Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

The mammalian G protein rhoC is ADP‐ribosylated by Clostridium botulinum exoenzyme C3 and affects actin microfilaments in Vero cells.

The mammalian G protein rhoC is ADP‐ribosylated by Clostridium botulinum exoenzyme C3 and affects... Journal vol.8 no.4 pp. 1 087 - 1092, 1989 The EMBO The mammalian G protein rhoC is ADP-ribosylated by Clostridium botulinum exoenzyme C3 and affects actin microfilaments in Vero cells named rhoA (formerly clone 1 or 12) (Yeramian humans, P.Chardin, P.Boquet', P.Madaule2, et al., 1987), rhoB (clone 6) and rhoC (clone 9) (Chardin M.R.Popoff3, E.J.Rubin4 and D.M.Gill4 Hitherto their functions have not been well et al., 1988). Institut National de la Sante et de la Recherche Medicale U-248, but now some information can be inferred from the defined Facult6 de Medecine Lariboisiere Saint-Louis, 10, Avenue de Verdun, of ADP-ribosylation. C3 treatment of Vero cells effects 75010 Paris, 'Unit6 des Antigenes Bacteriens, Institut Pasteur, 28 rue results in the disappearance of actin filaments which suggests du Dr Roux, 75724 Paris, 2Laboratoire de Physiologie Nerveuse, that the rho product may control microfilament assembly. CNRS, 91198 Gif-sur-Yvette, 3Unite des Anaerobies, Institut Pasteur, The change in actin status presumably accounts for some 25 rue de Dr Roux, 75015 Paris, France and 4Department of Molecular Biology and Microbiology, Tufts University, 136 Harrison morphological consequences of C3 action. Avenue, Boston, MA 021 1 1, USA Communicated by S.Olsnes Results botulinum C3 is a recently discovered exo- Clostridium ADP-ribosylation of rhoC protein that ADP-ribosylates a eukaryotic GTP-binding enzyme protein of the ras superfamily. We show now that the The human rhoC cDNA was subcloned in the bacterial ptac (de Boer et al., 1983) and the rhoC bacterially-expressed product of the human rhoC gene expression vector was expressed in Escherichia coli strain JM 109 upon is ADP-ribosylated by C3 and corresponds in size, charge protein of isopropyl-fl-D-thiogalactoside (IPTG), as and behavior to the dominant C3 substrate of eukaryotic the addition in Materials and methods. There was too little cells. C3 treatment of Vero cells results in the disappear- described to detect directly by gel staining but the presump- ance of microfilaments and in actinomorphic shape product rhoC was detected by [32P]GTP binding after changes without any apparent direct effect upon actin. tive protein transfer of proteins to nitrocellulose. The rho Thus the ADP-ribosylation of a rho protein seems to be electrophoretic microfilament disassembly and we infer was readily ADP-ribosylated when a bacterial extract responsible for protein form of a rho protein may be incubated with [32P]NAD and C3 (Figure 1). At an that the unmodified was in control. elevated NAD concentration (100 the amount of ADP- involved cytoskeletal IM) words: actin/ADP-ribosylation/C3/Clostridium botu- ribosylated rhoC protein reached a stable plateau with time Key of incubation or C3 concentration. The limit was typically linumIG protein - 0.2 (4 ng/yl) of bacterial extract. Two orders of pmol/lil rhoC protein was available for reaction in magnitude less The ADP-ribosylated product had the same uninduced cells. Introduction mobility as the main C3 product of human electrophoretic molecular mass of both was -21 000 Bacterial enzymes that ADP-ribosylate proteins have proven cells; the apparent Two-dimensional gel analysis showed that the invaluable for studying the functions of their eukaryotic (Figure 1). and human cell products also had the same charge particularly of G proteins. Exoenzyme C3 is a newly bacterial targets, A less intense spot of the same charge (pl 5.6) but described ADP-ribosyl transferase secreted by C and D (pl 5.6). size (Mr 23 000) is present in both cases strains of Clostridium botulinum but distinct from the of a larger apparent 2) and provides additional evidence of identity. It classical C1 and D neurotoxins. In many eukaryotic cells, (Figure substrate which is found mostly in runs closer to the anticipated position of ADP-ribosylated C3 has a single protein - the same charge the two forms fraction et al., 1988; Rubin et al., rhoC (Mr 22 500). Having the soluble (Aktories and may simply represent of this substrate, - 21 000 Mr, and its inter- cannot be related by proteolysis 1988). The size isomers. Partial proteolysis by the method us to suggest that it is a member of conformational action with GTP led et al. generated the same series of of proteins (Rubin et al., 1988). of Cleveland (1977) the ras superfamily from each of the apparent size variants causes maturational changes radioactive peptides C3-catalysed ADP-ribosylation and distinct morphological and physio- (not shown). in Xenopus oocytes mouse and calf tissues, including thymus, in cultured cells (Rubin et al., 1988) which Homogenized logical changes and patterns similar heart, kidney spleen, gave those caused by microinjected activated ras muscle, lung, partially overlap cells in 2. to that of the HeLa Figure The C3 substrate is, however, distinct from p21rI protein. in its choice of protein substrate. C3 is selective We show now that the bacterially-expressed rhoC highly itself. substrate was detected in the parental E. coli No C3 of human origin is ADP-ribosylated by C3 and specific gene product or HB11. there was no ADP- substrate. The rho genes strains JM109 Likewise, with the eukaryotic coincides substrate in extracts of E. coli cells that and evolutionarily conserved branch of ribosylatable a distinct represent of the ras-related of other members family proteins a 30% sequence identity to the the ras showing expressed superfamily rab4 and ral, rab2, rabS) (Chardin and Axel, 1985; Madaule (c-H-ras, raplA, rap2, classical ras proteins (Madaule Touchot et by al., 1987, expression a slow GTPase activity (Anderson and and Tavitian, 1986; et 1987) and al., and A.Zahraoui, in V.Pizon, N.Touchot, At least three rho have been found in P.Chardin, genes Lacal, 1987). ©IRL Press et al. P.Chardin of mam- now expect, the [32P]ADP-ribosylated product malian cells was not precipitated by antisera against various ko b *g small G proteins, namely Gp (antibody given by John Nor- thup), SEC4 (Goud et al., 1988; antibody given by Peter Nava Novick), YPTI (Segev et al., 1988, antibody given by Segev) or ARF (Kahn et al., 1988, antibody given by Rick Kahn, and also prepared by ourselves). The rhoC protein is susceptible to bacterial proteolysis: several fragments of > 18 500, <21 000 Mr are ADP- in ribosylated and these differ from the starting material charge (pl 5.2-5.5) as well as in size (Figure 2). Proteolysis in was more severe when the protein was synthesized bacterial strain HB101 than in JM109. Such fragmentation in nor was there was not evident the HeLa cell product, evidence that the HeLa cells performed any other kind of which some other processing, such as the fatty acylation by proteins of the ras family are modified. of rhoC 1. in Ecoli and in Fig. [32P]ADP-ribosylation expressed HeLa cells. Incubation conditions are described in Materials and is for the Magnesium required ADP-ribosylation each 108 bacteria or 2 methods. 