Entrapment of Rho ADP-ribosylated by Clostridium botulinum C3 exoenzyme in the Rho-guanine nucleotide dissociation inhibitor-1 complex.

RhoA, -B, and -C are ADP-ribosylated by Clostridium botulinum exoenzyme C3 to induce redistribution of the actin filaments in intact cells, a finding that has led to the notion that the ADP-ribosylation blocks coupling of Rho to the downstream effectors. ADP-ribosylation, however, does not alter nucleotide binding, intrinsic, and GTPase-activating protein-stimulated GTPase activity. ADP-ribosylated Rho is even capable of activating the effector protein ROK in a recombinant system. Treatment of cells with a cell-permeable chimeric C3 toxin led to complete localization of modified Rho to the cytosolic fraction based on the complexation of ADP-ribosylated Rho with the guanine-nucleotide dissociation inhibitor-1 (GDI-1). The modified complex turned out to be resistant to phosphatidylinositol 4,5-bisphosphate- and GTPgammaS-induced release of Rho from GDI-1. Thus, ADP-ribosylation leads to entrapment of Rho in the GDI-1 complex. The increased stability of the GDI complex prevented binding of Rho to membrane-associated players of the GTPase cycle such as the activating guanine nucleotide exchange factors and effector proteins.

The Rho subfamily of low molecular mass GTPases comprises RhoA/B/C, Rac1/2/3, Cdc42, RhoD, RhoE/Rnd, RhoG, TC10, and TTF. The best characterized of them, Rho, Rac, and Cdc42, have been shown to be master regulators of the actin cytoskeleton. Whereas Rho is involved in the formation of focal adhesions and stress fibers, Rac participates in the control of lamellipodia and Cdc42 in the formation of filopodia. In addition to their function as regulators of the microfilament system, the Rho proteins are involved in the management of several cellular functions such as membrane trafficking, smooth muscle contraction, phospholipid metabolism, cell cycle progression, cell transformation, apoptosis, and transcriptional activation (1,2).
In the GTP-bound form, Rho translocates to plasma membranes where it interacts with effectors to transduce the signal downstream. In the inactive GDP-bound form, Rho is com-plexed by the guanine nucleotide dissociation inhibitor (GDI), 1 thereby localizing it to the cytosolic fraction. Three isoforms of GDI have been described: Rho-GDI-1 that interacts with Rho, Rac, and Cdc42 and the D4-GDI or Ly-GDI that is selectively expressed in hematopoetic cell lines (3)(4)(5). Rho-GDI␥ and the murine form Rho-GDI3 possess a hydrophobic N-terminal elongation to be associated with membranous compartments (6,7). GDI␥ seems to function with RhoG but not with Rac. It is poorly understood how Rho is released from the GDI complex. Phospholipids, especially PIP 2 , are reported to liberate Rho from the GDI complex (8,9). Recently, moesin, a member of the ERM (ezrin, radixin, moesin) family, that binds to CD44 or intercellular adhesion molecules (ICAMs) has been reported to bind GDI thereby releasing monomeric Rho (10 -13).
Because Clostridium botulinum exoenzyme C3 turns off the Rho-dependent signal pathway by selective ADP-ribosylation of RhoA, -B, and -C but not of other low molecular mass GTPases, C3 has been classified as an indispensable tool in cell biology. The disadvantage of poor cell accessibility because of failure of receptor binding and translocation domain has been overcome by different approaches: 1) microinjection of C3 (14); 2) electroporation of cells in the presence of C3 (15); 3) permeabilization by digitonin (16) or streptolysin O (17); 4) intracellular expression of C3 (18 -20); 5) using chimeric toxins that recruit the cell entry machinery of other toxins, C3 is fused to enzyme-deficient diphtheria toxin or C. botulinum C2 toxin (21)(22)(23). The latter possibility, a chimera with the binary C. botulinum C2 toxin, was applied in this study.
