Munc18 Interacting Proteins
2003; Elsevier BV; Volume: 278; Issue: 38 Linguagem: Inglês
10.1074/jbc.m301632200
ISSN1083-351X
AutoresKaren K. Hill, Yawei Li, Matt Bennett, Melissa McKay, Xinjun Zhu, Jack F. Shern, Enrique Torre, James J. Lah, Allan I. Levey, Richard Kahn,
Tópico(s)Autophagy in Disease and Therapy
ResumoCoat proteins cycle between soluble and membrane-bound locations at the time of vesicle biogenesis and act to regulate the assembly of the vesicle coat that determines the specificity in cargo selection and the destination of the vesicle. A transmembrane cargo protein, an Arf GTPase, and a coat protein (e.g. COPs, APs, or GGAs) are minimal components required for budding of vesicles. Munc18 interacting proteins (MINTs) are a family of three proteins implicated in the localization of receptors to the plasma membrane. We show that MINTs bind Arfs directly, co-localize with Arf and the Alzheimer's precursor protein (β-APP) to regions of the Golgi/trans-Golgi network, and can co-immunoprecipitate clathrin. We demonstrate that MINTs bind Arfs through a region of the PTB domain and the PDZ2 domain, and Arf-MINT interaction is necessary for the increased cellular levels of β-APP produced by MINT overexpression. Knockdown (small interference RNA) experiments implicate β-APP as a transmembrane cargo protein that works together with MINTs. We propose that MINTs are a family of Arf-dependent, vesicle-coat proteins that can regulate the traffic of β-APP. Coat proteins cycle between soluble and membrane-bound locations at the time of vesicle biogenesis and act to regulate the assembly of the vesicle coat that determines the specificity in cargo selection and the destination of the vesicle. A transmembrane cargo protein, an Arf GTPase, and a coat protein (e.g. COPs, APs, or GGAs) are minimal components required for budding of vesicles. Munc18 interacting proteins (MINTs) are a family of three proteins implicated in the localization of receptors to the plasma membrane. We show that MINTs bind Arfs directly, co-localize with Arf and the Alzheimer's precursor protein (β-APP) to regions of the Golgi/trans-Golgi network, and can co-immunoprecipitate clathrin. We demonstrate that MINTs bind Arfs through a region of the PTB domain and the PDZ2 domain, and Arf-MINT interaction is necessary for the increased cellular levels of β-APP produced by MINT overexpression. Knockdown (small interference RNA) experiments implicate β-APP as a transmembrane cargo protein that works together with MINTs. We propose that MINTs are a family of Arf-dependent, vesicle-coat proteins that can regulate the traffic of β-APP. Eukaryotes require specialized, membrane-bounded compartments for many essential functions; e.g. aspects of protein secretion, post-translational protein processing, lipid metabolism, protein degradation, energy metabolism, and the regulation of cell surface protein expression. To achieve the specificity required to establish and maintain these different compartments a system exists in which transmembrane and luminal cargo is recruited into specialized vesicles that can be targeted to specific destinations. These vesicles move in both anterograde and retrograde directions and together comprise membrane traffic. Membrane traffic is orchestrated through the specific recruitment of cargo into budding vesicles at a donor compartment. Delivery to the appropriate destination requires specific proteins to coat the vesicle and provide targeting information and sites of nucleation for the binding of other factors required for maturation of the vesicle. Activation of the ADP-ribosylation factor (Arf) 1The abbreviations used are: Arf, ADP-ribosylation factor; β-APP, β-amyloid precursor