H-Ras Dynamically Interacts with Recycling Endosomes in CHO-K1 Cells

H-, N-, and K-Ras are isoforms of Ras proteins, which undergo different lipid modifications at the C terminus. These post-translational events make possible the association of Ras proteins both with the inner plasma membrane and to the cytosolic surface of endoplasmic reticulum and Golgi complex, which is also required for the proper function of these proteins. To better characterize the intracellular distribution and sorting of Ras proteins, constructs were engineered to express the C-terminal domain of H- and K-Ras fused to variants of green fluorescent protein. Using confocal microscopy, we found in CHO-K1 cells that H-Ras, which is palmitoylated and farnesylated, localized at the recycling endosome in addition to the inner leaflet of the plasma membrane. In contrast, K-Ras, which is farnesylated and nonpalmitoylated, mainly localized at the plasma membrane. Moreover, we demonstrate that sorting signals of H- and K-Ras are contained within the C-terminal domain of these proteins and that palmitoylation on this region of H-Ras might operate as a dominant sorting signal for proper subcellular localization of this protein in CHO-K1 cells. Using selective photobleaching techniques, we demonstrate the dynamic nature of H-Ras trafficking to the recycling endosome from plasma membrane. We also provide evidence that Rab5 and Rab11 activities are required for proper delivery of H-Ras to the endocytic recycling compartment. Using a chimera containing the Ras binding domain of c-Raf-1 fused to a fluorescent protein, we found that a pool of GTP-bound H-Ras localized on membranes from Rab11-positive recycling endosome after serum stimulation. These results suggest that H-Ras present in membranes of the recycling endosome might be activating signal cascades essential for the dynamic and function of the organelle. H-, N-, and K-Ras are isoforms of Ras proteins, which undergo different lipid modifications at the C terminus. These post-translational events make possible the association of Ras proteins both with the inner plasma membrane and to the cytosolic surface of endoplasmic reticulum and Golgi complex, which is also required for the proper function of these proteins. To better characterize the intracellular distribution and sorting of Ras proteins, constructs were engineered to express the C-terminal domain of H- and K-Ras fused to variants of green fluorescent protein. Using confocal microscopy, we found in CHO-K1 cells that H-Ras, which is palmitoylated and farnesylated, localized at the recycling endosome in addition to the inner leaflet of the plasma membrane. In contrast, K-Ras, which is farnesylated and nonpalmitoylated, mainly localized at the plasma membrane. Moreover, we demonstrate that sorting signals of H- and K-Ras are contained within the C-terminal domain of these proteins and that palmitoylation on this region of H-Ras might operate as a dominant sorting signal for proper subcellular localization of this protein in CHO-K1 cells. Using selective photobleaching techniques, we demonstrate the dynamic nature of H-Ras trafficking to the recycling endosome from plasma membrane. We also provide evidence that Rab5 and Rab11 activities are required for proper delivery of H-Ras to the endocytic recycling compartment. Using a chimera containing the Ras binding domain of c-Raf-1 fused to a fluorescent protein, we found that a pool of GTP-bound H-Ras localized on membranes from Rab11-positive recycling endosome after serum stimulation. These results suggest that H-Ras present in membranes of the recycling endosome might be activating signal cascades essential for the dynamic and function of the organelle. Ras proteins are signal transducers that regulate cellular growth, differentiation, and apoptosis in response to different stimulus (1Campbell S.L. Khosravi-Far R. Rossman K.L. Clark G.J. Der C.J. Oncogene. 