In intact cells, ADP-ribosylation of Rho is accompanied by disaggregation of actin filaments leading to the notion that ADP-ribosylation turns Rho inactive. It has been reported that the typical feature of the disrupted actin filament system caused by C3-catalyzed ADP-ribosylation is reversed by the cytotoxic necrotizing factor (CNF) (22,24) that is known to transform Rho constitutively active by deamidation of glutamine 63 (25,26). Because deamidation merely blocks GTP hydrolysis, ADP-ribosylated Rho is able to interact functionally with its effectors. Thus, inhibition of Rho-effector coupling seems not to be the mode of how ADP-ribosylation inactivates Rho functions. Therefore, we studied the influence of ADP-ribosylation on the Rho-GDI interaction and on the subcellular distribution of Rho and report here that ADP-ribosylation prevents membrane binding and traps Rho in the GDI complex thereby blocking Rho cycling between cytosol and membranes.

EXPERIMENTAL PROCEDURES
Materials-[ 32 P]NAD was purchased from PerkinElmer Life Sciences. Phosphatidylinositol 4,5-bisphosphate and GTP␥S were obtained from Roche Molecular Biochemicals. The antibody against RhoA and GDI-1 were purchased from Santa Cruz Biotechnology, Inc.
Recombinant Proteins-RhoA was expressed in Sf9 insect cells using a baculovirus expression system (Pharmingen). Sf9 cells (1 ϫ 10 6 cells/ ml) were cultured at 25°C in TNM-FH insect medium containing 10% fetal calf serum, 100 units/ml penicillin G, and 100 mg/ml streptomycin. Infected with pAcGHLT containing RhoA baculovirus transfer vectors (multiplicity of infection of 5-10) was performed at 25°C for 48 h. Thereafter, the cells were lysed by sonication in buffer A (50 mM NaCl, 10 mM MgCl 2 , 1 mM dithiothreitol, 1% (v/v) Triton X-100, 10 M phenylmethylsulfonyl fluoride, and 10 mM Tris-HCl, pH 7.5) and centrifuged at 20,000 ϫ g for 1 h. GST-RhoA from the supernatant was precipitated with glutathione-Sepharose beads at 4°C for 30 min followed by intensive washing with buffer A to remove unbound proteins. RhoA was released from the parent GST fusion proteins by incubation with thrombin (Pharmingen, 10 units) at 4°C overnight in buffer (150 mM NaCl, 5 mM MgCl 2 , 2.5 mM CaCl 2 , 1 mM dithiothreitol, and 50 mM Tris-HCl, pH 8.0). The beads were removed by centrifugation, and thrombin was precipitated using p-aminobenzamidine beads. The homogeneity of the recombinant RhoA was proven by SDS-PAGE.
Toxin Treatment-NIH3T3 fibroblasts were grown in Dulbecco's medium supplemented with 10% fetal calf serum, 4 mM glutamine, 100 g/ml penicillin, and 100 M streptomycin. Fibroblasts on 10-cm dishes were treated with C2IN-C3 fusion toxin or C2 toxin (C2I plus C2II) (each 1 g/ml) for 3 h. The medium was removed and the cells were washed with 5 ml of ice-cold phosphate-buffered saline and scraped in 300 l/dish lysis buffer (50 mM HEPES, pH 7.5, 100 mM NaCl, 5 mM MgCl 2 , 40 g/ml aprotinin, 0.1 mM phenylmethylsulfonyl fluoride, 20 g/ml leupeptin, 80 g/ml benzamidine, 1 mM dithiothreitol, 1 mM EDTA). The cells were disrupted mechanically by sonication (five times on ice), followed by centrifugation for 10 min at 2000 ϫ g to remove the nuclear fraction. The supernatant was used as cell lysate.
Lysates from NIH3T3 cells were centrifuged at 100,000 ϫ g for 1 h to prepare cytosolic and total particulate fractions. The high speed pellet, which consists of the heavy and the light membrane fractions, was washed and resuspended in lysis buffer (28). Cytosolic and membranous fractions were brought to the same concentration of protein measured by the method of Bradford (51).