protein; PBS, phosphate-buffered saline; siRNA, short-interfering RNA; TGN, trans-Golgi network; EGF, epidermal growth factor; HA, hemagglutinin; GTPγS, guanosine 5′-3-O-(thio)triphosphate; CCV, clathrin-coated vesicles; PNS, post-nuclear supernatant; GST, glutathione S-transferase; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside. GTPase is the critical step in the process of vesicle budding that leads to the recruitment of specific coat proteins, whether monomeric (e.g. GGAs) or protein complex (e.g. heptameric COPI or tetrameric APs). Coat proteins also bind transmembrane cargo proteins (1Rothman J.E. Orci L. Nature. 1992; 355: 409-415Crossref PubMed Scopus (744) Google Scholar, 2Kirchhausen T. Annu. Rev. Cell Dev. Biol. 1999; 15: 705-732Crossref PubMed Scopus (422) Google Scholar, 3Robinson M.S. Bonifacino J.S. Curr. Opin. Cell Biol. 2001; 13: 444-453Crossref PubMed Scopus (442) Google Scholar). The more recent conclusions that GGA1–3, AP-3, and AP-4 join AP-1 and COPI as Arf-dependent coat proteins has revealed a greater complexity in the number and composition of vesicles than was previously appreciated (4Boman A.L. Zhang C.J. Zhu X. Kahn R.A. Mol. Biol. Cell. 2000; 11: 1241-1255Crossref PubMed Scopus (223) Google Scholar, 5Hirst J. Lui W.W. Bright N.A. Totty N. Seaman M.N. Robinson M.S. J. Cell Biol. 2000; 149: 67-80Crossref PubMed Scopus (270) Google Scholar, 6Dell'Angelica E.C. Puertollano R. Mullins C. Aguilar R.C. Vargas J.D. Hartnell L.M. Bonifacino J.S. J. Cell Biol. 2000; 149: 81-94Crossref PubMed Scopus (323) Google Scholar, 7Boehm M. Bonifacino J.S. Mol. Biol. Cell. 2001; 12: 2907-2920Crossref PubMed Scopus (361) Google Scholar, 8Faundez V. Horng J.T. Kelly R.B. Cell. 1998; 93: 423-432Abstract Full Text Full Text PDF PubMed Scopus (247) Google Scholar, 9Hirst J. Bright N.A. Rous B. Robinson M.S. Mol. Biol. Cell. 1999; 10: 2787-2802Crossref PubMed Scopus (219) Google Scholar, 10Simpson F. Peden A.A. Christopoulou L. Robinson M.S. J. Cell Biol. 1997; 137: 835-845Crossref PubMed Scopus (308) Google Scholar, 11Stepp J.D. Huang K. Lemmon S.K. J. Cell Biol. 1997; 139: 1761-1774Crossref PubMed Scopus (177) Google Scholar). As the numbers of types of vesicles has increased, models have had to change from a bulk flow mechanism to ones that include specificity in selection of cargo and coats. With the increased number of coat proteins, and the possibility of them acting combinatorially, there are increasing opportunities for specificity in selection of cargo transported from the Golgi/TGN. There are four characteristics that all Arf-dependent coats share: direct binding to Arf·GTP, recruitment to the Golgi membrane in a brefeldin A-sensitive manner (which further indicates Arf involvement), binding of cargo proteins, and interactions with accessory proteins, e.g. clathrin (4Boman A.L. Zhang C.J. Zhu X. Kahn R.A. Mol. Biol. Cell. 2000; 11: 1241-1255Crossref PubMed Scopus (223) Google Scholar, 7Boehm M. Bonifacino J.S. Mol. Biol. Cell. 2001; 12: 2907-2920Crossref PubMed Scopus (361) Google Scholar, 12Boehm M. Aguilar R.C. Bonifacino J.S. EMBO J. 2001; 20: 6265-6276Crossref PubMed Scopus (79) Google Scholar, 13Donaldson J.G. Klausner R.D. Curr. Opin. Cell Biol. 1994; 6: 527-532Crossref PubMed Scopus (232) Google Scholar, 14Donaldson J.G. Cassel D. Kahn R.A. Klausner R.D. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6408-6412Crossref PubMed Scopus (380) Google Scholar, 15Stamnes M.A. Rothman J.E. Cell. 1993; 73: 999-1005Abstract Full Text PDF PubMed Scopus (339) Google Scholar). To test whether additional coat proteins exist that act at the Golgi, we focused on the direct binding to Arfs as the common feature of such proteins. We report here the identification of the three MINT family proteins as functional homologs to GGAs and AP complexes as coat proteins that regulate aspects of vesicle traffic. MINTs are a family of three proteins. MINT1 and MINT2 are expressed only in neuronal tissues (16Okamoto M. Sudhof T.C. J. Biol. Chem. 1997; 272: 31459-31464Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar), whereas MINT3 is ubiquitously expressed (17Okamoto M. Sudhof T.C. Eur. J. Cell Biol. 1998; 77: 161-165Crossref PubMed Scopus (68) Google Scholar). All MINTs share a conserved central PTB domain and two C-terminal PDZ domains (16Okamoto M. Sudhof T.C. J. Biol. Chem. 1997; 272: 31459-31464Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar, 17Okamoto M. Sudhof T.C. Eur. J. Cell Biol. 1998; 77: 161-165Crossref PubMed Scopus (68) Google Scholar, 18McLoughlin D.M. Miller C.C. FEBS Lett. 1996; 397: 197-200Crossref PubMed Scopus (134) Google Scholar, 19Borg J.P. Ooi J. Levy E. Margolis B. Mol. Cell. Biol. 1996; 16: 6229-6241Crossref PubMed Scopus (436) Google Scholar). Although MINT1 was originally described and named for its ability to bind Munc18, a neuronal protein acting at the synapse, it also binds a number of other proteins, most notably the β-Alzheimer's protein (β-APP), neurexins, and a number of transmembrane receptors (16Okamoto M. Sudhof T.C. J. Biol. Chem. 1997; 272: 31459-31464Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar, 18McLoughlin D.M. Miller C.C. FEBS Lett. 1996; 397: 197-200Crossref PubMed Scopus (134) Google Scholar, 19Borg J.P. Ooi J. Levy E. Margolis B. Mol. Cell. Biol. 1996; 16: 6229-6241Crossref PubMed Scopus (436) Google Scholar, 20Borg J.P. Lopez-Figueroa M.O. de Taddeo-Borg M. Kroon D.E. Turner R.S. Watson S.J. Margolis B. J. Neurosci. 1999; 19: 1307-1316Crossref PubMed Google Scholar, 21Biederer T. Sudhof T.C. J. Biol. Chem. 2000; 275: 39803-39806Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar, 22Lau K.F. McLoughlin D.M. Standen C. Miller C.C. Mol. Cell Neurosci. 2000; 16: 557-565Crossref PubMed Scopus (81) Google Scholar, 23Lee D.-S. Tomita S. Kirino Y. Suzuki T. J. Biol. Chem. 2000; 275: 23134-23138Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 24Riento K. Kauppi M. Keranen S. Olkkonen V.M. J. Biol. Chem. 2000; 275: 13476-13483Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar, 25Setou M. Nakagawa T. Seog D.H. Hirokawa N. Science. 2000; 288: 1796-1802Crossref PubMed Scopus (607) Google Scholar, 26Tomita S. Fujita T. Kirino Y. Suzuki T. J. Biol. Chem. 2000; 275: 13056-13060Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar, 27McLoughlin D.M. Standen C.L. Lau K.F. Ackerley S. Bartnikas T.P. Gitlin J.D. Miller C.C. J. Biol. Chem. 2001; 276: 9303-9307Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 28Maximov A. Sudhof T.C. Bezprozvanny I. J. Biol. Chem. 1999; 274: 24453-24456Abstract Full Text Full Text PDF PubMed Scopus (260) Google Scholar). Mammalian MINT1 and its ortholog in Caenorhabditis elegans, LIN-10, are also found in a stable complex with Cask/LIN-2 and Velis/LIN-7 (20Borg J.P. Lopez-Figueroa M.O. de Taddeo-Borg M. Kroon D.E. Turner R.S. Watson S.J. Margolis B. J. Neurosci. 1999; 19: 1307-1316Crossref PubMed Google Scholar, 29Butz S. Okamoto M. Sudhof T.C. Cell. 1998; 94: 773-782Abstract Full Text Full Text PDF PubMed Scopus (468) Google Scholar, 30Borg J.P. Straight S.W. Kaech S.M. de Taddeo-Borg M. Kroon D.E. Karnak D. Turner R.S. Kim S.K. Margolis B. J. Biol. Chem. 1998; 273: 31633-31636Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar). MINT2 lacks the N-terminal Cask binding domain but binds XB51 and NFκB and can also influence β-APP processing (23Lee D.