10 of product, representing x105 ADP-ribosylation of both the cellular and the cloned forms I1 HeLa cells, was analysed by SDS-polyacrylamide gel electrophoresis of rhoC protein was inhibited substantially by EDTA but h with no Lane which is a and autoradiography (5 screen). a, negative much less by EGTA (Figure 1) suggesting that magnesium an extract of that a different control, represents E.coli expressed rab4, the rhoC N.Touchot and A.Zahraoui, in is required. One possible interpretation is that ras-related gene (P.Chardin, V.Pizon, Lanes b-e extract of JM109 transformed with preparation). represent product binds magnesium and that magnesium supports a b: incubated without C3. c: control the ptac-rhoC. (with C3), major for protein conformation which is suitable ADP-ribosylation. has an Mr of 21 000. d: 10 mM EGTA. e: product apparent plus plus known that is involved in the of It is magnesium binding control. HeLa 10 EDTA. f: HeLa cell mM 10 mM extract, g: plus Guanine 10 mM EDTA. guanyl nucleotides to the related protein p2lras. EGTA. h: HeLa plus rhoC denaturation nucleotides stabilize the protein against loss of substrate at for and prevent ADP-ribosylatable 45°C et but we have not encountered example (Rubin al., 1988) condition in which nucleotides affect the ADP- any guanine ribosylation rate per se. Action of C3 upon Vero cells In cell lines tested previously, C3 altered the shapes of intact cells only if it was inserted into the cells by osmotic lysis of or pinosomes by microinjection (Rubin et al., 1988). We have since found, however, that Vero cells respond directly to C3 added to the culture medium. As was found for NIH3T3 and PC12 cells, the effect on Vero cells is dose- and dependent non-lethal. At the dose used in Figure 3 (5 the Vero cells start changing shape within 30 min Itg/ml) of the addition of C3 and gradually contract over the next 12-24 h. The cell bodies condense except for beaded processes which span the area previously covered by the flat cell (Figure Figure 3, panels b and c, show a field in 3f). which the cells are contracted to various degrees (compare to panel a): it is clear that cells in various stages of con- traction still contain microtubules (Figure 3b) which are re- organized only as a consequence of the shape change. However, actinomorphic cells can no longer be stained by the fluorescent reagent NBD-phallacidin which is specific for filamentous actin (Figure 3e). The result implies that the Fig. 2. Two-dimensional polyacrylamide gel analysis of the [32p]_ of ADP-ribosylated products HeLa cell extract (above) and E.coli cells no longer contain actin microfilaments and that the JM109 ptac-rhoC extract (below) as in Figure 1. The two radioactive disassembly of microfilaments accounts for the shape HeLa cell spots align precisely with the bacterial spots indicated by changes. Cells regain their normal morphology and growth slender arrows and have apparent molecular weights of 23 000 and rate a few days after the removal of C3 from the bathing 21 000. The broad arrow heads indicate the horizontal migration of medium. (left) fl-lactoglobulin (pl 5.1) and (right) human carbonic anhydrase Vero cells treated with C3 were lysed and portions of the (pI 6.5). lysate were [32P]ADP-ribosylated. The exposure to C3 preparation). We show this for the ptac-rab4 clone in substantially reduced the amount of rho available for ADP- Figure 1, lane a. Gp, a ras-related protein purified from ribosylation by C3 in vitro (Table I). The available actin, placenta (Evans et al., 1986), was not a substrate. As we in contrast, was changed little-sometimes slightly increased 1088 ADP-ribosylation of rhoC protein by exoenzyme C3 9- 4$ /0 9 U Fig. 3. C3 disrupts actin microfilaments but not microtubules. Vero cells treated with 5 ug/ml purified exoenzyme C3 for 15 h were stained for tubulin and filamentous actin as described in Materials and methods. (a) Control cells stained for tubulin: fluorescence mnicroscopy. (b) C3-treated cells stained with NBD phallacidin for actin filaments: cells stained for tubulin: fluorescence. (c) Same cells as b: phase contrast. (d) Control NBD cells as e: phase contrast. fluorescence. (e) C3-treated cells stained with phallacidin: fluorescence. (f) Same or in slightly decreased. C2-treated of the rhoC sometimes (as Table I) p21 .bot, is the product gene. Interestingly, little or no actin extract contains a C3 substrate of cells which change shape have available Saccharomyces cerevisiae and 1986). size for the rho] 209 (Ohishi Tsuyami, 23 500 Mr, the expected protein, which has about a 70% similarity to the residues, sequence Discussion also has a C3 mammalian rho Dictyostelium proteins. that a C3 in substrate of - 21 kd. These lead to the expectation We have shown major target protein many findings in a which we have named that the rho be ADP-ribosylated highly mammalian provisionally proteins may cells, 1089 P.Chardin et al. responsible for the effects seen. botulinum Apparently C. Table I. Reduction in the amount of rho available for ADP- ribosylation in vitro secretes two transferases ADP-ribosyl (exoenzyme C3 and C2 which microfilament component I) promote disassembly Treatment of intact cells in different It should be noted that actin is also ADP- ways. Control C3 several ribosylated by clostridial enzymes that are distinct ADP-ribose incorporated in vitro, pmol/mg protein from C2 toxin. Such enzymes include the iota toxin of C.perfringens type (Vandekerkhove et al., 1987), Actin 59 54 C.spiroforme toxin and and an ADP- Rho (Popoff Boquet, 1988) 1.9 0.6 ribosyl transferase from a strain of (Popoff et C.difficile al., Vero cell monolayers 0 or were exposed to 5 C3 for 24 h, then 1988). Furthermore, toxin B of causes C.difficile actin jtg/ml washed in fresh medium. Cells were scraped from the plates, washed filaments to break down at by yet another, present unknown, in mM 130 NaCI, 10 mM Hepes, pH 7.4, and collected centri- by mechanism (Wedel et al., 1983; Mitchell et al., 1987). fugation. Portions of the cells were lysed by detergent and ADP- Clearly microfilaments are a common for target clostridia. ribosylated with C3 (for rho) or with transferase for actin at C.difficile for 37°C 30 min. The enzyme actin Since C3 does not appear to affect actin C.difficile ADP-ribosylates directly, the directly with the same as C2 component I et specificity (Popoff al., breakdown of filaments is probably a of consequence ADP- 1988). Incubation conditions were: tl 7.5 x 104 volume, cells, ribosylation of rho protein. It is likely, therefore, that rho mM 10 Tris, pH 8.0, 0.07% Triton X-100, either C3 or (40 tLgImn) is directly or involved with microfilament indirectly assembly transferase (50 and 5 000 ,tg/ml) ,tM [32P]NAD (26 C.difficile or -7 disassembly. Microfilaments can be disassembled c.p.m./pmol). The cells also provided unlabeled NAD. These ytM by conditions were expected to be saturating (Rubin et and this al., 1988) increasing intracellular calcium or by decreasing intracellular was confirmed by finding the same degree of after ADP-ribosylation phosphatidylinositol-4,5-bis phosphate (PIP2) (Janmey and min as after 30 min of incubation. The ADP-ribosylated proteins Stossel, 1987). However, PC12 cells had normal calcium were identified by gel electrophoresis and and were autoradiography levels and experienced a normal increase in intracellular quantitated by