ADP-ribosylation-RhoA (50 g/ml) was subjected to ADP-ribosylation by C3 exoenzyme (1 g/ml) and 10 M [ 32 P]NAD for 30 min at 37°C. Nucleotide exchange reaction: isoprenylated RhoA (200 g/ml) dissolved in lysis buffer was incubated in the presence of 100 M GTP␥S or 100 M GDP for 1 h on ice.
Immunoblot Analysis-SDS-polyacrylamide gel electrophoresis was performed on 12.5% acrylamide gels. The gels were analyzed by Phos-phorImager SI from Molecular Dynamics. Proteins were separated on 12.5% acrylamide gels and transferred onto nitrocellulose for 2 h at 250 mA. The membranes were blocked for 1 h with 5% (w/v) nonfat dried milk. Blots were incubated for 2 h with anti-RhoA monoclonal antibody (Santa Cruz), anti-GDI-1 polyclonal antibody (both diluted 1:2.000) and anti-PLD1 (1:500) in phosphate-buffered saline containing 0.05% Tween and then for 45 min with a horseradish peroxidase-conjugated secondary antibody. The proteins were visualized by enhanced chemiluminescence. Western blotting of purified recombinant proteins confirmed that the antibodies were specific (results not shown).
Binding of RhoA to Membranes-Recombinant RhoA from Sf9 cells was ADP-ribosylated as described above followed by pre-binding of GDP or GTP␥S. RhoA (2 g) was incubated with NIH3T3 membranes (50 g) on ice for 30 min. The reaction mixture was layered onto 200 l of sucrose (20% (w/v) sucrose supplemented with 0.1 mg/ml bovine serum albumin and 0.02% (w/v) sodium azide) using microcentrifugation tips and centrifuged for 1 h at 30,000 ϫ g at 4°C. The supernatant was precipitated and resolved in 30 l of Laemmli sample buffer. The pellet was dissolved in the same volume of sample buffer. Supernatants and pellets were analyzed for RhoA content by immunoblot.
Separation of RhoA from the GDI Complex-Cytosol was prepared as described above. 1 mg of protein dissolved in 500 l of separation buffer was loaded onto a Superdex 75 column (Amersham Biosciences) previously equilibrated with separation buffer (10 mM imidazole, pH 6.8, 400 mM NaCl, 250 mM sucrose). The flow rate was 0.2 ml/min, and the fraction size was 500 l. The fractions were precipitated with chloroform/methanol and resolved in 30 l of Laemmli buffer. Fractions were analyzed by immunoblot for RhoA and GDI-1. Calibration of the column was performed with bovine serum albumin (66 kDa, fractions 4 and 5), chicken albumin (45 kDa, fractions 7 and 8), and chymotrypsin (25 kDa, fractions 11 and 12). Cytosols (0.5 mg/ml) prepared in lysis buffer supplemented with 250 mM sucrose were incubated in the presence of 0.2 mg/ml PIP 2 or buffer at 37°C for 10 min and applied to a 30-kDa cut-off membrane (Microcon 30, Amicon, Beverly) at 7000 ϫ g for 30 min at room temperature. Supernatant and filtrate were brought to the same protein concentration followed by immunoblot analysis for RhoA and GDI-1.
Pull-down Experiments with GST-C21-Pull-down experiments were performed as described by Reid et al. (29). Ice-cold lysis buffer (500 l; 50 mM NaCl, 20 mM Tris-HCl, pH 7.4, 3 mM MgCl 2 , 1% Nonidet P-40, 0.25% Triton X-100, 10 mM dithiothreitol, 100 M phenylmethylsulfonyl fluoride) was added to cell lysates containing 1 mg of protein in total. For recombinant assay, 1 g of GDP-or GTP␥S-loaded RhoA was added to 500 l of lysis buffer. Samples were centrifuged for 10 min at 15,000 ϫ g, and the supernatant was used for pull-down assay. To this end, 20 l of beads slurry of the Rho binding domain GST-C21 from rhotekin bearing ϳ30 g of fusion protein were added to each sample and rotated at 4°C for 30 min. The beads were collected by centrifugation at 10,000 rpm and washed twice with lysis buffer. To each sample 20 l of Laemmli buffer were added to the beads, and 12% SDS-PAGE with subsequent transfer of proteins onto nitrocellulose was performed. RhoA was detected by Western blot using specific antibody.

