QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bulbarelli, A. Articles by Borgese, N. Search for Related Content PubMed PubMed Citation Articles by Bulbarelli, A. Articles by Borgese, N.

Trafficking of tail-anchored proteins: transport from the endoplasmic reticulum to the plasma membrane and sorting between surface domains in polarised epithelial cells

Alessandra Bulbarelli^1,*, Teresa Sprocati¹, Massimo Barberi¹, Emanuela Pedrazzini¹ and Nica Borgese^1,2,

¹ Consiglio Nazionale delle Ricerche Cellular and Molecular Pharmacology Center and Department of Medical Pharmacology, University of Milan, Milan, Italy
² Faculty of Pharmacy, University of Catanzaro `Magna Graecia', Catanzaro, Italy
^* Present address: Centre for Study and Research on Obesity, Department of Preclinical Sciences, L. Sacco Hospital, University of Milan, Italy

View larger version (29K): [in a new window]   Fig. 1. Representation of the constructs used in this study. (A) Schematic representation of the basic TA reporter construct, GFP-17. An enhanced version of GFP (oval) is attached at its C-terminus to a linker (filled thin rectangle, L), consisting of the myc epitope followed by a repeated Gly-Ser sequence. This is attached to the entire tail region of rat b(5), which contains the TMD, flanked upstream and downstream by polar sequences (UPS and DPS, respectively). The regions between the indicated unique restriction sites in the cDNA can be easily substituted with paired synthetic oligonucleotides. The expected topology of the construct after insertion into membranes is also indicated. (B) Amino acid sequence (one-letter code) of the Linker and the UPS region present in all the constructs listed in panel (C); within the linker the residues of the myc epitope are in italics. Also shown are the sequences of the myc linker, present in the myc constructs, and the UPS linker, present in a construct in which b(5)'s UPS was deleted (these constructs are not listed in panel C — see Text). (C) Sequences of the TMD and DPS region of GFP TA constructs. The first six constructs have TMDs derived from b(5) in which amino acids have been deleted, inserted or substituted. The TMD of all these constructs is followed by the DPS of b(5). The numbers 14 through 25 in the construct names indicate the length of the TMD as predicted by hydrophilicity analysis with the scale of Engelman et al. (Engelman et al., 1986) over a window of seven residues. The residues that are predicted to be in a hydrophobic environment are shown in boldface. The construct GFP-17 contains the wild-type tail region of b(5). In the construct GFP-17-HH, the number of residues was not changed, but four substitutions (underlined) increase the TMD hydrophobicity (this results in the inclusion of one more residue in the hydrophobic region, as predicted by hydrophilicity analysis). In GFP-ST, the TMD of sialyl transferase replaces the one of b(5). In GFP-22-Nglyc, the DPS of b(5) is connected, via the couple SR, to the N-terminal sequence of bovine opsin, which contains two N-glycosylation consensus sites. The site that is predicted to be sufficiently distant from the bilayer to be used is boxed. In GFP-22-MutNglyc, both N-glycosylation sites of the preceding construct have been eliminated by AsnGln substitution (boldface). In GFP-Syn3 and Syn4, b(5)'s UPS connects to the last 29 residues of rat syntaxin 3 and 4, respectively. For all constructs, the numbers in italics enclosed by parentheses in the left column indicate the hydrophobicity of the TMD region, calculated by using the value of the scale of Engelman et al. (Engelman et al., 1986) for each residue in the sequence between the last upstream charged amino acid and the Arg residue in the DPS or for GFP-Syn 3 and 4, between the last upstream charged amino acid and the C-terminal residue.

View larger version (92K): [in a new window]   Fig. 2. Fluorescence analysis of CV-1 cells transiently transfected with GFP TA proteins. Cells transfected with the indicated constructs were fixed and viewed by conventional fluorescence microscopy. Arrows in panel GFP-19 and GFP-ST indicate regions where surface localisation is visible. Bar, 20 µm.

