, a type I transmembrane protein, is localized to the AP membrane in FRT cells, as was previously demonstrated in MDCK and Caco-2 cells [24] (Le Bivic, personal communication). A deletion of the cytoplasmic tail that leaves only five amino acids (mutant XI) does not alter the AP polarity of this receptor in either MDCK or FRT cells. We have also found that the secretory form (SEC), which consists of only the ectodomain of hNTR, is mainly secreted in the AP medium (figure 2B) in FRT cells. All these data suggest that the AP sorting motif identified within the ectodomain of this protein, that is recognized by MDCK cells [19], is functional also in FRT cells. In fact, although we have not directly investigated the role of O-glycan for the AP sorting of p75hNTR in FRT cells our previous data indicated that AP transport in FRT cells does not occur by default [45].


Lipardi et al.

Figure 5. Purification of WT of p75hNTR enriched fractions on sucrose density gradients. FRT cells expressing wild type p75hNTR were labeled with [35S]cysteine for 30 min and then chased for 3 h. Cells were lysed in TNE/TX-100 buffer and then run through a step 5–40% sucrose gradient. Fractions of 1 mL were collected from the top to bottom after centrifugation at equilibrium, and p75hNTR was immunoprecipitated.

We also found that p75hNTR remains completely soluble in TX-100 and is not able to float into lighter fractions on sucrose density gradients, indicating that it does not interact with GSLs and cholesterol rafts for its AP sorting. These data reinforce the idea that an AP sorting signal is present within the ectodomain of p75hNTR and that its transmembrane domain, which would be the one interacting with GSLs and cholesterol rafts, does not play a role in its AP delivery. Furthermore we show that a direct pathway is used by p75hNTR to enrich it on the AP surface (figure 3), confirming the similarity of the targeting phenotype of FRT and MDCK cells. However, one could formulate several questions as to the AP specific determinants: the mechanisms by which carbohydrates mediate to cell surface transport, the system by which the clustering of GSLs and GPI-anchored proteins occurs and leads to AP sorting, and their incorporation into vesicles. In the light of the current knowledge and of the fact that in FRT cells AP membrane proteins are apically delivered independently from glycolipid and GPI-anchored protein sorting (this paper) [41], it is very difficult to believe that one single AP mechanism for AP sorting exists [46].

5. Conclusion Targeting to the AP membrane of epithelial cells could be mediated by two independent mechanisms, one involving association with glycolipid-cholesterol and the other by N-linked and/or O-linked glycans. The capacities and/or the directionality of these pathways could vary from one epithelial tissue to another, leading to the

plasticity of AP membrane protein sorting in different epithelial cell types. One could hypothesize that different determinants sort different proteins to particular cell surface domains in a tissue specific manner that depends on the particular needs of a given cell type. Acknowledgments We thank Franco D’Agnello and Mario Berrdone for the art work. This work was supported by grants from Associazione Italiana per la Ricerca sul Cancro (AIRC), and from Ministerio Universita` e Ricerca Scientifica e Tecnologica (MURST) cofinanziamento programmi di ricerca di intersse nazionale to CZ.

References [1] Wollman S.H., Turnover of plasma membrane in thyroid epithelium and review of evidence for the role of micropinocytosis, Eur. J. Cell. Biol. 50 (1989) 247–256. [2] VanMeer G., Simons K., Lipid polarity and sorting in epithelial cells, J. Cell. Biochem. 36 (1988) 51–58. [3] Rodriguez-Boulan E., Powell S.K., Polarity of epithelial and neuronal cells, Annu. Rev. Cell Biol. 8 (1992) 395–427. [4] Ericson L.E., Exocytosis and endocytosis in the thyroid follicle cell, Mol. Cell. Endocrinol. 22 (1981) 1–4. [5] Herzog V., Berndorfer U., Saber Y., Isolation of insoluble secretory product from bovine thyroid: extracellular storage of thyroglobulin in covalently cross-linked form, J. Cell Biol. 118 (1992) 1071–1083. [6] Golstein P., Abramow M., Dumont J.E., Beauwens R., The iodide channel of the thyroid: a plasma membrane vesicle study, Am. J. Physiol. 263 (1992) C590–C597 [7] Dai G., Levy O., Carrasco N., Cloning and characterization of the thyroid iodide transporter, Nature 379 (1996) 458–460.