-S. Tomita S. Kirino Y. Suzuki T. J. Biol. Chem. 2000; 275: 23134-23138Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 26Tomita S. Fujita T. Kirino Y. Suzuki T. J. Biol. Chem. 2000; 275: 13056-13060Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar), whereas MINT3 lacks both the Munc18- and the Cask binding domains and has been less well studied. Except for Munc18 and Cask, binding to other proteins occurs through the PTB and/or PDZ domains of MINTs. The best-studied activities of MINT proteins include roles in traffic and/or processing of β-APP and the EGF receptor. MINT family proteins coordinately increase cellular β-APP levels and half-life and alter its processing (31Sastre M. Turner R.S. Levy E. J. Biol. Chem. 1998; 273: 22351-22357Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 32Biederer T. Cao X. Sudhof T.C. Liu X. J. Neurosci. 2002; 22: 7340-7351Crossref PubMed Google Scholar, 33Borg J.P. Yang Y. De Taddeo-Borg M. Margolis B. Turner R.S. J. Biol. Chem. 1998; 273: 14761-14766Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar) when overexpressed in mammalian cells, and MINT1/LIN-10 is required for the proper localization of the EGF receptor (LIN-23) to the basolateral surface of the vulva progenitor cell in C. elegans (34Whitfield C.W. Benard C. Barnes T. Hekimi S. Kim S.K. Mol. Biol. Cell. 1999; 10: 2087-2100Crossref PubMed Scopus (91) Google Scholar, 35Kaech S.M. Whitfield C.W. Kim S.K. Cell. 1998; 94: 761-771Abstract Full Text Full Text PDF PubMed Scopus (316) Google Scholar). The functional importance of monomeric MINTs, heterodimeric MINT-Munc18, or the heterotrimeric MINT·Cask·Velis complex (and the possibility of interplay between these three states of MINTs) to neuronal or more general eukaryotic cells is not known. Using a two-hybrid strategy, we identified MINTs as binding partners of activated Arf3. Additional data confirmed that MINTs bind directly and preferentially to activated Arfs and that this interaction occurs through their PTB and PDZ2 domains. Treatment with brefeldin A, a specific inhibitor of Arf exchange factors, rapidly reversed MINT localization to Golgi membranes. Knockdown in the expression of either MINT3 or β-APP results in commensurate changes in the distribution of the other protein, extending the functional linkage of Arf to MINT to β-APP. Thus, these data reveal that MINTs share all the characteristics of Arf-dependent coat proteins and β-APP is implicated as a transmembrane cargo for MINT vesicles. The implications of novel cellular roles for MINTs and β-APP in vesicle traffic to pathological conditions leading to Alzheimer's disease are discussed. Antibodies—The β-APP antibodies used in this study were mouse monoclonals 26D6 (Sibia) and 22C11 (Chemicon), raised against sequences in the Aβ and N-terminal portions, respectively. Mouse monoclonal antibodies raised against MINT1 (#M75920, BD Transduction Laboratories), MINT2 (#M76120, BD Transduction), MINT3 (#M93620, BD Transduction), and HA epitope (12CA5, BAbCO), or rabbit polyclonal antisera directed against giantin (BAbCO), γ-adaptin (M-300, Santa Cruz Biotechnologies), clathrin heavy chain (H-300, Santa Cruz Biotechnologies), or rabbit IgG (1–5006, Sigma Chemical Co.) were each obtained from commercial sources. The β-COP rabbit polyclonal antibody was raised against "EAGE peptide" as previously described (36Scheel J. Pepperkok R. Lowe M. Griffiths G. Kreis T.E. J. Cell Biol. 1997; 137: 319-333Crossref PubMed Scopus (72) Google Scholar). Co-immunoprecipitation—COS-7 cells were transiently transfected using FuGENE 6 (Roche Applied Science), according to the manufacturer's specifications. Twenty-four hours after transfection, cells were collected and lysed with lysis buffer (20 mm Tris, pH 7.4, 100 mm NaCl, 1 mm MgCl2, 1 mm CaCl2, 10% glycerol, 1% Nonidet