Catch the μ1B Train to the Basolateral Surface
1999; Cell Press; Volume: 99; Issue: 2 Linguagem: Inglês
10.1016/s0092-8674(00)81643-8
ISSN1097-4172
AutoresKeith E. Mostov, Martin B.A. ter Beest, Steven J. Chapin,
Tópico(s)Microtubule and mitosis dynamics
ResumoTransport between the numerous organelles of the cell is largely mediated by sorting of cargo proteins into carrier vesicles, vesicular budding, tethering of the vesicles at the target membrane, and finally fusion. Delivery of proteins to the distinct apical and basolateral plasma membrane domains of polarized epithelial cells is an attractive system to study how specificity of transport is achieved. Proteins involved in transport to these surfaces are summarized in Figure 1. For instance, the exocyst is involved in tethering vesicles at the basolateral surface (8Grindstaff K. Yeaman C. Anandasabapathy N. Hsu S.-C. Rodriguez-Boulan E. Scheller R.H. Nelson W.J Cell. 1998; 93: 731-740Abstract Full Text Full Text PDF PubMed Scopus (449) Google Scholar). SNARES may supply part of the specificity for fusion since epithelial cells utilize distinct t-SNAREs for apical and basolateral traffic (a situation dramatically different from neurons, where no known syntaxins are found uniquely in axons or dendrites) (10Low S.-H. Chapin S.J. Wimmer C. Whiteheart S.W. Kómüves L.G. Mostov K.E. Weimbs T J. Cell Biol. 1998; 141: 1503-1513Crossref PubMed Scopus (158) Google Scholar, 5Foletti D.L. Prekeris R. Scheller R.H Neuron. 1999; 23: 641-644Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 9Lafont F. Verkade P. Galli T. Wimmer C. Louvard D. Simons K Proc. Natl. Acad. Sci. USA. 1999; 96: 3734-3738Crossref PubMed Scopus (208) Google Scholar). Many vesicles traveling between the plasma membrane, trans-Golgi network (TGN), and endosomes utilize clathrin coats. Between the clathrin coat and the lipid bilayer is a layer of clathrin adaptors. These multisubunit adaptor complexes may facilitate selective incorporation of cargo into the nascent vesicle and recruitment of clathrin to the membrane. In this issue of Cell, 6Fölsch H. Ohno H. Bonifacino J.S. Mellman I Cell. 1999; 99: 189-198Abstract Full Text Full Text PDF PubMed Scopus (442) Google Scholar have used a cell line lacking an alternate subunit of a clathrin adaptor, μ1B, to demonstrate its role in sorting of proteins to the basolateral surface of epithelial cells. In the TGN newly made proteins are segregated into vesicles that ultimately carry them to either the basolateral or apical surface. A decade ago it was believed that sorting to the basolateral surface was by default, i.e., no signal was needed. This was disproved by showing that mutations in the cytoplasmic domains of many basolateral proteins resulted in their misdirection to the apical surface (reviewed in 11Matter K. Mellman I Curr. Opin. Cell Biol. 1994; 6: 545-554Crossref PubMed Scopus (391) Google Scholar). More importantly, short sequences in the cytoplasmic domains of the polymeric immunoglobulin receptor (pIgR) and LDL receptor (LDLR) were sufficient to redirect heterologous proteins to the basolateral surface (3Casanova J.E. Apodaca G. Mostov K.E Cell. 1991; 66: 65-75Abstract Full Text PDF PubMed Scopus (226) Google Scholar, 12Matter K. Hunziker W. Mellman I Cell. 1992; 71: 741-753Abstract Full Text PDF PubMed Scopus (304) Google Scholar). Some basolateral sorting signals resemble the endocytic signals used for incorporation of membrane proteins into clathrin-coated pits at the plasma membrane, suggesting that basolateral sorting uses a similar clathrin-based mechanism. The molecular basis for recognition of endocytic signals is partly understood. The AP2 class of clathrin adaptors, containing α, β2, μ2, and σ2 subunits, are used for endocytosis. One type of endocytotic signal is based on a Y-X-X-hydrophobic motif, which binds to the μ2 subunit (reviewed in 2Bonifacino J.S. Dell'Angelica E.C J. Cell Biol. 1999; 145: 923-926Crossref PubMed Scopus (369) Google Scholar). AP1 and AP3 adaptors mediate delivery from the TGN to endosomes. AP1 consists of γ, β1, μ1, and σ1 subunits, and the μ1 subunit recognizes Tyr-based sorting signals. Fölsch et al. studied a novel isoform of μ1, μ1B, whose expression is confined to epithelial cells (13Ohno H. Tomemori T. Nakatsu F. Okazaki Y. Aguilar R.C. Foelsch H. Mellman I. Saito T. Shirasawa T. Bonifacino J.S FEBS Lett. 1999; 449: 215-220Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar). This suggested that μ1B, as part of the AP1 complex, could be the long sought adaptor responsible for basolateral sorting. One polarized epithelial cell line, LLC-PK1, lacks μ1B. These cells are not very well polarized, and missort several basolateral proteins, e.g., LDLR and transferrin receptor (TfR), to the apical surface (6Fölsch H. Ohno H. Bonifacino J.S. Mellman I Cell. 1999; 99: 189-198Abstract Full Text Full Text PDF PubMed Scopus (442) Google Scholar, 16Roush D.L. Gottardi C.J. Naim H.Y. Roth M.G. Caplan M.J J. Biol. Chem. 1998; 273: 26862-26869Crossref PubMed Scopus (106) Google Scholar). Fölsch et al. exogenously expressed μ1B in LLC-PK1 cells and showed that LDLR and TfR are redirected to the basolateral surface. The key conclusion is that the μ1B subunit is sufficient to sort a protein to the basolateral surface. How, precisely, this is achieved remains to be determined, but it will be interesting to see