Biochemical, structural, and functional properties of Rab5 wild-type (WT) protein were compared with those of Q79L and N133I mutants. The detergent 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate increased guanine nucleotide binding to Rab5 WT ≈10-fold. The single-step catalytic rate of Rab5 WT exceeded that of Q79L 12.2-fold, but the steady-state GTPase rate was only 2.8-fold greater because GDP dissociation was rate-limiting and GDP dissociation was 3.6-fold slower than for Q79L. In contrast, dissociation rates of GTP were indistinguishable. Binding to Rab5 N133I was not detectable. GTP protected Rab5 WT and Q79L from any apparent proteolysis by trypsin. A 20-kDa fragment was the major product of digestion in the presence of GDP, and 12- and 8-kDa fragments were the major products in the absence of added guanine nucleotides. Rab5 N133I underwent no apparent proteolysis with 10 mM GTP or GDP, suggesting a "triphosphate" conformation may be induced in Rab5 N133I by either GTP or GDP. Partially geranylgeranylated Rab5 WT stimulated endosome fusion in vitro, whereas unmodified Rab5 WT did not. Processed Rab5 Q79L failed to inhibit endosome fusion, and Rab5 N133I could not be geranylgeranylated. These findings identify biochemical and structural features of Rab5 proteins, providing data for the interpretation of functional assays. Biochemical, structural, and functional properties of Rab5 wild-type (WT) protein were compared with those of Q79L and N133I mutants. The detergent 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate increased guanine nucleotide binding to Rab5 WT ≈10-fold. The single-step catalytic rate of Rab5 WT exceeded that of Q79L 12.2-fold, but the steady-state GTPase rate was only 2.8-fold greater because GDP dissociation was rate-limiting and GDP dissociation was 3.6-fold slower than for Q79L. In contrast, dissociation rates of GTP were indistinguishable. Binding to Rab5 N133I was not detectable. GTP protected Rab5 WT and Q79L from any apparent proteolysis by trypsin. A 20-kDa fragment was the major product of digestion in the presence of GDP, and 12- and 8-kDa fragments were the major products in the absence of added guanine nucleotides. Rab5 N133I underwent no apparent proteolysis with 10 mM GTP or GDP, suggesting a "triphosphate" conformation may be induced in Rab5 N133I by either GTP or GDP. Partially geranylgeranylated Rab5 WT stimulated endosome fusion in vitro, whereas unmodified Rab5 WT did not. Processed Rab5 Q79L failed to inhibit endosome fusion, and Rab5 N133I could not be geranylgeranylated. These findings identify biochemical and structural features of Rab5 proteins, providing data for the interpretation of functional assays. INTRODUCTIONEukaryotic cells maintain a highly compartmented organization, and are capable of ordered and specific transport among different intracellular compartments. A large number of Ras-related GTPases, termed Rabs, have been implicated in distinct steps of intercompartmental transport (for review, see (1Novick P. Brennwald P. Cell. 1993; 75: 597-601Abstract Full Text PDF PubMed Scopus (315) Google Scholar). Regulatory GTPases shuttle between two activity states, which are determined by the phosphorylation status of bound guanine nucleotides(2Dickey B.F. Birnbaumer L. GTPases in Biology. Springer-Verlag, Heidelberg, Germany1993Crossref Google Scholar). It has been proposed that a cycle of regulated nucleotide exchange and GTP hydrolysis is superimposed on the cycling of Rab proteins between donor and acceptor compartments to ensure accurate and directional vesicle transport.Among the known Rab GTPases, Rab5 is of great interest because it appears to be rate-influencing for receptor-mediated and fluid-phase endocytosis. Lateral fusion between endocytic vesicles is stimulated by Rab5 both in vitro and in vivo, and antibodies against Rab5 inhibit fusion in vitro(3Gorvel J.-P. Chavrier P. Zerial M. Gruenberg J. Cell. 1991; 64: 915-925Abstract Full Text PDF PubMed Scopus (856) Google Scholar, 4Bucci C. Parton R.G. Mather I.H. Stunnenberg H. Simons K. Hoflack B. Zerial M. Cell. 1992; 70: 715-728Abstract Full Text PDF PubMed Scopus (1116) Google Scholar, 5Li G. Stahl P.D. J. Biol. Chem. 1993; 268: 24475-24480Abstract Full Text PDF PubMed Google Scholar). Endosome fusion in vitro is inhibited by cytosol containing overexpressed mutant Rab5 N133I protein that has impaired guanine nucleotide binding (3Gorvel J.