counting appropriate portions of the gels. calcium in response to even bradykinin after all of their conserved region and therefore that the very similar rhoA available rho had been ADP-ribosylated C3 by (not shown). rhoB and proteins are likely to be substrates as well as rhoC. This argues against a role for rho in the PIP2-IP3 pathway. Indeed two-dimensional gel patterns of the ADP-ribosylation Two actin-associated proteins of the approximate size of rho of products certain tissues (brain for example) do contain were described recently (Shapland et al., 1988) but they of the size and for rhoA spots charge expected and rhoB in appear to have a considerably more basic charge than the addition to rhoC. Thus it is that C3 possible modifies rho C3 substrate. In fact, we have estimated the amount of ADP- proteins in and we will now refer general to the substrate ribosylatable rho protein in tissues at 0.01 % of the total as rho. generically protein (Rubin et al., 1988) and if this amount truly reflects most information relevant to the function of Hitherto, rho the total amount of this protein in the cells, rho is much less concerned the S. cerevisiae rho] gene product. Disruption abundant than other actin-binding proteins and may not be of rho] results in loss of viability while a dominant mutation detected easily in actin association. Alternatively, it may act in rho] renders cells unable to sporulate (Madaule et al., through an intermediate such as a severing protein. 1987). Our identification of the C3 substrate provides further In addition to the actinomorphic changes, C3 has other insights into the role of rho. C3 biological delivered to the effects on cells and we may consider whether these depend cytosol by osmotic of NIH lysis pinosomes caused 3T3 cells on the disassembly of microfilaments. Certainly a micro- to assume an actinomorphic morphology (Rubin et al., filament defect which blocked cytokinesis could adequately 1988). We describe here a similar response of Vero cells, explain the high frequency of binucleate cells seen in cultures but in this case the cells to C3 in the respond culture medium treated with C3 (Rubin et al., 1988). Microfilament changes without any special procedure to allow it to enter the cells. also be might involved in the migration of germinal vesicles It presumably enters the unaided at a slow cytosol rate. Its which is the principal effect of C3 when injected into effect inside the cells is evident both as a reduction in the Xenopus oocytes (Rubin et al., 1988). It is less obvious how amount of rho protein available for subsequent [32p]_ microfilament changes might relate to the C3 response of ADP-ribosylation and as a collapse of the microfilament PC 12 cells. Shortly after C3 administration, PC 12 cells network. sprout small processes that resemble the initial neurites The C3 was secreted by bacteria which also produce C2, formed in the presence of nerve growth factor (Rubin et al., a binary toxin consisting of component II required for cell Of 1989). course it is possible that the individual rho and I entry component which efficiently ADP-ribosylates proteins, and certainly the group of rho proteins, may have actin. Together cause they microfilament breakdown in intact several effects of which cytoskeletal control is only one. Thus cells. Thus it was important to show that the microfilament the of phenotypes yeast underexpressing rho] or producing disassembly shown in Figure 3 was truly caused by C3 and a hyperactive form of rho] protein overlap, but only not C2. by There are three relevant observations. Firstly, partially, the phenotype observed with actin mutants (Novick the C3 used is electrophoretically pure and we can detect and Botstein, 1985). Pleiotropic effects have been described no trace of actin ADP-ribosyl transferase activity. Secondly, for YPT1 (Schmitt et al., 1986, 1988; Segev and Botstein, whereas C2-II is activated by trypsin treatment, the effect 1987) and are well known for p21r,,. of C3 on Vero cells does not depend on, nor is even Some confusion has arisen in the literature because C3 increased by, prior trypsinization of the enzyme preparation. is a minor contaminant of certain preparations of the C and Nor is it potentiated by native or trypsin-treated C2-II. D botulinal neurotoxins and mixtures of C3 with a neurotoxin Thirdly, affected cells have experienced a reduction in both catalyse ADP-ribosylation and block neurosecretion available rho but have suffered no and protein (Ohashi Kurihara and significant change Narumiya, 1987; 1987; Kubo, in available actin. Thus we conclude that C3 is Adam-Vizi and Kikuchi et truly Knight, 1987; Pure al., 1988). 1090 of rhoC ADP-ribosylation protein by exoenzyme C3 NcoI C 1 or D neurotoxins do not, ADP- CCGGAGCGAACCCCAACCATGGCTGCA... RhoC cDNA however, catalyse sequence MetAlaAla ribosylation (Rosener et al., 1987; Rubin et al., 1988) and CATGGCTGCA ... RhoC, NcoI filled-in cut, ... CAGTGAATT M13 EcoRI mpll, cut, filled-in pure C3 does not affect secretion et Adam- (Rubin al., 1989; ... CAGTGAATTCATGGCTGCA ... Resulting sequence after ligation Vizi et al., 1988). ... CAGTGAATTCTATGGCTGCA.. Introduction of a "T" before the initiation EcoRI MetAlaAla codon M13 in by vitro mutagenesis. C3 is the latest addition to the list of bacterial ADP-ribosyl transferases which have GTP as binding proteins substrates. Pst pOly A 100b.p. Nco Other examples include diphtheria cholera toxin toxin, and rhoC pertussis toxin which have been particularly useful reagents for investigating the biology of elongation factor 2, Gs and Fig. 4. Principal in the stages construction of ptac-rhoC, involving the Gi, respectively. It may be anticipated likewise that C3 will insertion of a single T residue before the initiating AUG and the continue to be of in value the elucidation of the role of rho creation of an EcoRI site. Details are in given Materials and methods. Below, is a sketch of the human rlhoC DNA. L indicates the proteins. EcoRI linkers at each end of the cDNA. The heavy line indicates coding sequence. A classical AATAAA polyadenylation signal sequence is located 13 before the tail. bp poly(A) Materials and methods For two-dimensional gel analyses (O'Farrell. as modified by 1975, Ames Construction of ptac-rhoC and Nikaido. 