View larger version (65K): [in a new window]   Fig. 3. GFP-ST localises both to the cell surface and to the Golgi complex. (A) Confocal analysis of CV-1 cells expressing GFP-ST and stained either for surface glycoproteins with Con A (a-c) or immunostained for giantin (d-f) (see Materials and Methods). Single confocal sections viewed under the fluorescein filter for GFP (a, d) and under the Texas Red filter for surface glycoproteins (b) or for the Golgi complex (e) are shown. The resulting merged images are shown in (c) and (f), respectively. The yellow colour indicates colocalisation. Bar, 10 µm. Inset in panels (d-f) is magnified 2.5 times. (B) Golgi localisation of GFP-ST is not affected by the level of expression of the construct and decreases after cycloheximide treatment. Transfected cells were either fixed or first incubated with cycloheximide (50 µg/ml) for 4 hours before fixation. After immunostaining with anti-giantin antibodies, Z series were acquired (see Materials and Methods). Average fluorescence and the percentage of total fluorescence colocalizing with giantin were determined for 25 cells of each group. The mean percentage of GFP in the Golgi is given for groups of cells classified according to their summed average fluorescence intensity over the stack. The numbers inside the columns indicate the number of cells in each group. Gray columns, untreated cells; pink columns, cells treated with cycloheximide; bars indicate the standard deviation.

View larger version (109K): [in a new window]   Fig. 4. TMD-dependent sorting of GFP TA proteins in polarised MDCK cells. (A-C) Filter-grown cells stably expressing GFP-17 were fixed and immunostained for PDI, with the use of Texas-Red-conjugated secondary antibodies. A single confocal section, visualised for GFP (A), PDI (B) or as a merge of GFP and Texas Red (PDI) fluorescence (C) is shown. (D-I) Filter-grown cells stably expressing GFP-22 were either fixed and immunostained for a 58 kDa basolateral protein of MDCK cells (Keller and Simons, 1998) with the use of Texas-Red-conjugated secondary antibodies (D-F) or incubated in the cold with biotinylated ConA administered in the upper compartment of a Transwell filter chamber, followed by Texas-Red-conjugated streptavidin and fixation (G-I). Panels (D-F) show a single confocal section (D, GFP alone; E, basolateral marker; F, merge of images in D and E). Note colocalization of the GFP-22 construct with the basolateral marker in this lateral plane. (G-I) x-z section, with GFP fluorescence shown in G, apical Con A staining in H and merged image I. Bar, 10 µm.

View larger version (47K): [in a new window]   Fig. 6. N-glycosylation causes an apical shift of a GFP TA construct. Filter-grown MDCK cells stably expressing GFP-22-Nglyc (A), GFP-22-MutNglyc (B) or GFP-22-MutNglyc-myc (C) were analysed by confocal microscopy. x-z sections acquired with the fluorescein filter are shown. GFP-22-MutNglyc and GFP-22-MutNglyc-myc are mainly basolateral, whereas GFP-22-Nglyc shows an apical localisation. Bar, 10 µm.

View larger version (22K): [in a new window]   Fig. 5. A GFP TA construct tagged with a lumenal N-glycosylation consensus sequence acquires an N-linked oligosaccharide that undergoes processing in the Golgi complex. Lysates prepared from cells expressing GFP-22-Nglyc (lanes 1, 2, 4, 5) or from non-transfected MDCK cells (lane 3) were either mock-digested (lanes 1, 3, 4) or digested with PNGase F or Endo H, as indicated, and then analysed by SDS-PAGE followed by western blotting with antiopsin antibodies. The glycosylated forms of the construct are both converted by PNGase F to a more rapidly migrating form, whereas the major, higher Mr band is resistant to Endo H. The vertical arrow in lanes 1 and 4 indicates the Endo-H-sensitive polypeptide, presumably localised in the ER. Numbers on the right indicate the position and size (kDa) of markers (Bio-Rad).