-P. Chavrier P. Zerial M. Gruenberg J. Cell. 1991; 64: 915-925Abstract Full Text PDF PubMed Scopus (856) Google Scholar). In cells overexpressing Rab5 N133I, the rate of receptor-mediated and fluid-phase endocytosis is significantly decreased compared with control(4Bucci C. Parton R.G. Mather I.H. Stunnenberg H. Simons K. Hoflack B. Zerial M. Cell. 1992; 70: 715-728Abstract Full Text PDF PubMed Scopus (1116) Google Scholar, 5Li G. Stahl P.D. J. Biol. Chem. 1993; 268: 24475-24480Abstract Full Text PDF PubMed Google Scholar). The N133I mutant is believed to interfere with endocytosis by interacting nonproductively and competitively with some important component of the endocytic apparatus.If Rabs must cycle between GDP- and GTP-bound forms in order to function, then a mutation reducing the GTPase activity (such as the cognate of H-Ras Q61L) is predicted to also be a dominant inhibitor. This is the case with Rab2, but not with Rab1(6Tisdale E.J. Bourne J.R. Khosravi-Far R. Der C.J. Balch W.E. J. Cell Biol. 1992; 119: 749-761Crossref PubMed Scopus (417) Google Scholar), and the Sec4 mutation has an intermediate phenotype(7Walworth N.C. Brennwald P. Kabcenell A.K. Garrett M. Novick P. Mol. Cell. Biol. 1992; 12: 2017-2028Crossref PubMed Scopus (107) Google Scholar). Wild-type Rab25 actually contains Lys at the cognate position(8Goldenring J.R. Shen K.R. Vaughan H.D. Modlin I.M. J. Biol. Chem. 1993; 268: 18419-18422Abstract Full Text PDF PubMed Google Scholar). The molecular basis for the functional differences among these related proteins is not known.Although several in vivo and in vitro functional studies of Rab5 have been published, this protein has not yet been subjected to a detailed biochemical analysis. Correct interpretation of in vivo and in vitro experiments using WT 1The abbreviations used are: WTwild-typeCHAPS3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonatePAGEpolyacrylamide gel electrophoresisGTPγSguanosine 5′-O-(3-thiotriphosphate)GAPGTPase activating protein. and mutant Rab5 proteins depends on thorough knowledge of the biochemistry of these reagents. As a step toward understanding the role of Rab5 in endosome fusion and endocytosis, we have expressed in Escherichia coli Rab5 WT, its N133I mutant, and a putative GTPase defective mutant (Q79L). The biochemical properties, particularly the kinetics of nucleotide binding and GTPase activities, of purified recombinant proteins were characterized; functional properties of the purified proteins were studied using an in vitro endosome fusion assay to verify predicted phenotypes; and nucleotide-dependent structural properties of the proteins were analyzed by limited proteolysis.MATERIALS AND METHODSConstruction of Rab5 MutantsRecombinant plasmids with human Rab5 and Rab4 cDNAs were obtained from A. Tavitian(9Zahraoui A. Touchot N. Chardin P. Tavitian A. J. Biol. Chem. 1989; 264: 12394-12401Abstract Full Text PDF PubMed Google Scholar). The cDNA inserts were amplified by add-on polymerase chain reaction and cloned into M13mp18 for nucleotide sequencing(10Sanger F. Nicklen S. Coulson A.R. Proc. Natl. Acad. Sci. U. S. A. 1977; 74: 5463-5467Crossref PubMed Scopus (52357) Google Scholar). DNA sequence analysis of Rab5 clones from several independent polymerase chain reactions identified three discrepancies from the published sequence(9Zahraoui A. Touchot N. Chardin P. Tavitian A. J. Biol. Chem. 1989; 264: 12394-12401Abstract Full Text PDF PubMed Google Scholar) : Arg replacing Gly at position 81, Val replacing Ala-86, and Arg replacing Gly-197. The Gly to Arg conversions are identical to residues found in these positions (amino acids 81 and 197) in canine Rab5(11Chavrier P. Gorvel J.P. Stelzer E. Simons K. Gruenberg J. Zerial M. Nature. 1991; 353: 769-772Crossref PubMed Scopus (318) Google Scholar). The remaining alteration, Val to Ala, is conservative and most likely does not affect the functional integrity of the protein; in fact, the product of this cDNA can bind and hydrolyze GTP, can be geranylgeranylated, and functions in an endosome fusion assay (see "Results and Discussion"). Site-directed point mutagenesis of Rab5 was performed by the method of Kunkel et al.(12Kunkel T.A. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 488-492Crossref PubMed Scopus (4886) Google Scholar). Two point mutants, Q79L and N133I, were generated using the oligonucleotides 5′-GGTATCGTTCTAGACCAGCTGTA-3′ and 5′-AGGTCCGGCCTTGATTCCCGATAAAG-3′, respectively. Mutants were verified by nucleotide sequence analysis in both directions. WT and mutant Rab5 cDNAs and the Rab4 WT cDNA were cloned into the T7-polymerase expression plasmid pT7.7 (13Tabor S. Richardson C.C. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 1074-1078Crossref PubMed Scopus (2437) Google Scholar) utilizing the NdeI cloning site to ensure the production of recombinant proteins with native amino termini.Purification of Rab ProteinsE. coli BL21(DE3) cells were transformed with recombinant plasmids and grown in LB medium containing 50 μg/ml ampicillin. When the A600 was ≈ 1.0, isopropyl-1-thio-β-D-galactopyranoside was added to a final concentration of 0.8 mM. Cells were harvested 3-4 h later by centrifugation at 1500 × g for 30 min. The cell pellet from a 2-liter culture was resuspended in Tris-buffered saline (20 mM Tris-HCl, pH 8.0, 150 mM NaCl), centrifuged, and then washed with the same buffer. The washed pellet was suspended in 10 ml of hypertonic buffer (2.4 M glucose, 20 mM Tris-HCl, pH 8.0, 1 mM EDTA) and kept on ice for 30 min. One hundred ml of a lysozyme solution (0.5 mg/ml lysozyme, 0.02 mg/ml DNase, 40 mM Tris-HCl, pH 8.0, 5 mM MgCl2, 1 mM EDTA) was then added, and the incubation was continued on ice for a further 15 min. One hundred ml of lysis buffer (20 mM Tris-HCl, pH 8.0, 1 mM EDTA, 5 mM MgCl2, 1 μM GDP, 2 mM β-mercaptoethanol, 0.2 mM phenylmethylsulfonyl fluoride) was then added. The lysate was centrifuged in a GSA rotor (Sorvall) at 11,000 rpm for 30 min, and the supernatant was filtered through a glass fiber filter. CHAPS was added to a final concentration of 0.1% at this point, except where noted. This solution was applied to a 500-ml DEAE-Sepharose FF column equilibrated with several volumes of Buffer A (20 mM Tris-HCl, pH 8.0, 1 mM EDTA, 2 mM β-mercaptoethanol, 5 mM MgCl2, 1 μM GDP, 0.1% CHAPS). Protein that did not bind to the resin was collected and concentrated on an Amicon YM-10 membrane and then applied to a 1500-ml Sephacryl S-200 column equilibrated with Buffer A plus 150 mM NaCl. Rab proteins in eluting fractions were identified by SDS-PAGE, and concentrated by membrane filtration. The N133I mutant was insoluble when expressed in E. coli, similar to the cognate Ras mutant(14Clanton D.J. Hattori S. Shih T.Y. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 5076-5080Crossref PubMed Scopus (67) Google Scholar). Therefore, the protein was extracted from the bacterial pellet with 8 M urea, which was removed during chromatography on the gel filtration column. Rab4 was purified by sequential chromatography as for Rab5 WT, except that Rab4 bound to DEAE-Sepharose and was eluted with a linear gradient of NaCl in Buffer A. Protein concentrations were determined with the Coomassie Plus assay (Pierce), using bovine serum albumin as a standard. Proteins were stored at 2-5 mg/ml in Buffer A at −80°C.Binding of Nucleotides in SolutionNucleotide binding was determined by a rapid filtration technique(15Northup J. Smigel M. Gilman A. J. Biol. Chem. 1982; 257: 11416-11423Abstract Full Text PDF PubMed Google Scholar). Binding buffer contained 20 mM Tris-HCl, pH 8.0, 250 mM NaCl, 1 mM EDTA, 5 mM MgCl2, 1 mM dithiothreitol, 0.1% CHAPS, 500 nM guanine nucleotide, and either [35S]GTPγS, [α-32P]GTP, or [3H]GDP (all 2,000 cpm/pmol). The protein concentration was 50 nM in a volume of 100 μl. After incubation at 30°C for the indicated times, samples were diluted with 4 ml of iced buffer, filtered through BA85 nitrocellulose (Schleicher and Schuell), and then washed twice with 4 ml of iced buffer. Filters were then dried, immersed in scintillation fluid, and counted by scintillation counting. Data points in all figures represent the mean of triplicate determinations from a representative experiment that was repeated 2-4 times. Except where noted, maximum binding was determined by extrapolation from experimental data using the program Enzfitter (Biosoft, Ferguson, MO). In the presence of 0.1% CHAPS and 5 mM MgCl2, maximum binding averaged 1.2 ± 0.2 mol of nucleotide/mol of Rab5.Measurement of Dissociation RatesThese were measured as described previously(16Hall A. Self A.J. J. Biol. Chem. 1986; 261: 10963-10965Abstract Full Text PDF PubMed Google Scholar). Protein was preincubated for 3 h in the conditions described above for nucleotide binding, and then the bound labeled nucleotide was competed with 1 mM unlabeled nucleotide. At indicated times, samples were diluted with cold buffer, filtered, and counted, as above. In some experiments, Rab5 was rapidly loaded with radiolabeled guanine nucleotides by a modification of the procedure of Tucker et al.