1998; 17: 1395-1413Crossref PubMed Scopus (920) Google Scholar, 2Shields J.M. Pruitt K. McFall A. Shaub A. Der C.J. Trends Cell. Biol. 2000; 10: 147-154Abstract Full Text Full Text PDF PubMed Scopus (689) Google Scholar, 3Cox A.D. Der C.J. Oncogene. 2003; 22: 8999-9006Crossref PubMed Scopus (363) Google Scholar). Ras proteins act as binary molecular switches, cycling between inactive GDP-bound and active GTP-bound forms that are regulated by the concerted action of different proteins such as guanine nucleotide exchange factors, which stimulate the GDP release from GDP-loaded Ras allowing the binding of the most abundant GTP and GTPase-activating proteins (GAPs), 3The abbreviations used are: GAP, GTPase-activating protein; Arf, ADP-ribosylation factor; CFP, cyan fluorescent protein; CHO, Chinese hamster ovary; DRM, detergent-resistant membrane(s); FRAP, fluorescence recovery after photobleaching; FLIP, fluorescence loss in photobleaching; GalNAc-T, UDP-GalNAc:LacCer/GM3/GD3 N-acetylgalactosaminyltransferase; GFP, green fluorescent protein; GPI, glycosylphosphatidylinositol; M6PR, mannose 6-phosphate receptor; RBD, Ras binding domain; YFP, yellow fluorescent protein; PBS, phosphate-buffered saline; PIM, protease inhibitor mixture; BSA, bovine serum albumin; HA, hemagglutinin; GM3, NeuAc-α2,3Galβ1,4Glc-ceramide; GD3, NeuAcα2,8NeuAcα2,3Galβ1,4Glc-ceramide. which enhances the intrinsic GTPase activity of Ras (4Quilliam L.A. Rebhun J.F. Castro A.F. Prog. Nucleic Acids Res. Mol. Biol. 2002; 71: 391-444Crossref PubMed Google Scholar, 5Zheng Y. Quilliam L.A. EMBO Rep. 2003; 4: 463-468Crossref PubMed Scopus (16) Google Scholar). There are three ubiquitous isoforms of Ras proteins, namely H-Ras, N-Ras, and K-Ras4B, referred as K-Ras for the rest of this work. These proteins are more than 90% homologues, and their functions are not redundant (6Hancock J.F. Nat. Rev. Mol. Cell. Biol. 2003; 4: 373-384Crossref PubMed Scopus (692) Google Scholar). Knock-out mice of the K-ras gene are not viable, whereas deletions of the N-ras and H-ras genes have no observable side effects (7Umanoff H. Edelmann W. Pellicer A. Kucherlapati R. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 1709-1713Crossref PubMed Scopus (206) Google Scholar, 8Johnson L. Greenbaum D. Cichowski K. Mercer K. Murphy E. Schmitt E. Bronson R.T. Umanoff H. Edelmann W. Kucherlapati R. Jacks T. Genes Dev. 1997; 11: 2468-2481Crossref PubMed Scopus (442) Google Scholar, 9Koera K. Nakamura K. Nakao K. Miyoshi J. Toyoshima K. Hatta T. Otani H. Aiba A. Katsuki M. Oncogene. 1997; 15: 1151-1159Crossref PubMed Scopus (288) Google Scholar, 10Esteban L.M. Vicario-Abejon C. Fernandez-Salguero P. Fernandez-Medarde A. Swaminathan N. Yienger K. Lopez E. Malumbres M. McKay R. Ward J.M. Pellicer A. Santos E. Mol. Cell. Biol. 2001; 21: 1444-1452Crossref PubMed Scopus (250) Google Scholar). In addition, distinct signal outputs from different Ras isoforms (11Yan J. Roy S. Apolloni A. Lane A. Hancock J.F. J. Biol. Chem. 1998; 273: 24052-24056Abstract Full Text Full Text PDF PubMed Scopus (387) Google Scholar) and different potencies of their corresponding activated alleles for oncogenic transformation (12Voice J.K. Klemke R.L. Le A. Jackson J.H. J. Biol. Chem. 1999; 274: 17164-17170Abstract Full Text Full Text PDF PubMed Scopus (230) Google Scholar) have been identified. It is widely assumed that the association of Ras proteins with membranes, particularly with the inner leaflet of the plasma membrane, is necessary and sufficient for the proper function of these proteins (13Hancock J.F. Paterson H. Marshall C.J. Cell. 1990; 63: 133-139Abstract Full Text PDF PubMed Scopus (846) Google Scholar). The association of these proteins with membranes is a consequence of their post-translational lipid modification. The polypeptide sequence of all Ras isoforms contains a C-terminal CAAX motif (where C represents cysteine, A is aliphatic, and X is any other amino acid), which is first modified in the cytosol with a farnesyl anchor to the cysteine residue, and then the AAX sequence is cleaved by an endopeptidase associated to the cytosolic surface of the endoplasmic reticulum and, finally, the farnesylated cysteine is carboxymethylated (6Hancock J.F. Nat. Rev. Mol. Cell. Biol. 2003; 4: 373-384Crossref PubMed Scopus (692) Google Scholar). Depending on the Ras isoform, a second signal of membrane anchor is present immediately to the farnesylated cysteine. H-Ras is dually palmitoylated at cysteine residues 181 and 184, whereas N-Ras is palmitoylated at cysteine residue 184 (6Hancock J.F. Nat. Rev. Mol. Cell. Biol. 2003; 4: 373-384Crossref PubMed Scopus (692) Google Scholar). Conversely, K-Ras is not further lipid-modified but contains a polybasic domain (six contiguous lysine residues) with binding capacity to phospholipid headgroups of lipid bilayers (14Ghomashchi F. Zhang X. Liu L. Gelb M.H. Biochemistry. 1995; 34: 11910-11918Crossref PubMed Scopus (106) Google Scholar, 15Leventis R. Silvius J.R. Biochemistry. 1998; 37: 7640-7648Crossref PubMed Scopus (80) Google Scholar). The CAAX motif and the second signal (palmitoylation or the polybasic domain) are essential for efficient attachment of Ras proteins to membranes, in particular plasma membrane (13Hancock J.F. Paterson H. Marshall C.J. Cell. 1990; 63: 133-139Abstract Full Text PDF PubMed Scopus (846) Google Scholar). As described above, different functions have been attributed to Ras proteins. It has been recognized that Ras isoforms are capable of activating different effectors, thus affecting different signaling pathways (2Shields J.M. Pruitt K. McFall A. Shaub A. Der C.J. Trends Cell. Biol. 2000; 10: 147-154Abstract Full Text Full Text PDF PubMed Scopus (689) Google Scholar, 3Cox A.D. Der C.J. Oncogene. 2003; 22: 8999-9006Crossref PubMed Scopus (363) Google Scholar, 16Reuther G.W. Der C.J. Curr. Opin. Cell Biol. 2000; 12: 157-165Crossref PubMed Scopus (347) Google Scholar). Because transmembrane receptors and Ras proteins are mainly localized in plasma membrane, one plausible explanation for these results is that they are localized in different plasma membrane microdomains (17Parton R.G. Hancock J.F. Trends Cell. Biol. 2004; 14: 141-147Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar), which are formed by the lateral segregation of lipids based on their dissimilar biophysical properties (18Simons K. Toomre D. Nat. Rev. Mol. Cell. Biol. 2000; 1: 31-39Crossref PubMed Scopus (5164) Google Scholar, 19Edidin M. Trends Cell. Biol. 2001; 11: 492-496Abstract Full Text Full Text PDF PubMed Scopus (224) Google Scholar). The differential activities of Ras proteins can be also explained if they are localized in different membrane subcompartments (20Bivona T.G. Philips M.R. Curr. Opin. Cell Biol. 2003; 15: 136-142Crossref PubMed Scopus (122) Google Scholar). In this sense, it has been demonstrated that intracellular pools of H-Ras localized in membranes of endoplasmic reticulum and Golgi apparatus become active when cells are stimulated with epidermal growth factor (21Chiu V.K. Bivona T. Hach A. Sajous J.B. Silletti J. Wiener H. Johnson II, R.L. Cox A.D. Philips M.R. Nat. Cell. Biol. 2002; 4: 343-350Crossref PubMed Scopus (517) Google Scholar, 22Arozarena I. Matallanas D. Berciano M.T. Sanz-Moreno V. Calvo F. Munoz