P-40) that also contained a protease inhibitor mixture (Sigma). Nuclei and insoluble material were removed from the cell lysate by centrifugation for 5 min at 14,000 × g. The supernatants were collected and incubated in fresh tubes with antibody for 4 h at 4 °C, before the addition of protein-G-Sepharose beads and a further 1-h incubation. The beads were then collected by centrifugation in a microcentrifuge and washed three times with phosphate-buffered saline (PBS) containing 1% Nonidet P-40. SDS sample buffer (37Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207159) Google Scholar) was added to the beads, and the samples were boiled for 5 min before loading onto polyacrylamide gels. Indirect Immunofluorescent Cell Staining—Cells were grown on coverslips and then fixed with 3.7% formaldehyde for 20 min at room temp before being permeabilized for 10 min with 0.2% saponin in PBS containing 10% goat serum. After permeabilization, the cells were incubated with PBS containing 0.2% saponin, 10% goat serum, and primary antibodies for 1 h at room temperature. The coverslips were washed three times with PBS containing 10% goat serum before a second incubation with Alexa 488 anti-rabbit IgG (Molecular Probes) and Alexa 594 anti-mouse IgG (Molecular Probes) in PBS/0.2% saponin/10% goat serum for 1 h. The coverslips were washed three times with PBS/10% goat serum, washed twice with PBS, and mounted in PBS/0.1 m N-propyl gallate/50% glycerol. Arf-effector Binding Assay—Direct binding of an effector to Arf often results in a change in the steady-state binding of GTP to Arf (38Zhu X. Boman A.L. Kuai J. Cieplak W. Kahn R.A. J. Biol. Chem. 2000; 275: 13465-13475Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar), and therefore radioligand binding can be used as an assay for effector binding (39Zhu X. Kahn R.A. J. Biol. Chem. 2001; 276: 25014-25021Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar, 40Van Valkenburgh H. Shern J.F. Sharer J.D. Zhu X. Kahn R.A. J. Biol. Chem. 2001; 276: 22826-22837Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar). Binding of [γ-35S]GTPγS to Arf was determined as previously described (38Zhu X. Boman A.L. Kuai J. Cieplak W. Kahn R.A. J. Biol. Chem. 2000; 275: 13465-13475Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar, 41Kahn R.A. Gilman A.G. J. Biol. Chem. 1986; 261: 7906-7911Abstract Full Text PDF PubMed Google Scholar, 42Kahn R.A. Methods Enzymol. 1991; 195: 233-242Crossref PubMed Scopus (19) Google Scholar). Arf3 (1 μm) was incubated with GTPγS (10 μm,[γ-35S]GTPγS, 2500 cpm/pmol) and GST-MINT2 or POR1 (5 μm) at 30 °C in binding buffer (20 mm HEPES, pH 7.4, 1 mm EDTA, 1 mm dithiothreitol, 100 mm NaCl, 0.5 mm MgCl2, 3 mm l-α-dimyristoyl phosphatidylcholine, 0.1% sodium cholate, and 100 μg/ml bovine serum albumin). Aliquots (10 μl) were diluted into 2 ml of ice-cold buffer (25 mm Tris-Cl, pH 7.4, 100 mm NaCl, 10 mm MgCl2, and 1 mm dithiothreitol). Samples were rapidly filtered onto 25-mm BA85 nitrocellulose filters, and retained radionucleotide was determined by liquid scintillation counting. Small Interference RNA—Knockdown in the expression of either human β-APP or MINT3 was achieved by transfection of cells with pSUPER-based plasmids containing inserts that generate snap-back RNAs 19 nucleotides long with a loop of 9 nucleotides, as described in Brummelkamp et al. (43Brummelkamp T.R. Bernards R. Agami R. Science. 