(17Tucker J. Sczakiel G. Feuerstein J. John J. Goody R.S. Wittinghofer A. EMBO J. 1986; 5: 1351-1358Crossref PubMed Scopus (225) Google Scholar) as follows. Protein was incubated in the presence of 1.25 mM EGTA for 10 min at 30°C to release prebound nucleotide and then radiolabeled nucleotide was added, followed by 5 mM MgCl2.GTPase ActivitySteady state GTPase activity was measured by the release of [32P]Pi using the "charcoal assay" as described(18Kabcenell A.K. Goud B. Northup J.K. Novick P.J. J. Biol. Chem. 1990; 265: 9366-9372Abstract Full Text PDF PubMed Google Scholar). Protein was incubated with 500 nM [γ-32P]GTP (2,000 cpm/pmol) in the presence of 5 mM MgCl2 at 30°C, and at the indicated times aliquots were withdrawn and mixed with charcoal. Following centrifugation, the [32P]Pi in the supernatants was assayed by scintillation counting. Pre-steady state GTPase rates were estimated by the charcoal method at early times following rapid loading of Rab5 as above. The single-step GTPase rate was determined by a filtration assay following rapid nucleotide loading. This assay was performed essentially as described for the measurement of dissociation rates, with the exceptions that [γ-32P]GTP was used, and the GTPase rate was calculated by subtracting the [35S]GTPγS dissociation rate from the apparent GTPase rate(19Nagata K.I. Suzuki T. Shibagaki Y. Mizumoto K. Okano Y. Kaziro Y. Nozawa Y. J. Biol. Chem. 1992; 267: 19600-19606Abstract Full Text PDF PubMed Google Scholar). All rates were calculated using the program Enzfitter.In Vitro Processing of Recombinant Rab5 ProteinsPost-translational modification of Rab proteins with geranylgeranyl was accomplished by an in vitro reaction supported by rabbit reticulocyte lysate (Promega, Madison, WI), as described(20Sanford J.C. Pan Y. Wessling-Resnick M. J. Biol. Chem. 1993; 268: 23773-23776Abstract Full Text PDF PubMed Google Scholar). Briefly, Rab proteins (final concentration, 0.1-0.8 mg/ml) were added to lysate together with 5 μM [3H]geranylgeranyl pyrophosphate (33,300 dpm/pmol, American Radiolabeled Chemicals) in 10 mM Tris, pH 8.0, 0.1% CHAPS, 1 μM GDP, 5 mM MgCl2, 1 mM EDTA, 2 mM β-mercaptoethanol; the final mixture contained 40-60% reticulocyte lysate (v/v). The reaction mixtures were incubated at 37°C for 3-4 h, after which time aliquots (3.4 μl) were removed for analysis by PAGE and fluorography. The amount of [3H]geranylgeranyl incorporated into the proteins was determined by excising the bands from the dried gel, dissolving samples in 30% H2O2 at 65°C, and scintillation counting. After processing but prior to the addition to in vitro endosome fusion assays, the remaining reaction mixture was chromatographed over a Sephadex G25 spin column to remove excess radioactivity and to exchange the Rab protein into the appropriate buffer (20 mM Hepes, pH 7.4, 0.1 M KCl, 85 mM sucrose, 20 μM EGTA).In Vitro Endosome Fusion AssayCell-free vesicle fusion assays were performed exactly as described previously(21Wessling-Resnick M. Braell W.A. J. Biol. Chem. 1990; 265: 690-699Abstract Full Text PDF PubMed Google Scholar, 22Wessling-Resnick M. Braell W.A. J. Biol. Chem. 1990; 265: 16751-16759Abstract Full Text PDF PubMed Google Scholar). Postnuclear supernatant fractions were prepared from K562 cells that had endocytosed either biotinylated transferrin (BTf) or avidin-β-galactosidase (AvβGal) at 20°C for 45 min to load early endosomes. The postnuclear supernatant fractions were then dialyzed against 20 mM Hepes, pH 7.4, 0.1 M KCl, 85 mM sucrose, 20 μM EGTA, and frozen at −80°C until use. Postnuclear supernatant fractions were rapidly thawed immediately prior to the assay, and 5-μl aliquots of each were added to a reaction mixture that contained 1 mM MgATP, 50 μg/ml creatine kinase, 8 mM phosphocreatine, 10 μg/ml biotin-insulin, 1 mM dithiothreitol, 400 μM GTP, with appropriate additions of Rab proteins and/or rabbit reticulocyte lysate as detailed in the legend to Table 1. Fusion activity was promoted by incubation of the reaction mixtures at 37°C, while control reactions (4°C) were held on ice. Specific vesicle fusion results in co-localization of AvβGal and BTf within fused vesicles, and the resulting avidin:biotin complex between the two probes was measured by a modified enzyme-linked immunosorbent assay using a fluorogenic β-galactosidase substrate