M.T. Egea G. Lafarga M. Crespo P. Mol. Cell. Biol. 2004; 24: 1516-1530Crossref PubMed Scopus (78) Google Scholar). Despite it becoming apparent that subcellular distribution of Ras isoforms is important for proper function, the underlying mechanisms of intracellular transport and distribution of these proteins are poorly understood. K-Ras, after it is synthesized in the cytosol and post-translationally modified in the endoplasmic reticulum, can reach the plasma membrane by a desorption-absortion mechanism. This mechanism is driven by the negative electrostatic potential of the plasma membrane, which is enriched in anionic lipids such as phophatidylserine (inner leaflet) and sialic acid-containing glycolipids and glycoproteins (outer leaflet) (23Silvius J.R. J. Membr. Biol. 2002; 190: 83-92Crossref PubMed Scopus (82) Google Scholar). Conversely, the current data suggest that H-Ras, and probably N-Ras, are transported to different subcompartments by vesicular traffic (24Choy E. Chiu V.K. Silletti J. Feoktistov M. Morimoto T. Michaelson D. Ivanov I.E. Philips M.R. Cell. 1999; 98: 69-80Abstract Full Text Full Text PDF PubMed Scopus (620) Google Scholar, 25Apolloni A. Prior I.A. Lindsay M. Parton R.G. Hancock J.F. Mol. Cell. Biol. 2000; 20: 2475-2487Crossref PubMed Scopus (345) Google Scholar, 26Roy S. Wyse B. Hancock J.F. Mol. Cell. Biol. 2002; 22: 5128-5140Crossref PubMed Scopus (114) Google Scholar) or by a nonvesicular pathway involving a constitutive de-/reacylation cycle (27Goodwin J.S. Drake K.R. Rogers C. Wright L. Lippincott-Schwartz J. Philips M.R. Kenworthy A.K. J. Cell. Biol. 2005; 170: 261-272Crossref PubMed Scopus (227) Google Scholar, 28Rocks O. Peyker A. Kahms M. Verveer P.J. Koerner C. Lumbierres M. Kuhlmann J. Waldmann H. Wittinghofer A. Bastiaens P.I. Science. 2005; 307: 1746-1752Crossref PubMed Scopus (648) Google Scholar). In the present work, we have investigated the membrane association, subcellular distribution, and intracellular trafficking of H- and K-Ras proteins in Chinese hamster ovary (CHO)-K1 cells. By using confocal microscopy and biochemical analysis, we demonstrate that H-Ras, at steady state, localized at the recycling endosome in addition to the cytoplasmic leaflet of the plasma membrane. In contrast, K-Ras mainly localized at the plasma membrane. Interestingly, we found that sorting signals of H- and K-Ras are contained within the C-terminal domain of these proteins and that palmitoylation on this region of H-Ras might operate as a dominant sorting signal for proper subcellular localization of this protein. Using selective photobleaching techniques and time lapse fluorescence microscopy, we demonstrate the dynamic association of H-Ras to the recycling endosome. We also found that Rab5 and Rab11 activities are required for delivery of H-Ras to this organelle. Using a chimera containing the Ras binding domain of c-Raf-1 fused to a fluorescent protein, we demonstrate that H-Ras can be activated (GTP-bound state) at the Rab11-positive recycling endosome after proper stimulation. This evidence supports roles for novel signaling pathways regulated by H-Ras in membranes from recycling endosome. Plasmids—The expression vectors pECFP-N1, pEYFP-N1, pECFP-C1, pEYFP-C1, and pEGFP-F, a plasmid encoding green fluorescent protein (GFP)-H-RasC20, were from Clontech. Expression plasmids for CFP-K-RasC14 and N13GAP43-YFP were generated by synthesizing complementary oligonucleotides and cloned in BglII/EcoRI sites of pECFP-C1 or HindIII/BamHI sites of pEYFP-N1. Plasmids encoding YFP-H-Rasfull and YFP-K-Rasfull were kindly supplied by M. Philips (New York University School of Medicine). The glycosylphosphatidylinositol (GPI)-YFP fusion construct was kindly supplied by