1976) the isoelectric focusing gels contained pH 3.5-10 Human cDNA encoding rhoC, homologous to human rhoA (Yeramian et al.. ampholines (LKB) and the second dimension gels contained 12.5% 1987) and rlho and Aplvsia (Madaule Axel, 1985) was described by Chardin acrylamide. cl ail. (I1988). The cloned cDNA is an EcoRI fragment containing -20 bp of 5'-non-coding region, 579 bp of - coding sequence and 600 bp of 3'-non- Vero cells coding sequence that includes a Pstl - site 100 bp before the poly(A) site. Cells were grown as monolayers in Dulbecco's-modified Eagle's medium The initiation codon for rhoC is of a NcoI part site (see Figure 4) which (DMEM) plus 10% new-born calf serum. Cells to be stained were grown was convenient for subcloning into the bacterial expression vector ptac (de on coverslips. After exposure to C3 they were fixed for 20 min with formalin Boer et ail., 1983) which has been used to successfully express high levels (3.7%) in phosphate-buffered saline (PBS). They were permeabilized with 1cHra, of (Tucker ei al., 1986). RhtoC cDNA was cut with NcoI and 5' p2 0.1I% Triton X-100 and covered with 100 of either diluted mouse ends were filled the Klenow of DNA using fragment polymerase I. The Al monoclonal anti-tubulin (Amersham) or N-(7-nitrobenz-2-oxa- 1,3-di- 3' end was then trimmed with PstI. The resultant fragment (- I kb) was azo-4-yl)phallacidin (NBD phallacidin: Molecular Probes, Eugene. OR) and subcloned into double-stranded M 13 11 that had been mp similarly treated incubated 20 min at room temperature. After four washes of 5 min each with EcoRI. Klenow fragment and PstI. Single-stranded M13 phage DNA with PBS containing gelatin, the cells exposed to 0.2% anti-tubulin were was prepared, its sequence was checked, and then hybridized with the oligo- incubated for 20 min at room temperature with rhodamine-labeled anti-mouse nucleotide 5'-CAGTGAATTCTATGGCTGCA-3' which was designed to IgG (Sigma). All cells were washed as before, mounted in and glycerol, introduce a T residue immediately before the initiating ATG in order to photographed using a fluorescence microscope. increase expression. Plaques of mutated M 13 were screened with kinase- labeled mutagenesis oligonucleotide under stringent conditions (hybridization in 5 x Denhardt's solution. 5 x 0.1% SDS at SSPE, 53°C. final wash Acknowledgements in I x SSC. 0.1% SDS at for 15 The of 55°C min). sequence probe-positive M 13 phages were checked and double-stranded phages were prepared. An We thank the following colleagues. A.Wittinghofer for a gift of the ptac- EcoRI - PstI insert possessing the expected sequence GAATTCTATG was cHras vector and for suggesting that expression may be increased by placing then subcloned into ptac-ral partially digested by EcoRI and completely by a T residue before the initiator codon. N.Touchot and A.Zahraoui for various PstI (the large fragment retaining the tac promoter was recovered by electro- ptac-rcab expression vectors, Veronique Pizon for a ptac-raplA vector, Simon The elution). ptac-ral expression vector includes a 3' PstI site. It was derived Dillon for generous help and Kristine Danowski for technical assistance. from ptac-c-Hras (Tucker et al.. 1986) and originally from ptac (de Boer Supported by NIH grant Al 16928 to DMG, by INSERM grants CRE eit al., 1983). 852004 to PB and CRE 871009 to PM. Expression of rhoC protein E.coli strain JM 109 (Yanisch-Perron et was al., 1985) transformed with References ptac-rhoC. An culture was induced 0. 1 mM overnight with IPTG. After 10 h of incubation, the cells were washed and with Adam-Vizi.V. and Knight,D.E. (1987) J. Phlsiol., lysed lysozyme exactly 394, 96. as described by Tucker et al. Cell debris Adam-Vizi,V.. Rosener,S., Aktories,K. and (1986). was removed by Knight,D.E. (1988) FEBS Lett, 238. 277-280. centrifugation. Aktories,K., Rosener,S., Blaschke,U. and Chhatwal.G.S. (1988) Eur. J. BiochemZ., 172, 445 -450. Exoenzyme C3 Ames.G.F.-L. and Nikaido,K. (1976) Biochemistrv, 15. 616-623. The enzyme was purified from the culture of strain supematant C.botrlooimiz0s Anderson,P. and Lacal,J. (1987) Mol. Cell. Biol., 7. 3620-3628. 1 873-D as described Rubin et al. The consists of a by (1988). product single Chardin,P. and Tavitian.A. (1986) EMBO J., 5. 2203-2208. electrophoretic 26 and exhibits no detectable actin ADP- species (Mr 000) and Chardin,P., Madaule,P. Tavitian,A. (1988) Nucleic Acids Res., 16. ribosyltransferase activity. Cleveland.D.W.. Fischer,S.G.. Kirschner,M.W. and Laemmli,U.K. Preparation of cell and tissue extracts (1977) J. Biol. 1102- 1106. Clhemti., 252. Pelleted HeLa cells were and then incubated at lysed by freezing thawing and de Boer.H.A.. Comstock.L.J. Vasser,M. (1983) Proc. Naitl. Acad. Sci. for 5 min to allow some of NAD. The 37°C hydrolysis endogenous lysate 21-25. USA, 80. was extracted with one cell volume of 50 mM packed Hepes buffer, pH 8. and Evans,T., Brown,M.L., Fraser,E.D. Northup,J.K. (1986) J. Biol. and a soluble fraction was of mouse prepared by centrifugation. Homogenates Chle,oi., 261, 7052-7059. M mM and calf were in five vols of 0.32 0. 1 organs prepared sucrose, Walworth,N.C. and Novick,P.J. Cell, 100 000 Goud.B., Salminen.A., (1988) 53, fluoride and were at phenylmethylsulphonyl (PMSF). centrifuged g 753 - 768. for I h to fractions. soluble prepare Janmey,P.A. and Stossel,T.P. (1987) Nature, 325. 362-364. Kahn,R.A., Goddard.C. and Newkirk,M. J. Biol. Chem., 263, (1988) and [32P]ADP-ribosylation gel analysis -8287. or tissue were diluted with 10 Soluble bacterial mammalian I 1l pA, samples, and Takai,Y. (1988) J. Biol. Kikuchi,A., Yamamoto,K.. Fujita,T. Chem.. 2.5 pM 20 000 of 50 mM buffer 8.0, (- Hepes pH [3-P]NAD d.p.m./ -16308. 263, 16303 3 of C3 and incubated for 30 at For extracts and min pmol) pg/ml 37°C. and FEBS 216. Kurihara,K. Kubo,S. (1987) 295-299. Lett., of mammalian cells. 10 mM thymidine was also included to suppress 227, Laemmli,U.K. (1970) Ncature, 680-685. poly(ADP-ribose) formation. One-dimensional gel electrophoresis (using and Cell, 31-40. Madaule,P. Axel,R. (1985) 41, 12.5% polyacrylamide) was performed as described by Laemmli (1970). 1091 P.Chardin et al. Axel,R. and Myers,A.M. (1987) Proc. Natl. Acad. Sci. USA, Madaule,P., 84, 779-783. Mitchell,M.J., Laughon,B.E. and Lin,S. (1987) Infect. Immun., 55, 1610-1615. 405-416. Novick,P. and Botstein,D. (1985) Cell, 40, O'Farrell,P.H. (1975) J. Biol. Chem., 250, 4007-4021. Ohashi,Y.T. and Narumiya,S. (1987) J. Biol. Chem., 262, 1430-1433. Biochem. Ohishi,I. and Tsuyama,S. (1986) Biophvs. Res. Commun., 136, -806. Res. Popoff,M.R. and Boquet,P. (1988) Biochem. Biophvs. Commun., 12, 1361- 1368. and Popoff,M.R., Rubin,E.R., Gill,D.M. Boquet,P. (1988) Infect. Immun., 56, 2299-2306. and FEBS Rosener,S., Chhatwal,G.S. Aktories,K. (1987) Lett., 224, 38-42. and Mol. Cell. Rubin,E.J., Gill,D.M., Boquet,P. Popoff,M.R. (1988) Biol., 8, 418-426. and In Rubin,E.J., Boquet,P., Popoff,M.R. Gill,D.M. (1989) Jacobson, ADP-Ribose M.K. and (eds), Transfer Reactions: Jacobson,E.L. Mechanisms and Biological Significance. Springer-Verlag, New York, pp. 422-429. Schmitt,H.D., Wagner,P., Pfaff,E. and Gallwitz,D. (1986) Cell, 47, 401-412. Schmitt,H.D., and Gallwitz,D. (1988) Cell, 53, 635-647. Puzicha,M. Segev,N. and Botstein,D. (1987) Mol. Cell. Biol., 7, 2367-2377. Segev,N., Mulholland,J. and Botstein,D. (1988) Cell, 52, 915-924. Shapland,C., Lowings,P. and Lawson,D. (1988) J. Cell Biol., 107, 153-161. Touchot,N., Chardin,P. and Tavitian,A. (1987) Proc. Natl. Acad. Sci. USA, 84, 8210-8214. Tucker,J., Sczakiel,G., Feuerstein,J., John,J., and Witting- Goody,R.S. (1986) hofer,A. EMBO J., 5, 1351-1358. Vandekerckhove,J., Schering,B., Barmann,M. and Aktories,K. (1987) FEBS Lett., 225, 48-52. Wedel,N., Toselli,P., Pothoulakis,C., Faris,B., Oliver,P., Franzblau,C. and Lamont,T. (1983) Exp. Cell Res., 148, 413-422. Yanisch-Perron,C., Vieira,J. and Messing,J. (1985) Gene, 33, 103- 119. Yeramian,P., Chardin,P., Madaule,P. and Tavitian,A. (1987) Nucleic Acids Res., 15, 1869. Received on November 11, revised on January 12, 1989 1988; http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The EMBO Journal Springer Journals

The mammalian G protein rhoC is ADP‐ribosylated by Clostridium botulinum exoenzyme C3 and affects actin microfilaments in Vero cells.

Download PDF
6 pages

Loading next page...