as detailed previously (21Wessling-Resnick M. Braell W.A. J. Biol. Chem. 1990; 265: 690-699Abstract Full Text PDF PubMed Google Scholar). The signal in fluorescence units is directly proportional to the extent of endocytic vesicle fusion in the in vitro reaction. Samples (15-20 μl) of desalted prenylation reaction mixtures were tested in each assay.Table I Open table in a new tab RESULTS AND DISCUSSIONExpression of Rab5 and Two of Its MutantsInduction of E. coli containing recombinant plasmids resulted in the accumulation of a major protein of 25 kDa in lysates as assessed by SDS-PAGE and Coomassie Blue staining (Fig. 1, lane2). Prior to purification, the predicted biochemical phenotypes of WT and mutant proteins were qualitatively confirmed by [32P]GTP overlay assays. Lysates were Western blotted onto nitrocellulose filters and incubated with [α-32P]GTP (Fig. 2A) or [γ-32P]GTP (Fig. 2B). Autoradiography revealed that Western blotted Rab5 WT bound and hydrolyzed GTP, as indicated by reduced density of the 25-kDa band with γ-labeled GTP as compared with α-labeled GTP. In contrast, Rab5 Q79L showed a much smaller difference in band intensity between γ-labeled and α-labeled GTP, suggesting a reduced GTPase activity as expected from similar effects of the cognate mutation in Ras (23Der C.J. Finkel T. Cooper G.M. Cell. 1986; 44: 167-176Abstract Full Text PDF PubMed Scopus (399) Google Scholar) and in other Rabs (7Walworth N.C. Brennwald P. Kabcenell A.K. Garrett M. Novick P. Mol. Cell. Biol. 1992; 12: 2017-2028Crossref PubMed Scopus (107) Google Scholar, 19Nagata K.I. Suzuki T. Shibagaki Y. Mizumoto K. Okano Y. Kaziro Y. Nozawa Y. J. Biol. Chem. 1992; 267: 19600-19606Abstract Full Text PDF PubMed Google Scholar, 24Brondyk W.H. McKiernan C.J. Burstein E.S. Macara I.G. J. Biol. Chem. 1993; 268: 9410-9415Abstract Full Text PDF PubMed Google Scholar). Rab5 N133I had a severe guanine nucleotide binding defect, as indicated by the complete absence of an autoradiographic signal with nucleotide labeled in either position, consistent with previous results (4Bucci C. Parton R.G. Mather I.H. Stunnenberg H. Simons K. Hoflack B. Zerial M. Cell. 1992; 70: 715-728Abstract Full Text PDF PubMed Scopus (1116) Google Scholar, 5Li G. Stahl P.D. J. Biol. Chem. 1993; 268: 24475-24480Abstract Full Text PDF PubMed Google Scholar). Cognate mutations in Ras (14Clanton D.J. Hattori S. Shih T.Y. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 5076-5080Crossref PubMed Scopus (67) Google Scholar, 25Walter M. Clark S.C. Levinson A.D. Science. 1986; 233: 649-652Crossref PubMed Scopus (85) Google Scholar) and in other Rabs (6Tisdale E.J. Bourne J.R. Khosravi-Far R. Der C.J. Balch W.E. J. Cell Biol. 1992; 119: 749-761Crossref PubMed Scopus (417) Google Scholar, 24Brondyk W.H. McKiernan C.J. Burstein E.S. Macara I.G. J. Biol. Chem. 1993; 268: 9410-9415Abstract Full Text PDF PubMed Google Scholar, 26Walworth N.C. Goud B. Kabcenell A.K. Novick P.J. EMBO J. 1989; 8: 1685-1693Crossref PubMed Scopus (197) Google Scholar, 27Schmitt H.D. Wagner P. Pfaff E. Gallwitz D. Cell. 1986; 47: 401-412Abstract Full Text PDF PubMed Scopus (169) Google Scholar) result in undetectable guanine nucleotide binding as assayed by rapid filtration or GTP overlay. This severe defect in nucleotide binding is not surprising in view of the central role of the highly conserved asparagine in linking three structural elements directly involved in nucleotide binding, as inferred from the crystal structures of Ras and EF-Tu(28Wittinghofer A. Pai E.F. Trends. Biochem. Sci. 1991; 16: 382-387Abstract Full Text PDF PubMed Scopus (231) Google Scholar).Figure 2:Binding of [32P]GTP to Western blotted proteins. Crude proteins (2 μg/lane) were resolved by SDS-PAGE, electrotransferred to nitrocellulose, and then incubated with (A) [α-32P]GTP or (B) [γ-32P]GTP (both 106 cpm/pmol) for 1 h at room temperature. The nitrocellulose was then washed for 1 h, dried, and exposed to x-ray film as described(57Dexter D. Rubins J.B. Manning E.C. Khachatrian L. Dickey B.F. J. Immunol. 1990; 145: 1845-1850PubMed Google Scholar).