P. Keller (Max-Plank Institute, Dresden, Germany). GFP-Rab11 fusion proteins were provided by M. Colombo (Universidad Nacional de Cuyo, Mendoza, Argentina). GFP-Rab5 and GFP-Dyn2K44A fusion proteins were kindly supplied by J. Bonifacino (NICHD, National Institutes of Health, Bethesda, MD). RBD51-131HA-CFP was generated by PCR amplification of the sequence encoding amino acids 51-131 of c-Raf-1 and later insertion in the NheI/SalI sites of pECFP-N1. Sequences of the oligonucleotides used for the different constructs are available on request. Cell Lines, Cell Culture, and DNA Transfections—The following CHO-K1 cell clones were used: wild type CHO-K1 cells (ATCC) and clone 5, a stable cell line that expresses CFP-H-RasC20. Cells were maintained at 37 °C in 5% CO2 in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and antibiotics. Cells were transfected with 0.6-1.2 μg/35-mm dish of expression plasmids using Lipofectamine (Invitrogen) according to the manufacturers' recommendations. 24 h after transfection, cells were labeled with endocytic markers or washed with cold phosphate-buffered saline (PBS), harvested by scrapping, and lysed as indicated below or washed with PBS and fixed for immunofluorescence microscopy. Subcellular Fractionation—Cells were washed with cold PBS and harvested by scraping. Extracts were centrifuged at 10,000 rpm for 5 min and resuspended in 400 μl of PBS in presence of 5 μg/ml aprotinin, 0.5 μg/ml leupeptin, 0.7 μg/ml pepstatin (PBS-protease inhibitor mixture, PBS-PIC). Pellets were dispersed by vortex, each for 10 min, and after 30 min, they were passed 20 times through a 25-gauge needle. After 60 min of treatment, nuclear fractions and unbroken cells were removed by centrifugation at 2,000 rpm, and supernatants were ultracentrifuged for 1 h at 100,000 × g at 4 °C, using a TLA 100.3 rotor (Beckman). The supernatant (S fraction) was removed, and the pellet (P fraction) was resuspended in 400 μl of PBS-PIC for subsequent Western blotting. Proteins from S and P fractions were precipitated with chloroform/methanol (1:4, v/v). The pellets were resuspended in Laemmli buffer (29Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207231) Google Scholar) and subjected to SDS-12% PAGE and Western blotting. Hydroxylamine Treatment—Homogenates from a stable cell line that expresses CFP-H-RasC20 were treated with 1 m hydroxylamine or 1 m Tris (control) for 3 h at room temperature as previously described (30Skene J.H. Virag I. J. Cell. Biol. 1989; 108: 613-624Crossref PubMed Scopus (315) Google Scholar). Samples were resolved by SDS-PAGE, and the expression of CFP-H-RasC20 was analyzed by Western blot as indicated below. The acylation state was analyzed by changes in the electrophoretic mobility of the recombinant proteins. Triton X-114 Partition Assay—S and P fractions were solubilized for 1 h at 4 °C in 1% Triton X-114. Then samples were incubated at 37 °C for 3 min and then centrifuged at 12,000 rpm. The aqueous upper phase and detergent-enriched lower phase were separated and extracted again with detergent and aqueous solutions, respectively. The four resulting samples were adjusted to equal volumes, and detergent content and proteins were precipitated with chloroform/methanol previous to Western blot analyses. Electrophoresis and Western Blotting—Proteins were resolved by electrophoresis through 12% SDS-PAGE gels under reducing conditions and then electrophoretically transferred to nitrocellulose membranes for 80 min at 300 mA. Protein bands in nitrocellulose membranes were visualized by Ponceau S staining. For immunoblotting, nonspecific binding sites on the nitrocellulose membrane were blocked with 5% nonfat dry milk or with 2.5% bovine serum albumin (BSA), 2.5% polyvinyl