 
/lp/springer-journals/the-mammalian-g-protein-rhoc-is-adp-ribosylated-by-clostridium-YaXZOvpGVX

References (0)

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 1989
ISSN
0261-4189
eISSN
1460-2075
DOI
10.1002/j.1460-2075.1989.tb03477.x
Publisher site
See Article on Publisher Site

Abstract

Journal vol.8 no.4 pp. 1 087 - 1092, 1989 The EMBO The mammalian G protein rhoC is ADP-ribosylated by Clostridium botulinum exoenzyme C3 and affects actin microfilaments in Vero cells named rhoA (formerly clone 1 or 12) (Yeramian humans, P.Chardin, P.Boquet', P.Madaule2, et al., 1987), rhoB (clone 6) and rhoC (clone 9) (Chardin M.R.Popoff3, E.J.Rubin4 and D.M.Gill4 Hitherto their functions have not been well et al., 1988). Institut National de la Sante et de la Recherche Medicale U-248, but now some information can be inferred from the defined Facult6 de Medecine Lariboisiere Saint-Louis, 10, Avenue de Verdun, of ADP-ribosylation. C3 treatment of Vero cells effects 75010 Paris, 'Unit6 des Antigenes Bacteriens, Institut Pasteur, 28 rue results in the disappearance of actin filaments which suggests du Dr Roux, 75724 Paris, 2Laboratoire de Physiologie Nerveuse, that the rho product may control microfilament assembly. CNRS, 91198 Gif-sur-Yvette, 3Unite des Anaerobies, Institut Pasteur, The change in actin status presumably accounts for some 25 rue de Dr Roux, 75015 Paris, France and 4Department of Molecular Biology and Microbiology, Tufts University, 136 Harrison morphological consequences of C3 action. Avenue, Boston, MA 021 1 1, USA Communicated by S.Olsnes Results botulinum C3 is a recently discovered exo- Clostridium ADP-ribosylation of rhoC protein that ADP-ribosylates a eukaryotic GTP-binding enzyme protein of the ras superfamily. We show now that the The human rhoC cDNA was subcloned in the bacterial ptac (de Boer et al., 1983) and the rhoC bacterially-expressed product of the human rhoC gene expression vector was expressed in Escherichia coli strain JM 109 upon is ADP-ribosylated by C3 and corresponds in size, charge protein of isopropyl-fl-D-thiogalactoside (IPTG), as and behavior to the dominant C3 substrate of eukaryotic the addition in Materials and methods. There was too little cells. C3 treatment of Vero cells results in the disappear- described to detect directly by gel staining but the presump- ance of microfilaments and in actinomorphic shape product rhoC was detected by [32P]GTP binding after changes without any apparent direct effect upon actin. tive protein transfer of proteins to nitrocellulose. The rho Thus the ADP-ribosylation of a rho protein seems to be electrophoretic microfilament disassembly and we infer was readily ADP-ribosylated when a bacterial extract responsible for protein form of a rho protein may be incubated with [32P]NAD and C3 (Figure 1). At an that the unmodified was in control. elevated NAD concentration (100 the amount of ADP- involved cytoskeletal IM) words: actin/ADP-ribosylation/C3/Clostridium botu- ribosylated rhoC protein reached a stable plateau with time Key of incubation or C3 concentration. The limit was typically linumIG protein - 0.2 (4 ng/yl) of bacterial extract. Two orders of pmol/lil rhoC protein was available for reaction in magnitude less The ADP-ribosylated product had the same uninduced cells. Introduction mobility as the main C3 product of human electrophoretic molecular mass of both was -21 000 Bacterial enzymes that ADP-ribosylate proteins have proven cells; the apparent Two-dimensional gel analysis showed that the invaluable for studying the functions of their eukaryotic (Figure 1). and human cell products also had the same charge particularly of G proteins. Exoenzyme C3 is a newly bacterial targets, A less intense spot of the same charge (pl 5.6) but described ADP-ribosyl transferase secreted by C and D (pl 5.6). size (Mr 23 000) is present in both cases strains of Clostridium botulinum but distinct from the of a larger apparent 2) and provides additional evidence of identity. It classical C1 and D neurotoxins. In many eukaryotic cells, (Figure substrate which is found mostly in runs closer to the anticipated position of ADP-ribosylated C3 has a single protein - the same charge the two forms fraction et al., 1988; Rubin et al., rhoC (Mr 22 500). Having the soluble (Aktories and may simply represent of this substrate, - 21 000 Mr, and its inter- cannot be related by proteolysis 1988). The size isomers. Partial proteolysis by the method us to suggest that it is a member of conformational action with GTP led et al. generated the same series of of proteins (Rubin et al., 1988). of Cleveland (1977) the ras superfamily from each of the apparent size variants causes maturational changes radioactive peptides C3-catalysed ADP-ribosylation and distinct morphological and physio- (not shown). in Xenopus oocytes mouse and calf tissues, including thymus, in cultured cells (Rubin et al., 1988) which Homogenized logical changes and patterns similar heart, kidney spleen, gave those caused by microinjected activated ras muscle, lung, partially overlap cells in 2. to that of the HeLa Figure The C3 substrate is, however, distinct from p21rI protein. in its choice of protein substrate. C3 is selective We show now that the bacterially-expressed rhoC highly itself. substrate was detected in the parental E. coli No C3 of human origin is ADP-ribosylated by C3 and specific gene product or HB11. there was no ADP- substrate. The rho genes strains JM109 Likewise, with the eukaryotic coincides substrate in extracts of E. coli cells that and evolutionarily conserved branch of ribosylatable a distinct represent of the ras-related of other members family proteins a 30% sequence identity to the the ras showing expressed superfamily rab4 and ral, rab2, rabS) (Chardin and Axel, 1985; Madaule (c-H-ras, raplA, rap2, classical ras proteins (Madaule Touchot et by al., 1987, expression a slow GTPase activity (Anderson and and Tavitian, 1986; et 1987) and al., and A.Zahraoui, in V.Pizon, N.Touchot, At least three rho have been found in P.Chardin, genes Lacal, 1987). ©IRL Press et al. P.Chardin of mam- now expect, the [32P]ADP-ribosylated product malian cells was not precipitated by antisera against various ko b *g small G proteins, namely Gp (antibody given by John Nor- thup), SEC4 (Goud et al., 1988; antibody given by Peter Nava Novick), YPTI (Segev et al., 1988, antibody given by Segev) or ARF (Kahn et al., 1988, antibody given by Rick Kahn, and also prepared by ourselves). The rhoC protein is susceptible to bacterial proteolysis: several fragments of > 18 500, <21 000 Mr are ADP- in ribosylated and these differ from the starting material charge (pl 5.2-5.5) as well as in size (Figure 2). Proteolysis in was more severe when the protein was synthesized bacterial strain HB101 than in JM109. Such fragmentation in nor was there was not evident the HeLa cell product, evidence that the HeLa cells performed any other kind of which some other processing, such as the fatty acylation by proteins of the ras family are modified. of rhoC 1. in Ecoli and in Fig. [32P]ADP-ribosylation expressed HeLa cells. Incubation conditions are described in Materials and is for the Magnesium required ADP-ribosylation each 108 bacteria or 2 methods. 