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Detergent Effects on Guanine Nucleotide Binding to Rab5During our initial purification of Rab5, much of the partially fractionated protein precipitated. Efforts were therefore directed at stabilizing Rab5 in solution without losing [35S]GTPγS binding activity. Several nonionic and zwitterionic detergents caused a dramatic increase in [35S]GTPγS binding at a concentration of 0.05%, as did 0.01% bovine serum albumin, but there was no detectable binding in the presence of the ionic detergent SDS (not shown). When the concentration of the detergents was increased to 0.5%, only CHAPS still supported high [35S]GTPγS binding. Hence, subsequent protein purification and biochemical experiments were performed in buffers containing 0.1% CHAPS unless indicated.The effect of 0.1% CHAPS on [35S]GTPγS binding to Rab5 was time-dependent as shown in Fig. 3A. To test the hypothesis that the increase in [35S]GTPγS binding induced by CHAPS might be due to acceleration of prebound nucleotide dissociation, 100 nM [3H]GDP was loaded onto Rab5 by transient magnesium chelation. However, the dissociation of [3H]GDP from Rab5 in the presence of 0.1% CHAPS was slower than in its absence (Fig. 3B, diamonds). When excess unlabeled GDP was not added after loading [3H]GDP, binding remained stable in the presence of CHAPS but declined rapidly in its absence (Fig. 3B, circles). These results suggest that without added detergent, Rab5 quickly assumes a conformation that does not bind guanine nucleotides with high affinity. This conclusion is consistent with the decline in the small amount of [35S]GTPγS initially bound to Rab5 in the absence of CHAPS during the course of prolonged incubation (Fig. 3A, opencircles). It should be noted that the recombinant protein is not post-translationally modified when expressed in E. coli, and therefore lacks geranylgeranyl groups that are attached to Rab proteins in eukaryotes(29Kinsella B.T. Maltese W.A. J. Biol. Chem. 1992; 267: 3940-3945Abstract Full Text PDF PubMed Google Scholar, 30Khosravi-Far R. Clark G.J. Abe K. Cox A.D. McLain T. Lutz R.J. Sinensky M. Der C.J. J. Biol. Chem. 1992; 267: 24363-24368Abstract Full Text PDF PubMed Google Scholar).Figure 3:Effects of CHAPS on guanine nucleotide binding to Rab5. A, association of [35S]GTPγ S. Recombinant Rab5 WT purified in the absence of CHAPS was incubated with 500 nM [35S]GTPγ S either in the absence or presence of 0.1% CHAPS for 240 min at 30°C or was incubated initially in the absence of CHAPS for 90 min and then in the presence of 0.1% CHAPS for the next 150 min. At the indicated times, aliquots were removed and rapidly filtered through nitrocellulose. B, dissociation of [3H]GDP. Rab5 WT was loaded with [3H]GDP by transient magnesium chelation followed immediately by the addition of binding buffer with or without 0.1 mM GDP and with or without 0.1% CHAPS (final concentrations). Aliquots were removed at the indicated times and assayed by vacuum filtration through nitrocellulose.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The findings with Rab5 contrast with the lack of effect of CHAPS on nucleotide binding to unprocessed Rab6 (31Yang C. Mollat P. Chaffotte A. McCaffrey M. Cabanié L. Goud B. Eur. J. Biochem. 1993; 217: 1027-1037Crossref PubMed Scopus (18) Google Scholar) but are not unique to Rab5 in that similar effects were observed by us with bacterial recombinant Rab4 (not shown). Of note, both nucleotide binding and GTPase activities of Rab6 differ dramatically between the processed and unprocessed forms(31Yang C. Mollat P. Chaffotte A. McCaffrey M. Cabanié L. Goud B. Eur. J. Biochem. 1993; 217: 1027-1037Crossref PubMed Scopus (18) Google Scholar). Together with our results, this suggests that conformational effects on the nucleotide binding and hydrolyzing site of Rab proteins may be induced either by covalently attached prenyl groups or by noncovalently associated detergents, and such effects may vary from one Rab protein to another. Supporting the involvement of the carboxyl terminus of Rab5 in guanine nucleotide binding was our finding that Rab5 truncated after amino acid 184 (cognate of H-Ras166) was insoluble in E. coli (not shown), similar to Rab5 N133I and to several Ras mutants that are both defective in nucleotide binding and insoluble in E. coli(14Clanton D.J. Hattori S. Shih T.Y. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 5076-5080Crossref PubMed Scopus (67) Google Scholar). In contrast, truncated H-Ras166 is soluble in E. coli and has guanine nucleotide binding and hydrolyzing properties indistinguishable from those of the full-length protein (32John J. Schlichting I. Schiltz E. Rösch P. Wittinghofer A. J. Biol. Chem. 1989; 264: 13086-13092Abstract Full Text PDF PubMed Google Scholar). In addition, reciprocal interactions between guanine nucleotide binding and guanine nucleotide dissociation inhibitor binding to carboxyl-terminally prenylated Rabs have been described for other Rab proteins(1Novick P. Brennwald P. Cell. 1993; 75: 597-601Abstract Full Text PDF PubMed Scopus (315) Google Scholar), suggesting an interaction between the carboxyl terminus and the guanine nucleotide-binding domains. Of interest, guanine nucleotide binding by ADP-ribosylation factor, another small GTPase that regulates vesicular traffic, is highly dependent on interactions with lipids and detergents when its acylated amino terminus is intact (33Kahn R.A. Randazzo P. Serafini T. Weiss O. Rulka C. Clark J. Amherdt M. Roller P. Orci L. Rothman J.E. J. Biol. Chem. 1992; 267: 13039-13046Abstract Full Text PDF PubMed Google Scholar).Purification of Bacterial Recombinant Rab5 and Two of Its MutantsAfter stabilization of Rab5 WT and Q79L in solution with 0.1% CHAPS, 1 μM GDP, and 5 mM MgCl2, these proteins were readily purified from the supernatants of E. coli lysates. Rab5 WT (calculated pI, 7.05) did not bind to DEAE-Sepharose at pH 8.0 and was recovered in the flow-through, while the bulk of contaminating E. coli protein remained on the column (Fig. 1, lane4). Most of the residual contaminating protein was resolved by molecular sieve chromatography, yielding Rab5 WT of approximately 95% purity as estimated by Coomassie staining after SDS-PAGE (Fig. 1, lane5). A minor 20-kDa contaminant appeared to be a degradation product of Rab5 due to carboxyl-terminal proteolysis since it bound [32P]GTP on a GTP overlay, it reacted with affinity-purified antiserum to Rab5 holoprotein, and the amino terminus was intact by Edman degradation (not shown). Rab5 Q79L was purified by an identical scheme and yielded protein of similar purity (see Fig. 8B). Rab5 N133I was insoluble in E. coli lysates and was extracted with 8 M urea. Urea was removed during the molecular sieve chromatography, but approximately 90% of the protein subsequently precipitated. The remaining Rab5 N133I (see Fig. 8D) was fairly stable in solution in Buffer A containing 1 mM GDP.Figure 8:Proteolysis of Rab5 proteins by trypsin. Purified recombinant Rab5 WT (A), Q79L (B), or N133I (D), were preincubated in the absence or the presence of 10 mM nucleotides and 5 mM MgCl2 for 1 h at 30°C. Proteins (2.5 μg) were then incubated for 1 h with or without 0.25 μg of trypsin in the presence of the indicated nucleotides in a total volume of 50 μl at 30°C, as described previously(42Goffenberg S.I. Panchenko M.P. Zhuravlev I.V. Tkachuk V.A. Biochemistry (Engl. Transl. Biokhimiya). 1990; 55: 653-660Google Scholar), and boiled for 5 min in sample buffer, and the re
ABSTRACT During prolonged exposure to agonist, β2-adrenergic receptors undergo downregulation, defined by the loss of radioligand binding sites. To determine the cellular basis for β2-adrenergic receptor downregulation, we examined HEK293 cells stably expressing β2-adrenergic receptors with an N-terminal epitope tag. Downregulation was blocked by leupeptin, a cysteine protease inhibitor, but not by pepstatin, an inhibitor of aspartate proteases. Immunofluorescence microscopy of cells treated with agonist for 3-6 hours in the presence of leupeptin showed β2-adrenergic receptors, but not transferrin receptors, localizing with the lysosomal protease cathepsin D, and with lysosomes labeled by uptake of a fluorescent fluid-phase marker. No localization of β2-adrenergic receptors with lysosomal markers was observed in the absence of leupeptin, most likely due to proteolysis of the epitope. The proton pump inhibitor, bafilomycin A1, significantly inhibited this agonist-induced redistribution of β2-adrenergic receptors into lysosomes, causing receptors to accumulate in the rab11-positive perinuclear recycling compartment and slowing the rate of β2-adrenergic receptor recycling. Control experiments showed that leupeptin had no nonspecific effects on the cellular trafficking of either β2-adrenergic receptors or transferrin receptors. Although cAMP alone caused a small decline in receptor levels without redistributing β2-adrenergic receptors from the plasma membrane, this effect was additive to that seen with agonist alone, suggesting that agonist-induced β2-adrenergic receptor downregulation resulted largely from cAMP-independent mechanisms. These results indicate that during agonist-induced downregulation, a significant fraction of β2-adrenergic receptors are specifically sorted to lysosomes via the endosomal pathway, where receptor degradation by cysteine proteases occurs. These results provide a cellular explanation for the loss of radioligand binding sites that occurs during prolonged exposure to agonist.