pyrrolidone 40 in Tris-buffered saline (200 mm NaCl, 50 mm Tris-HCl, pH 7.5). Anti-GFP polyclonal antibody (Roche Applied Science) was used at a dilution of 1:1000. Bands were detected by protein A coupled to horseradish peroxidase combined with the chemiluminescence detection kit (ECL plus Western blotting detection system; Amersham Biosciences) and Hyperfilm MP films (Amersham Biosciences). The relative contribution of individual bands was measured using the computer software Scion Image on scanned films of low exposure images. Endocytosis of Alexa-conjugated Transferrin—To label the endocytic recycling compartment, CHO-K1 cells grown in coverslips were incubated in Dulbecco's modified Eagle's medium, 0.1% BSA for 1 h in a CO2 incubator and then labeled in identical conditions with Alexa594- or Alexa633-human transferrin (14 μg/ml). After the labeling, cells were shifted to 4 °C, washed tree times with PBS plus 0.1% BSA, and finally washed with PBS before paraformaldehyde fixation (30 min at 4 °C). Immunofluorescence Microscopy—Cells grown in coverslips were washed twice with PBS, fixed in 3% paraformaldehyde at 4 °C for 30 min, washed with PBS, permeabilized with 0.1% Triton X-100, 200 mm glycine in PBS (10 min at 4 °C), and incubated in PBS plus 3% BSA for 1 h at 37 °C to block nonspecific binding sites. Coverslips were then incubated overnight at 4 °C with primary antibodies, washed five times with PBS plus 1% BSA, and exposed to secondary antibodies for 90 min at 37 °C. The primary antibodies were rabbit polyclonal anti-calnexin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) diluted 1:200, rabbit polyclonal anti-mannose-6-phosphate receptor (M6PR) diluted 1:150, and rabbit polyclonal anti-mannosidase II (from K. Moremen, University of Georgia, Athens, GA) diluted 1:300. Secondary antibodies were Alexa546-conjugated goat anti-mouse antibodies (Santa Cruz Biotechnology), diluted 1:1000, or rhodamine-conjugated donkey anti-rabbit antibody (Jackson InmunoResearch), diluted 1:700. After final washes with PBS plus 1% BSA, coverslips were mounted in FluorSave (Calbiochem). Confocal Microscopy and Image Acquisition and Processing—Confocal images were collected using a Carl Zeiss LSM5 Pascal laser-scanning confocal microscope. Excitation wavelengths and filter set for GFP, YFP, CFP, and rhodamine were described previously (31Crespo P.M. Zurita A.R. Giraudo C.G. Maccioni H.J. Daniotti J.L. Biochem. J. 2004; 377: 561-568Crossref PubMed Google Scholar). Images from cells coexpressing GFP- and YFP-tagged fluorescent proteins were acquired using a CFP filter set to capture GFP fluorescence and 514-nm laser line (excitation) and a 560-nm long pass filter (emission) to capture YFP fluorescence. Images of fixed cells (no quantitative purposes) were taken using 63 × 1.4 numerical aperture (Plan-Apochromat, Zeiss) or 100 × 1.4 numerical aperture (Plan-Apochromat; Zeiss) objectives and the confocal pinhole of the microscope set to obtain an optical slice of 0.8 μm. Images of living cells in fluorescence recovery after photobleaching (FRAP) and fluorescence loss in photobleaching (FLIP) experiments were acquired with a 20 × 0.5 numerical aperture (Plan Neofluoar, Zeiss) objective and the confocal pinhole of the microscope set fully open. Selective photobleaching of YFP was carried out on a Zeiss LSM510 META laser-scanning confocal microscopy (Instituto Milenio de Estudios Avanzados en Biología Molecular y Biotecnología, Facultad de Ciencias, Santiago, Chile) using 500 consecutive scans with a 514-nm laser line (18 milliwatts). Living cells were held at 37 °C in a 5% CO2 atmosphere during image acquisition. All images were processed for output purposes using Adobe Photoshop® software. Fluorescence