10 of product, representing x105 ADP-ribosylation of both the cellular and the cloned forms I1 HeLa cells, was analysed by SDS-polyacrylamide gel electrophoresis of rhoC protein was inhibited substantially by EDTA but h with no Lane which is a and autoradiography (5 screen). a, negative much less by EGTA (Figure 1) suggesting that magnesium an extract of that a different control, represents E.coli expressed rab4, the rhoC N.Touchot and A.Zahraoui, in is required. One possible interpretation is that ras-related gene (P.Chardin, V.Pizon, Lanes b-e extract of JM109 transformed with preparation). represent product binds magnesium and that magnesium supports a b: incubated without C3. c: control the ptac-rhoC. (with C3), major for protein conformation which is suitable ADP-ribosylation. has an Mr of 21 000. d: 10 mM EGTA. e: product apparent plus plus known that is involved in the of It is magnesium binding control. HeLa 10 EDTA. f: HeLa cell mM 10 mM extract, g: plus Guanine 10 mM EDTA. guanyl nucleotides to the related protein p2lras. EGTA. h: HeLa plus rhoC denaturation nucleotides stabilize the protein against loss of substrate at for and prevent ADP-ribosylatable 45°C et but we have not encountered example (Rubin al., 1988) condition in which nucleotides affect the ADP- any guanine ribosylation rate per se. Action of C3 upon Vero cells In cell lines tested previously, C3 altered the shapes of intact cells only if it was inserted into the cells by osmotic lysis of or pinosomes by microinjection (Rubin et al., 1988). We have since found, however, that Vero cells respond directly to C3 added to the culture medium. As was found for NIH3T3 and PC12 cells, the effect on Vero cells is dose- and dependent non-lethal. At the dose used in Figure 3 (5 the Vero cells start changing shape within 30 min Itg/ml) of the addition of C3 and gradually contract over the next 12-24 h. The cell bodies condense except for beaded processes which span the area previously covered by the flat cell (Figure Figure 3, panels b and c, show a field in 3f). which the cells are contracted to various degrees (compare to panel a): it is clear that cells in various stages of con- traction still contain microtubules (Figure 3b) which are re- organized only as a consequence of the shape change. However, actinomorphic cells can no longer be stained by the fluorescent reagent NBD-phallacidin which is specific for filamentous actin (Figure 3e). The result implies that the Fig. 2. Two-dimensional polyacrylamide gel analysis of the [32p]_ of ADP-ribosylated products HeLa cell extract (above) and E.coli cells no longer contain actin microfilaments and that the JM109 ptac-rhoC extract (below) as in Figure 1. The two radioactive disassembly of microfilaments accounts for the shape HeLa cell spots align precisely with the bacterial spots indicated by changes. Cells regain their normal morphology and growth slender arrows and have apparent molecular weights of 23 000 and rate a few days after the removal of C3 from the bathing 21 000. The broad arrow heads indicate the horizontal migration of medium. (left) fl-lactoglobulin (pl 5.1) and (right) human carbonic anhydrase Vero cells treated with C3 were lysed and portions of the (pI 6.5). lysate were [32P]ADP-ribosylated. The exposure to C3 preparation). We show this for the ptac-rab4 clone in substantially reduced the amount of rho available for ADP- Figure 1, lane a. Gp, a ras-related protein purified from ribosylation by C3 in vitro (Table I). The available actin, placenta (Evans et al., 1986), was not a substrate. As we in contrast, was changed little-sometimes slightly increased 1088 ADP-ribosylation of rhoC protein by exoenzyme C3 9- 4$ /0 9 U Fig. 3. C3 disrupts actin microfilaments but not microtubules. Vero cells treated with 5 ug/ml purified exoenzyme C3 for 15 h were stained for tubulin and filamentous actin as described in Materials and methods. (a) Control cells stained for tubulin: fluorescence mnicroscopy. (b) C3-treated cells stained with NBD phallacidin for actin filaments: cells stained for tubulin: fluorescence. (c) Same cells as b: phase contrast. (d) Control NBD cells as e: phase contrast. fluorescence. (e) C3-treated cells stained with phallacidin: fluorescence. (f) Same or in slightly decreased. C2-treated of the rhoC sometimes (as Table I) p21 .bot, is the product gene. Interestingly, little or no actin extract contains a C3 substrate of cells which change shape have available Saccharomyces cerevisiae and 1986). size for the rho] 209 (Ohishi Tsuyami, 23 500 Mr, the expected protein, which has about a 70% similarity to the residues, sequence Discussion also has a C3 mammalian rho Dictyostelium proteins. that a C3 in substrate of - 21 kd. These lead to the expectation We have shown major target protein many findings in a which we have named that the rho be ADP-ribosylated highly mammalian provisionally proteins may cells, 1089 P.Chardin et al. responsible for the effects seen. botulinum Apparently C. Table I. Reduction in the amount of rho available for ADP- ribosylation in vitro secretes two transferases ADP-ribosyl (exoenzyme C3 and C2 which microfilament component I) promote disassembly Treatment of intact cells in different It should be noted that actin is also ADP- ways. Control C3 several ribosylated by clostridial enzymes that are distinct ADP-ribose incorporated in vitro, pmol/mg protein from C2 toxin. Such enzymes include the iota toxin of C.perfringens type (Vandekerkhove et al., 1987), Actin 59 54 C.spiroforme toxin and and an ADP- Rho (Popoff Boquet, 1988) 1.9 0.6 ribosyl transferase from a strain of (Popoff et C.difficile al., Vero cell monolayers 0 or were exposed to 5 C3 for 24 h, then 1988). Furthermore, toxin B of causes C.difficile actin jtg/ml washed in fresh medium. Cells were scraped from the plates, washed filaments to break down at by yet another, present unknown, in mM 130 NaCI, 10 mM Hepes, pH 7.4, and collected centri- by mechanism (Wedel et al., 1983; Mitchell et al., 1987). fugation. Portions of the cells were lysed by detergent and ADP- Clearly microfilaments are a common for target clostridia. ribosylated with C3 (for rho) or with transferase for actin at C.difficile for 37°C 30 min. The enzyme actin Since C3 does not appear to affect actin C.difficile ADP-ribosylates directly, the directly with the same as C2 component I et specificity (Popoff al., breakdown of filaments is probably a of consequence ADP- 1988). Incubation conditions were: tl 7.5 x 104 volume, cells, ribosylation of rho protein. It is likely, therefore, that rho mM 10 Tris, pH 8.0, 0.07% Triton X-100, either C3 or (40 tLgImn) is directly or involved with microfilament indirectly assembly transferase (50 and 5 000 ,tg/ml) ,tM [32P]NAD (26 C.difficile or -7 disassembly. Microfilaments can be disassembled c.p.m./pmol). The cells also provided unlabeled NAD. These ytM by conditions were expected to be saturating (Rubin et and this al., 1988) increasing intracellular calcium or by decreasing intracellular was confirmed by finding the same degree of after ADP-ribosylation phosphatidylinositol-4,5-bis phosphate (PIP2) (Janmey and min as after 30 min of incubation. The ADP-ribosylated proteins Stossel, 1987). However, PC12 cells had normal calcium were identified by gel electrophoresis and and were