The expression of the X and Y pseudo-ovalbumin genes is stimulated by estrogen to a much lesser extent than the expression of the authentic ovalbumin gene in the chicken oviduct. Since it is possible that the primary structures of the 5' flanking regions of these genes are responsible for their differential hormonal responses, we have identified the 5' transcription domains of the X and Y pseudo-ovalbumin genes and determined their nucleotide sequences. Similar to many other eukaryotic genes, the X, Y, and ovalbumin genes each contain an (A + T)-rich heptamer located about 30 nucleotides upstream from the cap site. This sequence is TATATAT for the X and ovalbumin genes, but GATATAT for the Y gene. The 5' flanking sequences of all three genes are about 70% homologous when allowances are made for deletions and insertions. There is no obvious feature of the 5' flanking sequences of the pseudo-ovalbumin genes which can be related to differential hormonal responsiveness. Nevertheless, these 5' flanking regions appear to have been conserved relative to the intervening sequences of the pseudogenes, and thus may be important to gene function.
The mostly widely used bronchodilators in asthma therapy are β2-adrenoreceptor (β2AR) agonists, but their chronic use causes paradoxical adverse effects. We have previously determined that β2AR activation is required for expression of the asthma phenotype in mice, but the cell types involved are unknown. We now demonstrate that β2AR signaling in the airway epithelium is sufficient to mediate key features of the asthmatic responses to IL-13 in murine models. Our data show that inhibition of β2AR signaling with an aerosolized antagonist attenuates airway hyperresponsiveness (AHR), eosinophilic inflammation, and mucus-production responses to IL-13, whereas treatment with an aerosolized agonist worsens these phenotypes, suggesting that β2AR signaling on resident lung cells modulates the asthma phenotype. Labeling with a fluorescent β2AR ligand shows the receptors are highly expressed in airway epithelium. In β2AR-/- mice, transgenic expression of β2ARs only in airway epithelium is sufficient to rescue IL-13-induced AHR, inflammation, and mucus production, and transgenic overexpression in WT mice exacerbates these phenotypes. Knockout of β-arrestin-2 (βarr-2-/-) attenuates the asthma phenotype as in β2AR-/- mice. In contrast to eosinophilic inflammation, neutrophilic inflammation was not promoted by β2AR signaling. Together, these results suggest β2ARs on airway epithelial cells promote the asthma phenotype and that the proinflammatory pathway downstream of the β2AR involves βarr-2. These results identify β2AR signaling in the airway epithelium as capable of controlling integrated responses to IL-13 and affecting the function of other cell types such as airway smooth muscle cells.
Based upon analogies to recent changes in the treatment of heart failure, we previously showed that chronic administration of certain ‘beta‐blockers’ decreased airway hyperresponsiveness (AHR) in a murine model of allergic asthma. To elucidate the mechanism for the beneficial effects, we used the same murine model to examine the effects of chronic administration of beta‐blockers on inflammation and mucous metaplasia, cardinal features of asthma that may contribute to AHR. Chronic administration of beta‐blockers reduced the total cell counts and eosinophils in bronchoalveolar lavage (BAL) of antigen‐challenged and sensitized mice. There was also reduced mucin production and goblet cell formation (mucous metaplasia) as determined by periodic acid Schiff's staining. A similar effect was observed by a second laboratory using a different strain of mice (C57BL/6J instead of BALB/cJ), with ~ 90% reduction in eosinophils and mucous metaplasia. These results were confirmed by a decrease in Muc5AC staining of airway epithelium using immunohistochemistry and Muc5AC transcript levels. BAL cytokine levels of IL‐13, IL‐10, IL‐5, and TGF‐â1 were also decreased. These observations suggest that asthmatics may benefit from treatment with beta‐blockers given their ability to reduce various indices of airway inflammation including mucous metaplasia, and inflammatory cells and cytokines in BAL.