Lipid rafts and apical sorting of p75hNTR [8] Bjorkman U., Ekholm R., Accelerated exocytosis and H2O2 generation in isolated thyroid follicles enhance protein iodination, Endocrinology 122 (1988) 488–494. [9] Drubin D.G., Nelson W.J., Origins of cell polarity, Science 84 (1996) 335–344. [10] Wandinger-Ness A., Bennett M.K., Antony C., Simons K., Distinct transport vesicles mediate the delivery of plasma membrane proteins to the apical and basolateral domains of MDCK cells, J. Cell. Biol. 111 (1990) 987–1000. [11] Matter K., Mellman I., Mechanisms of cell polarity: sorting and transport in epithelial cells, Curr. Opin. Cell. Biol. 6 (1994) 545–554. [12] Mellman I., Endocytosis and molecular sorting, Annu. Rev. Cell. Dev. Biol. 12 (1996) 575–625. [13] Simons K., Ikonen E., Functional rafts in cell membranes, Nature 387 (1997) 569–572. [14] Brown D.A., Crise B., Rose J.K., Mechanism of membrane anchoring affects polarized expression of two proteins in MDCK cells, Science 245 (1989) 1499–1501. [15] Lisanti M.P., Caras I.W., Davitz M.A., Rodriguez-Boulan E., A glycophospholipid membrane anchor acts as an apical targeting signal in polarized epithelial cells, J. Cell. Biol. 109 (1989) 2145–2156. [16] Kundu A., Avalos R.T., Sanderson C.M., Nayak D.P., Transmembrane domain of influenza virus neuramidase, a type II protein, possesses an apical sorting signal in polarized MDCK cells, J. Virol. 70 (1996) 6508–6515. [17] Scheiffele P., Peranen J., Simons K., Interaction of influenza virus hemagglutinin with sphingolipid-cholesterol membrane domains via its transmembrane domain, EMBO J. 16 (1997) 5501–5508. [18] Scheiffele P., Peranen J., Simons K., N-glycan as apical sorting signals in epithelial cells, Nature 378 (1995) 96–98. [19] Yeaman C., Le Gall A.H., Baldwin A.N., Monlauzer L., Le Bivic A., Rodriguez-Boulan E., The O-glycosylated stalk domain of neurotrophin receptors is required for apical sorting of neurotrophin receptors in polarized MDCK cells, J. Cell. Biol. 139 (1997) 929–940. [20] Mauchamp J., Chabaud O., Chambard M., Gerard C., Penel C., Verrier B., Polarized properties of thyroid cells: a study with cultured porcine cells, Acta Endocrinol. (suppl.) 281 (1987) 220–224. [21] Grollman E.F., Doi S.Q., Weiss P., Ashwell G., Wajchenberg B.L., Medeiros-Neto G., Hyposialylated thyroglobin in a patient with congenital goiter and hypothyroidism, J. Clin. Endocrinol. Metab. 74 (1992) 43–48. [22] Schoenenberger C.A., Zuk A., Kendall D., Matlin K.S., Multilayering and loss of apical polarity in MDCK cells transformed with viral K-ras, J. Cell. Biol. 112 (1991) 873–889. [23] Eaton S., Simons K., Apical, basal, and lateral cues for epithelial polarization, Cell 82 (1995) 5–8. [24] Le Bivic A., Sambuy Y., Patzak A., Patil N., Chao M., RodriguezBoulan E., An internal deletion in the cytoplasmic tail reverses the apical localization of human NGF receptor in transfected MDCK cells, J. Cell. Biol. 115 (1991) 607–618. [25] Matter K., Brauchbar M., Bucher K., Hauri H.P., Sorting of endogenous plasma membrane proteins occurs from two sites in cultured human intestinal epithelial cells (Caco-2), Cell 60 (1990) 429–437. [26] Bartles J.R., Feracci H.M., Stieger B., Hubbard A.L., 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 (1987) 1241–1251. [27] Bartles J.R., Hubbard A.L., Plasma membrane protein sorting in epithelial cells: Do secretory pathways hold the key?, Trends Biochem. Sci. 13 (1988) 181–184.