quantification in selected regions of interest was carried out using MetaMorph® 4.5 software. All measurements of mean fluorescence intensity were background-corrected by subtracting the mean fluorescence of an area void of cells. When required in FRAP experiments, images were corrected for partial photobleaching during image acquisition. A kinetic model for transport of YFP-H-Ras to recycling endosome was developed to obtain the t1/2 of fluorescence recovery in this organelle. As a first approximation, this model includes two compartments (recycling endosome and plasma membrane) where Ras is located. First order differential equations were set for describing the exchange between these compartments. Protein synthesis and degradation were not included in the model, because both the short time duration of the recovery process and treatment with cycloheximide for 6 h did not alter the ratio of mean fluorescence associated with recycling endosomes and mean fluorescence in the rest of the cell. The exponential part of the recovery process was utilized to fit the model, and non-linear regression analysis was performed using Microcal Origin® software. Biotinylation of Plasma Membrane Proteins—Plasma membrane proteins were labeled using EZ-link N-hydroxysulfosuccinimide-SS-Biotin reagent (Pierce) according to the manufacturer's instructions. Briefly, cells were gently rinsed with Dulbecco's modified Eagle's medium and then incubated with 1.3 mg/ml EZ-link in culture medium at 4 °C for 30 min. Finally, cells were rinsed three times with 0.1% BSA in PBS in order to quench reactive N-hydroxysulfosuccinimide ester groups. Sucrose Density Gradient—Subcellular fractionation using a step flotation gradient was performed using a previously described method with few modifications (32Gorvel J.P. Chavrier P. Zerial M. Gruenberg J. Cell. 1991; 64: 915-925Abstract Full Text PDF PubMed Scopus (860) Google Scholar, 33Trischler M. Stoorvogel W. Ullrich O. J. Cell. Sci. 1999; 112: 4773-4783Crossref PubMed Google Scholar). Cells cultured in a 60-mm Petri dish were rinsed, gently scraped in PBS-PIC, and centrifuged for 5 min at 500 × g at 4 °C. The pellet was then resuspended in 1 ml of homogenization buffer (250 mm sucrose, 3 mm imidazole, pH 7.4) containing protease inhibitors and centrifuged at 1,200 × g for 10 min. The sediment was then resuspended in 200 μl of homogenization buffer containing 0.5 mm EDTA and passed 20 times through a 25-gauge needle. Under these conditions, more than 90% of cells were disrupted while nuclei remained intact as determined by double labeling with both a permeable (Hoescht 33258; Molecular Probes, Inc., Eugene, OR) and impermeable (Sytox green; Molecular Probes) DNA dyes. A postnuclear supernatant was obtained by sequential centrifugation at both 1,200 × g (10 min) and 1,500 × g (5 min). Then the postnuclear supernatant was adjusted to 50% (w/v) sucrose. 100 μl of this suspension was loaded on 225 μl of 62% sucrose in 3 mm imidazole, 0.5 mm EDTA in a TLS-55 centrifugation tube (Beckman) and then sequentially overlaid with 375, 800, 500, and 200 μl of 40.6/35/25 and 8.6% (w/v) sucrose, respectively. The gradient was centrifuged at 135,000 × g for 130 min at 4 °C. After centrifugation, 20 fractions were collected from the bottom using a fraction collector. Proteins in each fraction were precipitated by the addition trichloroacetic acid to a final concentration of 10%. The protein pellet was collected by centrifuging at 15,000 × g at 4 °C during 30 min for later SDS-PAGE analysis. The C-terminal Sequences of H-Ras and K-Ras Target a Soluble Protein to Plasma Membrane and Intracellular Compartments—To investigate the implication of lipid modifications of Ras proteins in