View larger version (60K): [in a new window]   Fig. 7. Quantitative analysis of the surface distribution of tagged GFP TA constructs with the use of extracellular reagents. (A) MDCK cells expressing GFP-22-Nglyc were exposed to anti-opsin mAbs from both the apical and the basolateral compartments of a Transwell filter chamber, followed by Cy5-conjugated anti-mouse antibodies. After fixation and permeabilisation, they cells were labelled with rat anti-E cadherin followed by Cy3-conjugated anti-rat antibodies. Single confocal sections of the same field in an apical (a-d) and lateral (e-h) plane are shown (positions on the z axis correspond to the third and tenth section of panel B), visualised for total GFP (a,e), extracellular opsin tag (b,f) or cadherin (c,g). Acquisition with each filter and image processing was carried out under the same conditions in the apical and lateral plane. Note the absence of intracellular staining with the anti-opsin mAb in panel (f). (d) and (h) show the merged image from panels (a-c) and (e-g) respectively (GFP in green, opsin tag in red and cadherin in blue). The yellow colour in (d) indicates colocalisation between apical GFP and the extracellular opsin tag. The white segments in (h) indicate colocalisation of basolateral GFP, extracellular tag and cadherin. Bar, 10 µm. (B) The distributions of extracellular opsin tag and of cadherin were determined in Z series of confocal sections acquired on GFP-22-Nglyc- (left panel) and GFP-22-MutNglyc- (right panel) expressing cells, treated as in (A). , extracellular opsin tag, , E-cadherin. The average of eight determinations, with s.e.m. (bars), are shown. See Materials and Methods for details. (C) Western blot analysis of surface biotinylated MDCK cells. MDCK cells, expressing GFP-22-Nglyc (lanes 3, 4) or GFP-22-MutNglyc (lanes 5, 6), or non-transfected cells (lanes 1, 2) were biotinylated either from the apical (lanes 1, 3, 5) or the basolateral (lanes 2, 4, 6) compartment of Transwell filter chambers. Biotinylated proteins were collected with streptavidin beads and analysed by western blotting with anti-GFP (upper panel) or anti-cadherin (lower panel) antibodies. The bracket, arrow and arrowhead in the upper panel indicate the mature glycosylated, glycosylated Endo-H-sensitive and unglycosylated forms of GFP-22-Nglyc, respectively. The mature glycosylated form is distributed both on the apical and basolateral surface, whereas the non glycosylated form of GFP-22-Nglyc, as well as GFP-22-MutNgly and cadherin, are detected exclusively on the basolateral side. Numbers on the left indicate the position and size (kDa) of markers (Bio-Rad).

View larger version (45K): [in a new window]   Fig. 8. Analysis of Triton solubility of GFP-22-Nglyc and GFP-22-MutNglyc. (A) Detergent lysates prepared from MDCK cells expressing GFP-22-Nglyc (upper part of panel) or GFP-22-MutNglyc (lower part of panel) were brought to 40% sucrose and analysed by flotation through a sucrose step gradient, as described in Materials and Methods. Half of each fraction was analysed by western blotting, using anti-GFP, anti caveolin I or anti-E cadherin antibodies, as indicated. (B) Half of fraction 4 and 1/20th of fraction 11 (11*) from the same gradient of panel (A) were analysed by western blotting with anti-GFP antibodies. Bracket, arrow and arrowhead in (A) and (B) have the same meaning as in Fig. 7C. Note enrichment of mature glycosylated form of GFP-22-Nglyc in fraction 4, which corresponds to the interface between the 30% and 5% sucrose layers and contains raft-associated caveolin.

View larger version (50K): [in a new window]   Fig. 9. Surface distribution in MDCK cells of GFP reporter constructs bearing the tails of syntaxin 3 or 4. MDCK cells were transiently transfected with GFP-Syn3 (A-C) or GFP-Syn4 (D-F) and then allowed to form polarised monolayers on filters. After fixation, the cells were immunostained for E-cadherin, using Cy3-conjugated secondary antibodies, and analysed by confocal microscopy. A vertical section for each construct is shown: (A and B), GFP fluorescence, (B and E), cadherin, (C and F) merged image of (A + B) and (D + E). GFP-Syn4 has a more pronounced basolateral distribution than GFP-Syn3. Bar, 10 µm.