if μ1B induces the formation of a new class of basolaterally directed vesicles, or causes the μ1B-dependent sorting of certain proteins into vesicles destined for the basolateral surface even in the absence of μ1B. Fölsch et al. showed that μ1B assembles with the γ, β1, and σ1 subunits, though it is unclear if individual vesicles homogeneously contain either μ1B or the original μ1 isoform (renamed μ1A), or if both isoforms are present on a single vesicle. Whether μ1B is strictly required for basolateral sorting is not yet known, but the missorting phenotype of LLC-PK1 cells lacking μ1B strongly suggests that this might be the case. μ1B can recognize Tyr-based basolateral signals (13Ohno H. Tomemori T. Nakatsu F. Okazaki Y. Aguilar R.C. Foelsch H. Mellman I. Saito T. Shirasawa T. Bonifacino J.S FEBS Lett. 1999; 449: 215-220Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar). However, there are multiple types of basolateral sorting signals and only a subset of proteins rely on μ1B for their sorting. A different type of basolateral signal, an LL sequence, interacts with the β rather than the μ adaptor subunit, and the IgG Fc receptor FcRII-B2 (which contains an LL signal) is basolateral even in LLC-PK1 cells. Yet another type of basolateral signal, H/R-X-X-V, is found in the pIgR and the mannose-6-phosphate receptor (1Aroeti B. Okhrimenko H. Reich V. Orzech E Biochim. Biophys. Acta. 1998; 1376: 57-90Crossref PubMed Scopus (51) Google Scholar, 4Distel B. Bauer U. Le Borgne R. Hoflack B J. Biol. Chem. 1998; 273: 186-193Crossref PubMed Scopus (25) Google Scholar). A related signal may sort the metabotrophic glutamate receptor to the soma and dendrites of neurons, a process similar to basolateral sorting in epithelial cells (17Stowell J.N. Craig A.M Neuron. 1999; 22: 525-536Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar). Even the same protein in the same cell can use multiple mechanisms for delivery to the basolateral surface. For instance, the pIgR interacts with AP1 (μ1 isotype unknown) during TGN to basolateral transport, but if phosphorylation of Ser726 in its cytoplasmic tail is prevented, the pIgR takes a different pathway (not involving AP1) to the basolateral surface (14Orzech E. Schlessinger K. Weiss A. Okamoto C.T. Aroeti B J. Biol. Chem. 1999; 274: 2201-2215Crossref PubMed Scopus (40) Google Scholar). Polarized sorting occurs in both the biosynthetic and endocytotic pathways, as endocytosed proteins are recycled to their original surface or transcytosed to the opposite surface. An unresolved question is whether μ1B works in either the biosynthetic or endocytotic pathways, or both? Fölsch et al. determined the steady-state distribution of LDLR and TfR in LLC-PK1 cells transfected with μ1B. Further work will now be necessary to address whether these proteins were directly targeted from the TGN to the basolateral surface, or targeted to both surfaces, then endocytosed and sorted basolaterally. A clue to the site of action of μ1B comes from previous data that for both LDLR and pIgR, mutations in the basolateral signal that affect polarity of biosynthetic sorting have similar effects on polarized sorting after endocytosis (reviewed in 1Aroeti B. Okhrimenko H. Reich V. Orzech E Biochim. Biophys. Acta. 1998; 1376: 57-90Crossref PubMed Scopus (51) Google Scholar). This suggests that polarized sorting might not occur in the TGN; instead proteins on their way to the cell surface may traverse endosomes where polarized sorting could occur. In nonpolarized fibroblasts a portion of newly made plasma membrane proteins move through endosomes. Similar experiments in polarized cells are thus needed to test this possibility. Alternatively, the same sorting machinery may act in both the TGN and endocytic pathways. In polarized MDCK cells, AP1 (μ1 isotype unknown) and clathrin have been found on TfR-containing buds on tubular endosomes, which may generate vesicles that recycle back to the basolateral surface (7Futter C.E. Gibson A. Allchin E.H. Maxwell S. Ruddock L.J. Odorizzi G. Domingo D. Trowbridge I.S. Hopkins C.R J. Cell Biol. 1998; 141: 611-623Crossref PubMed Scopus (193) Google Scholar). This process might use μ1B, and its absence in LLC-PK1 cells would thereby abrogate polarized recycling. That μ1B may be used for multiple sorting pathways and that there exist μ1B-independent mechanisms for basolateral transport, highlights the complexity of the sorting machinery. Indeed, it seems that the only "law" of membrane traffic is that there is always another pathway. What might be the physiologal significance of this complexity? The lack of μ1B in LLC-PK1 cells may serve to deliver endocytotic receptors to the apical surface of the kidney proximal tubule cell (the origin of LLC-PK1 cells), where they could function to retrieve proteins from the urinary filtrate. Similarly, during development of epithelial tubules, cells undergo a transient loss of polarity, but then reestablish apical and basolateral surfaces (15Pollack A.L. Runyan R.B. Mostov K.E Dev. Biol. 1998; 204: 64-79Crossref PubMed Scopus (195) Google Scholar). These are but two examples of how modulating polarized trafficking pathways can provide the flexibility to generate specialized cell types and enable the cell to adapt to changing developmental requirements.*To whom correspondence should be addressed (e-mail: [email protected]).
Referência(s)