2002; 296: 550-553Crossref PubMed Scopus (3966) Google Scholar). Sequences targeted were in the middle of the open reading frames in β-APP (5′-GAAGGCAGTTATCCAGCAT-3′) or MINT3 (5′-GATGCTCTGCCACGTATTC-3′). Each sequence is 100% identical to regions of the open reading frame of the targeted messages, and BLAST searches did not find those sequences in any other message. HeLa cells were transfected using LipofectAMINE reagents (Invitrogen), according to the manufacturer's instructions. After a series of time course experiments, 4 days was chosen as the optimal time for knockdown in protein expressions. Cells were then trypsinized, washed in PBS, and lysed in buffer containing 25 mm HEPES, pH 7.4, 100 mm NaCl, and 0.5% Triton X-100. Protein concentrations of whole cell lysates were determined using the Bio-Rad protein assay. Parallel coverslips were fixed and prepared for immunofluorescence studies, as described above. Brefeldin A Treatment—Cells were treated with 10 μm brefeldin A for 0, 2.5, 5, and 10 min before fixing with 3.7% formaldehyde. The cells were stained as indicated. Yeast Two-hybrid Assay—A human, fetal brain cDNA library was screened with an activating mutant of human Arf3 (Arf3-Q71L, I74S) as bait using a two-hybrid screening protocol as previously described (4Boman A.L. Zhang C.J. Zhu X. Kahn R.A. Mol. Biol. Cell. 2000; 11: 1241-1255Crossref PubMed Scopus (223) Google Scholar). β-Galactosidase and histidine auxotrophy assays were performed. Reverse two-hybrid screening was performed to identify point mutations in MINT2 that result in the loss of binding to Arf3-Q71L, I74S. Random mutations were generated in the MINT2-(500–749) insert by polymerase chain reactions performed under conditions of reduced stringency and gap repair of yeast vectors with PCR products. Yeast two-hybrid assays and selection of colonies made deficient in β-galactosidase activity were performed as previously described (39Zhu X. Kahn R.A. J. Biol. Chem. 2001; 276: 25014-25021Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar, 44Kuai J. Kahn R.A. FEBS Lett. 2000; 487: 252-256Crossref PubMed Scopus (11) Google Scholar). Mutated inserts were sequenced and moved into mammalian expression vectors for further analysis. Sixteen mutants were identified that expressed the MINT2 fragment to comparable levels as starting material. Three different mutations in PDZ2 were identified in this way, and those mutations were then moved into the full-length MINT2 open reading frame in the pcDNA3.1 mammalian expression vector. These point mutant full-length MINT2 inserts were also sub-cloned into pACT2, to confirm their loss of binding to activated Arfs. Membrane/Vesicle Preparations—Clathrin-coated vesicles (CCVs) were prepared from fresh rat liver as described previously (45Campbell C. Squicciarini J. Shia M. Pilch P.F. Fine R.E. Biochemistry. 1984; 23: 4420-4426Crossref PubMed Scopus (95) Google Scholar, 46Zhu Y. Traub L.M. Kornfeld S. Mol. Biol. Cell. 1998; 9: 1323-1337Crossref PubMed Scopus (88) Google Scholar). Briefly, a microsome fraction was prepared from fresh rat liver by differential centrifugation. Crude CCVs were isolated from this fraction by centrifugation in 12.5% Ficoll 400/12.5% sucrose and further purified using discontinuous sucrose gradients. TGN membranes were also prepared from fresh rat liver, using a protocol modified from Seaman et al. (47Seaman M.N.J. Sowerby P.J. Robinson M.S. J. Biol. Chem. 1996; 271: 25446-25451Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar). Briefly, a post-nuclear supernatant (PNS) was purified from rat liver homogenate by centrifugation (1500 × g for 10 min) and loaded onto a step gradient consisting of 15 ml of 6% Ficoll layered over 10 ml of 18% Ficoll layered over 4 ml of 45% Nycodenz, all dissolved in 0.25 m sucrose, 10 mm HEPES, pH 7.4, 1 mm MgCl2. The gradient was spun at 100,000 × g in a Beckman SW28 rotor for 2 h at 4 °C. TGN membrane fractions were collected at the 6–18% interface. MINT2 Truncations—Primers were designed to amplify a series of MINT2 truncations