autoradiography levels and experienced a normal increase in intracellular quantitated by counting appropriate portions of the gels. calcium in response to even bradykinin after all of their conserved region and therefore that the very similar rhoA available rho had been ADP-ribosylated C3 by (not shown). rhoB and proteins are likely to be substrates as well as rhoC. This argues against a role for rho in the PIP2-IP3 pathway. Indeed two-dimensional gel patterns of the ADP-ribosylation Two actin-associated proteins of the approximate size of rho of products certain tissues (brain for example) do contain were described recently (Shapland et al., 1988) but they of the size and for rhoA spots charge expected and rhoB in appear to have a considerably more basic charge than the addition to rhoC. Thus it is that C3 possible modifies rho C3 substrate. In fact, we have estimated the amount of ADP- proteins in and we will now refer general to the substrate ribosylatable rho protein in tissues at 0.01 % of the total as rho. generically protein (Rubin et al., 1988) and if this amount truly reflects most information relevant to the function of Hitherto, rho the total amount of this protein in the cells, rho is much less concerned the S. cerevisiae rho] gene product. Disruption abundant than other actin-binding proteins and may not be of rho] results in loss of viability while a dominant mutation detected easily in actin association. Alternatively, it may act in rho] renders cells unable to sporulate (Madaule et al., through an intermediate such as a severing protein. 1987). Our identification of the C3 substrate provides further In addition to the actinomorphic changes, C3 has other insights into the role of rho. C3 biological delivered to the effects on cells and we may consider whether these depend cytosol by osmotic of NIH lysis pinosomes caused 3T3 cells on the disassembly of microfilaments. Certainly a micro- to assume an actinomorphic morphology (Rubin et al., filament defect which blocked cytokinesis could adequately 1988). We describe here a similar response of Vero cells, explain the high frequency of binucleate cells seen in cultures but in this case the cells to C3 in the respond culture medium treated with C3 (Rubin et al., 1988). Microfilament changes without any special procedure to allow it to enter the cells. also be might involved in the migration of germinal vesicles It presumably enters the unaided at a slow cytosol rate. Its which is the principal effect of C3 when injected into effect inside the cells is evident both as a reduction in the Xenopus oocytes (Rubin et al., 1988). It is less obvious how amount of rho protein available for subsequent [32p]_ microfilament changes might relate to the C3 response of ADP-ribosylation and as a collapse of the microfilament PC 12 cells. Shortly after C3 administration, PC 12 cells network. sprout small processes that resemble the initial neurites The C3 was secreted by bacteria which also produce C2, formed in the presence of nerve growth factor (Rubin et al., a binary toxin consisting of component II required for cell Of 1989). course it is possible that the individual rho and I entry component which efficiently ADP-ribosylates proteins, and certainly the group of rho proteins, may have actin. Together cause they microfilament breakdown in intact several effects of which cytoskeletal control is only one. Thus cells. Thus it was important to show that the microfilament the of phenotypes yeast underexpressing rho] or producing disassembly shown in Figure 3 was truly caused by C3 and a hyperactive form of rho] protein overlap, but only not C2. by There are three relevant observations. Firstly, partially, the phenotype observed with actin mutants (Novick the C3 used is electrophoretically pure and we can detect and Botstein, 1985). Pleiotropic effects have been described no trace of actin ADP-ribosyl transferase activity. Secondly, for YPT1 (Schmitt et al., 1986, 1988; Segev and Botstein, whereas C2-II is activated by trypsin treatment, the effect 1987) and are well known for p21r,,. of C3 on Vero cells does not depend on, nor is even Some confusion has arisen in the literature because C3 increased by, prior trypsinization of the enzyme preparation. is a minor contaminant of certain preparations of the C and Nor is it potentiated by native or trypsin-treated C2-II. D botulinal neurotoxins and mixtures of C3 with a neurotoxin Thirdly, affected cells have experienced a reduction in both catalyse ADP-ribosylation and block neurosecretion available rho but have suffered no and protein (Ohashi Kurihara and significant change Narumiya, 1987; 1987; Kubo, in available actin. Thus we conclude that C3 is Adam-Vizi and Kikuchi et truly Knight, 1987; Pure al., 1988). 1090 of rhoC ADP-ribosylation protein by exoenzyme C3 NcoI C 1 or D neurotoxins do not, ADP- CCGGAGCGAACCCCAACCATGGCTGCA... RhoC cDNA however, catalyse sequence MetAlaAla ribosylation (Rosener et al., 1987; Rubin et al., 1988) and CATGGCTGCA ... RhoC, NcoI filled-in cut, ... CAGTGAATT M13 EcoRI mpll, cut, filled-in pure C3 does not affect secretion et Adam- (Rubin al., 1989; ... CAGTGAATTCATGGCTGCA ... Resulting sequence after ligation Vizi et al., 1988). ... CAGTGAATTCTATGGCTGCA.. Introduction of a "T" before the initiation EcoRI MetAlaAla codon M13 in by vitro mutagenesis. C3 is the latest addition to the list of bacterial ADP-ribosyl transferases which have GTP as binding proteins substrates. Pst pOly A 100b.p. Nco Other examples include diphtheria cholera toxin toxin, and rhoC pertussis toxin which have been particularly useful reagents for investigating the biology of elongation factor 2, Gs and Fig. 4. Principal in the stages construction of ptac-rhoC, involving the Gi, respectively. It may be anticipated likewise that C3 will insertion of a single T residue before the initiating AUG and the continue to be of in value the elucidation of the role of rho creation of an EcoRI site. Details are in given Materials and methods. Below, is a sketch of the human rlhoC DNA. L indicates the proteins. EcoRI linkers at each end of the cDNA. The heavy line indicates coding sequence. A classical AATAAA polyadenylation signal sequence is located 13 before the tail. bp poly(A) Materials and methods For two-dimensional gel analyses (O'Farrell. as modified by 1975, Ames Construction of ptac-rhoC and Nikaido. 