membrane association and subcellular distribution in CHO-K1 cells, constructs were engineered to express the C-terminal domains of human H- and K-Ras fused to GFP and its spectral variants (CFP or YFP). Initially, the C-terminal 20 amino acids (KLNPPDESGPGCMSCKCVLS) of H-Ras were fused to the carboxyl terminus of YFP (YFP-H-RasC20), whereas the C-terminal 14 amino acids (KKKKKKSKTKCVIM) of K-Ras were fused to the carboxyl terminus of CFP (CFP-K-RasC14). To evaluate the expression of these fusion proteins, CHO-K1 cells were transiently transfected with the corresponding DNA constructs. The expression of the chimeric proteins was confirmed by Western blotting with an antibody directed to the fluorescent protein. The antibody detected YFP-H-RasC20 as a band of ∼28 kDa and CFP-K-RasC14 as a band of 27.5 kDa. Soluble YFP was revealed as a band of the expected molecular mass (27 kDa) (Fig. 1A). To evaluate whether the fusion proteins are associated to membranes of CHO-K1 cells, cells expressing YFP or YFP-H-RasC20 or CFP-K-RasC14 were mechanically lysed, and the corresponding homogenates were centrifuged at 100,000 × g (Fig. 1B). As expected, more than 95% of YFP was recovered in the soluble fraction. By the contrary, YFP-H-RasC20 and CFP-K-RasC14 were found mainly associated with the particulate fraction (95 and 70%, respectively), strongly suggesting a membrane association. As a control, we analyzed membrane association of GPI-YFP protein, a fusion protein containing the total sequence of YFP fused to a GPI attachment signal (34Crespo P.M. Zurita A.R. Daniotti J.L. J. Biol. Chem. 2002; 277: 44731-44739Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar). As expected, more than 91% of GPI-YFP was found mainly associated to the membrane fraction. To discard an association of YFP-H-RasC20 and CFP-K-RasC14 with insoluble components like cytoskeleton, nuclear remnants, or extracellular matrix, we performed the Triton X-114 partitioning assay, which has been used to monitor the extent of post-translational lipid modifications of prenylated proteins (35Bordier C. J. Biol. Chem. 1981; 256: 1604-1607Abstract Full Text PDF PubMed Google Scholar, 36Gutierrez L. Magee A.I. Marshall C.J. Hancock J.F. EMBO J. 1989; 8: 1093-1098Crossref PubMed Scopus (330) Google Scholar) (Fig. 1C). YFP was found to partition completely into the aqueous phase, according to its hydrophilic nature. In contrast, 81 and 56% of YFP-H-RasC20 and CFP-K-RasC14, respectively, partitioned into the detergent phase of Triton X-114, indicating that a fraction of the expressed proteins are hydrophobic and therefore that YFP-H-RasC20 and CFP-K-RasC14 have been post-translationally modified by lipidation. As expected, the control GPI-YFP was found to partition almost completely into the detergent phase. Palmitoylation of YFP-H-RasC20 was further confirmed by cleavage of thioester-linked fatty acids by hydroxylamine treatment (supplemental Fig. 1). We conclude that YFP-H-RasC20 and CFP-K-RasC14 are correctly expressed and acylated in CHO-K1 cells and mainly associated to the membrane fraction of these cells. To further characterize membrane association and compare subcellular distribution of YFP-H-RasC20 and CFP-K-RasC14, CHO-K1 cells transiently expressing the fusion proteins were fixed and visualized by confocal microscopy. As controls, we also included the analysis of YFP and GPI-YFP subcellular distributions. As shown in Fig. 1D, both YFP-H-RasC20 and CFP-K-RasC14 proteins were associated with plasma membrane, which is in agreement with the subcellular distribution observed for the untagged full-length version of these proteins (24Choy E. Chiu V.K. Silletti J. Feoktistov M. Morimoto T. Michaelson D. Ivanov I.E. Philips M.R. Cell