using the polymerase chain reaction. After sequencing, these truncated MINT2 mutants were sub-cloned into pACT2 for use in yeast two-hybrid assays. Image Acquisition—Images were acquired using a Zeiss LSM 510 Axiovert 100M confocal scanning laser fluorescence microscope with ×63 optics. The graphics in the figures were assembled using Adobe Photoshop. Replication of Experiments—Every experiment described herein was repeated at least twice with essentially the same results. Most were repeated more than twice. MINTs Bind Directly to Activated Arf—Previous screens of mammalian cDNA libraries with activating mutants of Arf3 (Arf3-Q71L) have yielded novel effectors, including the GGAs, MKLP1, and Arfaptins (4Boman A.L. Zhang C.J. Zhu X. Kahn R.A. Mol. Biol. Cell. 2000; 11: 1241-1255Crossref PubMed Scopus (223) Google Scholar, 48Kanoh H. Williger B.T. Exton J.H. J. Biol. Chem. 1997; 272: 5421-5429Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar, 49Boman A. Kuai J. Zhu X. Chen J. Kuriyama R. Kahn R. Cell Motil. Cytoskeleton. 1999; 44: 119-132Crossref PubMed Scopus (58) Google Scholar). Reverse two-hybrid screens have identified a number of residues that are critical to the binding of different Arf effectors (39Zhu X. Kahn R.A. J. Biol. Chem. 2001; 276: 25014-25021Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar, 50Kuai J. Boman A. Arnold R. Zhu X. Kahn R. J. Biol. Chem. 2000; 275: 4022-4032Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). One such residue is isoleucine 74 as mutation to serine resulted in the loss of binding of Arf3-Q71L, I74S to all previously established Arf effectors yet this protein retained the ability to alter Golgi morphology when expressed in mammalian cells (50Kuai J. Boman A. Arnold R. Zhu X. Kahn R. J. Biol. Chem. 2000; 275: 4022-4032Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). This suggested the presence of additional Arf binding partners that are capable of altering Golgi when stimulated. A human, fetal brain cDNA library was screened with human Arf3-Q71L, I74S as bait, using the same yeast Gal4 two-hybrid screening protocol that earlier identified the GGAs (4Boman A.L. Zhang C.J. Zhu X. Kahn R.A. Mol. Biol. Cell. 2000; 11: 1241-1255Crossref PubMed Scopus (223) Google Scholar). Three library plasmids were cloned that had the ability to support growth in the absence of histidine and presence of 25 mm 5-aminotriazole and were positive in yeast colony β-galactosidase assays but each only in the presence of the Arf3-Q71L, I74S, and not with an unrelated bait. The inserts from the three positives were sequenced and found to encode different lengths of a single open reading frame, encoding human MINT2 (residues 252–749, 475–749, and 500–749). There are three human MINT proteins (MINT1–3, also known as X11α, -β, and -γ or X11, X11-like, and X11-like2; accession numbers Q02410, Q99767, and O96018, respectively) that all share a conserved central PTB domain and two C-terminal PDZ domains (16Okamoto M. Sudhof T.C. J. Biol. Chem. 1997; 272: 31459-31464Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar, 17Okamoto M. Sudhof T.C. Eur. J. Cell Biol. 1998; 77: 161-165Crossref PubMed Scopus (68) Google Scholar, 19Borg J.P. Ooi J. Levy E. Margolis B. Mol. Cell. Biol. 1996; 16: 6229-6241Crossref PubMed Scopus (436) Google Scholar). The shortest insert identified from the library screen (MINT2-(500–749)) encoded both PDZ domains plus an additional 30 residues at the C terminus of the (∼200 residues) PTB domain. The domain organization of the MINT proteins and some of the N-terminal truncation mutants are depicted in Table I. Although MINT1 and -2 are the most highly related family members and each is expressed only in neurons, MINT3 