1976) the isoelectric focusing gels contained pH 3.5-10 Human cDNA encoding rhoC, homologous to human rhoA (Yeramian et al.. ampholines (LKB) and the second dimension gels contained 12.5% 1987) and rlho and Aplvsia (Madaule Axel, 1985) was described by Chardin acrylamide. cl ail. (I1988). The cloned cDNA is an EcoRI fragment containing -20 bp of 5'-non-coding region, 579 bp of - coding sequence and 600 bp of 3'-non- Vero cells coding sequence that includes a Pstl - site 100 bp before the poly(A) site. Cells were grown as monolayers in Dulbecco's-modified Eagle's medium The initiation codon for rhoC is of a NcoI part site (see Figure 4) which (DMEM) plus 10% new-born calf serum. Cells to be stained were grown was convenient for subcloning into the bacterial expression vector ptac (de on coverslips. After exposure to C3 they were fixed for 20 min with formalin Boer et ail., 1983) which has been used to successfully express high levels (3.7%) in phosphate-buffered saline (PBS). They were permeabilized with 1cHra, of (Tucker ei al., 1986). RhtoC cDNA was cut with NcoI and 5' p2 0.1I% Triton X-100 and covered with 100 of either diluted mouse ends were filled the Klenow of DNA using fragment polymerase I. The Al monoclonal anti-tubulin (Amersham) or N-(7-nitrobenz-2-oxa- 1,3-di- 3' end was then trimmed with PstI. The resultant fragment (- I kb) was azo-4-yl)phallacidin (NBD phallacidin: Molecular Probes, Eugene. OR) and subcloned into double-stranded M 13 11 that had been mp similarly treated incubated 20 min at room temperature. After four washes of 5 min each with EcoRI. Klenow fragment and PstI. Single-stranded M13 phage DNA with PBS containing gelatin, the cells exposed to 0.2% anti-tubulin were was prepared, its sequence was checked, and then hybridized with the oligo- incubated for 20 min at room temperature with rhodamine-labeled anti-mouse nucleotide 5'-CAGTGAATTCTATGGCTGCA-3' which was designed to IgG (Sigma). All cells were washed as before, mounted in and glycerol, introduce a T residue immediately before the initiating ATG in order to photographed using a fluorescence microscope. increase expression. Plaques of mutated M 13 were screened with kinase- labeled mutagenesis oligonucleotide under stringent conditions (hybridization in 5 x Denhardt's solution. 5 x 0.1% SDS at SSPE, 53°C. final wash Acknowledgements in I x SSC. 0.1% SDS at for 15 The of 55°C min). sequence probe-positive M 13 phages were checked and double-stranded phages were prepared. An We thank the following colleagues. A.Wittinghofer for a gift of the ptac- EcoRI - PstI insert possessing the expected sequence GAATTCTATG was cHras vector and for suggesting that expression may be increased by placing then subcloned into ptac-ral partially digested by EcoRI and completely by a T residue before the initiator codon. N.Touchot and A.Zahraoui for various PstI (the large fragment retaining the tac promoter was recovered by electro- ptac-rcab expression vectors, Veronique Pizon for a ptac-raplA vector, Simon The elution). ptac-ral expression vector includes a 3' PstI site. It was derived Dillon for generous help and Kristine Danowski for technical assistance. from ptac-c-Hras (Tucker et al.. 1986) and originally from ptac (de Boer Supported by NIH grant Al 16928 to DMG, by INSERM grants CRE eit al., 1983). 852004 to PB and CRE 871009 to PM. Expression of rhoC protein E.coli strain JM 109 (Yanisch-Perron et was al., 1985) transformed with References ptac-rhoC. An culture was induced 0. 1 mM overnight with IPTG. After 10 h of incubation, the cells were washed and with Adam-Vizi.V. and Knight,D.E. (1987) J. Phlsiol., lysed lysozyme exactly 394, 96. as described by Tucker et al. Cell debris Adam-Vizi,V.. Rosener,S., Aktories,K. and (1986). was removed by Knight,D.E. (1988) FEBS Lett, 238. 277-280. centrifugation. Aktories,K., Rosener,S., Blaschke,U. and Chhatwal.G.S. (1988) Eur. J. BiochemZ., 172, 445 -450. Exoenzyme C3 Ames.G.F.-L. and Nikaido,K. (1976) Biochemistrv, 15. 616-623. The enzyme was purified from the culture of strain supematant C.botrlooimiz0s Anderson,P. and Lacal,J. (1987) Mol. Cell. Biol., 7. 3620-3628. 1 873-D as described Rubin et al. The consists of a by (1988). product single Chardin,P. and Tavitian.A. (1986) EMBO J., 5. 2203-2208. electrophoretic 26 and exhibits no detectable actin ADP- species (Mr 000) and Chardin,P., Madaule,P. Tavitian,A. (1988) Nucleic Acids Res., 16. ribosyltransferase activity. Cleveland.D.W.. Fischer,S.G.. Kirschner,M.W. and Laemmli,U.K. Preparation of cell and tissue extracts (1977) J. Biol. 1102- 1106. Clhemti., 252. Pelleted HeLa cells were and then incubated at lysed by freezing thawing and de Boer.H.A.. Comstock.L.J. Vasser,M. (1983) Proc. Naitl. Acad. Sci. for 5 min to allow some of NAD. The 37°C hydrolysis endogenous lysate 21-25. USA, 80. was extracted with one cell volume of 50 mM packed Hepes buffer, pH 8. and Evans,T., Brown,M.L., Fraser,E.D. Northup,J.K. (1986) J. Biol. and a soluble fraction was of mouse prepared by centrifugation. Homogenates Chle,oi., 261, 7052-7059. M mM and calf were in five vols of 0.32 0. 1 organs prepared sucrose, Walworth,N.C. and Novick,P.J. Cell, 100 000 Goud.B., Salminen.A., (1988) 53, fluoride and were at phenylmethylsulphonyl (PMSF). centrifuged g 753 - 768. for I h to fractions. soluble prepare Janmey,P.A. and Stossel,T.P. (1987) Nature, 325. 362-364. Kahn,R.A., Goddard.C. and Newkirk,M. J. Biol. Chem., 263, (1988) and [32P]ADP-ribosylation gel analysis -8287. or tissue were diluted with 10 Soluble bacterial mammalian I 1l pA, samples, and Takai,Y. (1988) J. Biol. Kikuchi,A., Yamamoto,K.. Fujita,T. Chem.. 2.5 pM 20 000 of 50 mM buffer 8.0, (- Hepes pH [3-P]NAD d.p.m./ -16308. 263, 16303 3 of C3 and incubated for 30 at For extracts and min pmol) pg/ml 37°C. and FEBS 216. Kurihara,K. Kubo,S. (1987) 295-299. Lett., of mammalian cells. 10 mM thymidine was also included to suppress 227, Laemmli,U.K. (1970) Ncature, 680-685. poly(ADP-ribose) formation. One-dimensional gel electrophoresis (using and Cell, 31-40. Madaule,P. Axel,R. (1985) 41, 12.5% polyacrylamide) was performed as described by Laemmli (1970). 1091 P.Chardin et al. Axel,R. and Myers,A.M. (1987) Proc. Natl. Acad. Sci. USA, Madaule,P., 84, 779-783. Mitchell,M.J., Laughon,B.E. and Lin,S. (1987) Infect. Immun., 55, 1610-1615. 405-416. Novick,P. and Botstein,D. (1985) Cell, 40, O'Farrell,P.H. (1975) J. Biol. Chem., 250, 4007-4021. Ohashi,Y.T. and Narumiya,S. (1987) J. Biol. Chem., 262, 1430-1433. Biochem. Ohishi,I. and Tsuyama,S. (1986) Biophvs. Res. Commun., 136, -806. Res. Popoff,M.R. and Boquet,P. (1988) Biochem. Biophvs. Commun., 12, 1361- 1368. and Popoff,M.R., Rubin,E.R., Gill,D.M. Boquet,P. (1988) Infect. Immun., 56, 2299-2306. and FEBS Rosener,S., Chhatwal,G.S. Aktories,K. (1987) Lett., 224, 38-42. and Mol. Cell. Rubin,E.J., Gill,D.M., Boquet,P. Popoff,M.R. (1988) Biol., 8, 418-426. and In Rubin,E.J., Boquet,P., Popoff,M.R. Gill,D.M. (1989) Jacobson, ADP-Ribose M.K. and (eds), Transfer Reactions: Jacobson,E.L. Mechanisms and Biological Significance. Springer-Verlag, New York, pp. 422-429. Schmitt,H.D., Wagner,P., Pfaff,E. and Gallwitz,D. (1986) Cell, 47, 401-412. Schmitt,H.D., and Gallwitz,D. (1988) Cell, 53, 635-647. Puzicha,M. Segev,N. and Botstein,D. (1987) Mol. Cell. Biol., 7, 2367-2377. Segev,N., Mulholland,J. and Botstein,D. (1988) Cell, 52, 915-924. Shapland,C., Lowings,P. and Lawson,D. (1988) J. Cell Biol., 107, 153-161. Touchot,N., Chardin,P. and Tavitian,A. (1987) Proc. Natl. Acad. Sci. USA, 84, 8210-8214. Tucker,J., Sczakiel,G., Feuerstein,J., John,J., and Witting- Goody,R.S. (1986) hofer,A. EMBO J., 5, 1351-1358. Vandekerckhove,J., Schering,B., Barmann,M. and Aktories,K. (1987) FEBS Lett., 225, 48-52. Wedel,N., Toselli,P., Pothoulakis,C., Faris,B., Oliver,P., Franzblau,C. and Lamont,T. (1983) Exp. Cell Res., 148, 413-422. Yanisch-Perron,C., Vieira,J. and Messing,J. (1985) Gene, 33, 103- 119. Yeramian,P., Chardin,P., Madaule,P. and Tavitian,A. (1987) Nucleic Acids Res., 15, 1869. Received on November 11, revised on January 12, 1989 1988;

Journal

The EMBO Journal – Springer Journals

Published: Apr 1, 1989

There are no references for this article.