353 [28] Nitsch L., Wollman S.H., Suspension culture of separated follicles consisting of differenziated thyroid epithelial cells, Proc. Natl. Acad. Sci. USA 77 (1980) 472–476. [29] Nitsch L., Wollman S.H., Ultrastructure of intermediate stages in polarity reversal of thyroid epithelium in follicles in suspension culture, J. Cell. Biol. 86 (1980) 875–880. [30] Tacchetti C., Zurzolo C., Monticelli A., Nitsch L., Functional properties of normal and inverted rat thyroid follicles in suspension culture, J. Cell. Physiol. 126 (1986) 93–98. [31] Nitsch L., Tramontano D., Ambesi-Impiombato F.S., Morphological and functional polarity of an epithelial thyroid cell line, Eur. J. Cell. Biol. 38 (1985) 57–66. [32] Zurzolo C., Le Bivic A., Quaroni A., Nitsch L., Rodriguez-Boulan E., Modulation of transcytotic and direct targeting pathways in a polarized thyroid cell line, EMBO J. 11 (1992) 2337–2344. [33] Zurzolo C., Polistina C., Saini M., Gentile R., Aloj L., Migliaccio G., Bonatti S., Nitsch L., Opposite polarity of virus budding and of viral envelope glycoprotein distribution in epithelial cells derived from different tissues, J. Cell. Biol. 117 (1992) 551–564. [34] Zurzolo C., Rodriguez-Boulan E., Delivery of Na+, K+, -ATPase in polarized epithelial cells, Science 260 (1993) 550–556. [35] Calì G., Retta S.F., Negri R., Damiano I., Gentile R., Tarone G., Nitsch L., Garbi C., β1B integrin interferes with matrix assembly but not with confluent monolayer polarity, and alters some morphogenetic properties of FRT epithelial cells, Eur. J. Cell. Biol. 75 (1998) 107–117. [36] Caplan M.J., Palade G.E., Jamieson J.D., Newly synthesized Na, K-ATPase alpha-subunit has no cytosolic intermediate in MDCK cells, J. Biol. Chem. 261 (1986) 2860–2865. [37] Kuliawat R., Lisanti M.P., Arvan P., Polarized distribution and delivery of plasma membrane proteins in thyroid follicular epithelial cells, J. Biol. Chem. 270 (1995) 2478–82. [38] Zurzolo C., Lisanti M.P., Caras I.W., Nitsch L., Rodriguez-Boulan E., Glycosylphosphatidylinositol-anchored proteins are preferentially targeted to the basolateral surface in Fischer rat thyroid epithelial cells, J. Cell. Biol. 121 (1993) 1031–1039. [39] Hempstead B.L., Patil N., Thiel B., Chao M.V., Deletion of the cytoplasmic sequences of the nerve growth factor receptor leads to loss of high affinity ligand binding, J. Biol. Chem. 265 (1990) 9595–9598 . [40] Zurzolo C., Le Bivic A., Rodriguez-Boulan E., Cell surface biotinylation techniques, in: Cell Biology, Vol. 3, Academic Press, 1994, 188 p. [41] Zurzolo C., van ‘t Hof W., van Meer G., Rodriguez-Boulan E., VIP21/caveolin, glycosphingolipid clusters and the sorting of glycosylphosphatidylinositol-anchored proteins in epithelial cells, EMBO J. 13 (1994) 42–53. [42] Sargiacomo M., Lisanti M., Graeve L., Le Bivic A., RodriguezBoulan E., Integral and peripheral protein composition of the apical and basolateral membrane domains in MDCK cells, J. Membr. Biol. 107 (1989) 277–286. [43] Roth M.G., Gunderson D., Patil N., Rodriguez-Boulan E., The large external domain is sufficient for the correct sorting of secreted or chimeric influenza virus hemagglutinins in polarized monkey kidney cells, J. Cell. Biol. 104 (1987) 769–782. [44] Fiedler K., Simons K., The role of N-glycans in the secretory pathway, Cell 81 (1995) 309–312. [45] Canipari R., Zurzolo C., Polistina C., Aloj L., Calì G., Gentile R., Nitsch L., Polarized secretion of plasminogen activators by epithelial cell monolayers, Biochim. Biophys. Acta 1175 (1992) 1–6. [46] Weimbs T., Low S.H., Chapin S.J., Mostov K.E., Apical targeting in polarized epithelial cells: there’s more afloat than rafts, Trends Cell. Biol. 7 (1997) 393–399.