is ubiquitously expressed and shorter (575 residues), lacking both the Munc18 and Cask binding domains from the N terminus (17Okamoto M. Sudhof T.C. Eur. J. Cell Biol. 1998; 77: 161-165Crossref PubMed Scopus (68) Google Scholar). To test whether MINT1 and MINT3 also bind to activated Arfs, constructs homologous to the shortest MINT2 insert were made and, like MINT2, bound activated Arf3 and Arf4 but not the unactivated Arfs, activated Arl1 (see Table I), or unrelated bait proteins (data not shown). Activated Arfs also bound full-length MINT2 (see Fig. 2) and binding does not require the second mutation, Arf3-I74S. Thus, the interactions between Arfs and MINTs are promoted by GTP binding and are different from those previously described in that mutation of isoleucine 74 does not interfere with binding. Representatives of both class I (Arf1–3) and class II (Arf4 and -5) Arfs bind MINTs, suggesting that it is likely a feature of all the soluble Arf (Arf1–5).Table IActivated Arfs bind to MINTs in yeast two-hybrid assays Yeast strains carrying plasmids directing the expression of the indicated Arf protein, fused at their C-termini to the Gal4 binding domain, and full-length MINT2 or N-terminally truncated MINT1–3, fused at their N termini to the Gal4 activation domain, were assayed for β-galactosidase activity, as described under "Materials and Methods." Reactivity (+) was detected within 60 min of exposure of lysed yeast cells to X-gal, and lack of reactivity (-) indicates no blue color detection within 4 h. The domain organization of each MINT and residues involved in each defined domain are depicted above. Mint2-(500–749) was the shortest MINT2 clone identified from library screening, and N-terminal truncation mutants of MINT1 and MINT3 were designed after alignment of the three proteins. MKLP1 was included as a positive control for activated Arfs and Arl1 and negative control for Arf3-Q71L, I74S. Abbreviations include MID, Munc18-interacting domain; CID, Cask-interacting domain; PTB, phosphotyrosine binding domain; PDZ, domain found in PSD-95, DLG, and ZO-1 proteins; MKLP1, mitotic kinesin-like protein 1.Fig. 2MINTs bind Arfs through PDZ2 and PTB domains. MINT2 truncation mutants were assayed in yeast two-hybrid assays with Arf3-Q71L as bait. The different truncation mutants are showed pictorially, with terminating residues in comparison to full-length proteins and protein interaction domains. The three clones pulled from library screening are shown, with MINT2-(500–749) the shortest fragment that yielded maximal activities in both β-galactosidase and histidine auxotrophy assays. Truncations at either end resulted in decreased activity in two-hybrid assays, shown on the right and performed as described under "Materials and Methods." -, no activity; +++, strong blue color developed in X-gal assays within 30 min; ++, strong blue developed within 2 h; +, blue color developed within 4 h. Negative controls (not shown) remained lacking in blue product development after overnight incubations.View Large Image Figure ViewerDownload (PPT) MINTs Bind Arfs Directly—Preference for binding activated Arfs over wild type proteins in two-hybrid assays is a strong indication that the binding is GTP-dependent (4Boman A.L. Zhang C.J. Zhu X. Kahn R.A. Mol. Biol. Cell. 2000; 11: 1241-1255Crossref PubMed Scopus (223) Google Scholar, 39Zhu X. Kahn R.A. J. Biol. Chem. 2001; 276: 25014-25021Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar, 49Boman A. Kuai J. Zhu X. Chen J. Kuriyama R. Kahn R. Cell Motil. Cytoskeleton. 1999; 44: 119-132Crossref PubMed Scopus (58) Google Scholar, 50Kuai J. Boman A. Arnold R. Zhu X. Kahn R. J. Biol. Chem. 2000; 275: 4022-4032Abstract Full Text Full T
Referência(s)