Activation of ROK by ADP-ribosylated RhoA-
The consequences of ADP-ribosylation on the Rho-dependent activation of the effector protein ROK were tested in a recombinant system. The activity of ROK was determined in terms of phosphorylation of the myosin binding subunit in the presence or absence of isoprenylated RhoA. As shown in Fig. 1, unmodified RhoA-GTP␥S as well as ADP-ribosylated RhoA-GTP␥S stimulated ROK activity, the latter to a lesser extent. Thus, ADPribosylation does not interfere with Rho-dependent ROK activation per se. Subcellular Distribution of RhoA in C3-treated NIH3T3 Fibroblasts-In terms of the cytosol membrane model of the Rho-GTPase cycle, active Rho is translocated to the membranes where effector proteins are also present (30). To study whether ADP-ribosylated Rho does in fact bind to membranes, the subcellular distribution of Rho after treatment of NIH3T3 cells with the chimeric C3 toxin was determined ( Fig. 2A). When cytosolic and membrane fractions of control NIH3T3 cells were adjusted to the same protein concentration, about one-half of the cellular Rho was localized in the cytosolic fraction and the other half in the membranes. ADP-ribosylation, however, resulted in a complete shift to the cytosol, and no Rho was detected anywhere in the membrane fractions ( Fig. 2A). To exclude that the observed translocation of Rho was based on the reorganization of the actin cytoskeleton, that is the consequence of ADP-ribosylation of Rho, the actin filaments were depolymerized by C. botulinum C2 toxin, that directly ADPribosylates actin thereby rendering it incapable of polymerization (31,32). Rho from these cells showed almost the same subcellular distribution as control cells (Fig. 2A). Thus, ADPribosylation rendered Rho incapable of binding to the membranes and led to its accumulation in the cytosol. The total localization of the Rho regulatory protein GDI-1 in the cytosol was altered neither by C3 nor by C2 toxin (Fig. 2A).
Rho Activity State-To check the activity state of Rho within the C3-treated cell, the amount of active ADP-ribosylated Rho was determined by pull-down experiments using the Rho binding domain of rhotekin (C21) (29). To exclude that ADP-ribosylation hampers the Rho-C21 interaction, this experiment was first carried out with recombinant proteins. As shown in Fig.  2B, recombinant GTP␥S-bound RhoA was pulled down with GST-C21 independently of the ADP-ribosylation. Subse-quently, Rho from lysates of C3-intoxicated cells was subjected to the pull-down assay with C21. As expected, some active Rho was present in fetal calf serum-cultured control cells, whereas Rho from C3-treated cells was completely inactive (Fig. 2B).
Binding of ADP-ribosylated RhoA to GDI-1-Because cellular ADP-ribosylated Rho was completely cytosolic, we addressed the question whether the modified Rho was monomeric or complexed with GDI-1 (33,34). To separate monomeric RhoA from the Rho-GDI complex, cytosolic fractions were passed through an ultra-centrifugation membrane with a 30-kDa cut off. As shown in Fig. 3A, both RhoA (molecular mass ϳ 20 kDa) and GDI-1 (molecular mass ϳ 24 kDa) were detected in the supernatant but not in the filtrate suggesting that RhoA was completely complexed with GDI-1 (molecular mass of about 45 kDa). Addition of PIP 2 to control cytosol led to the detection of RhoA and GDI-1 in the filtrate (Fig. 3A), indicating that PIP 2 indeed released RhoA from the complex and induced the for- NIH3T3 fibroblasts were treated with the indicated toxins (each 1 g/ml for 4 h) until the cells were completely rounded; the lysates were fractionated by ultracentrifugation (60 min at 100,000 ϫ g). Cytosol (cyt) and membranes (mem) were adjusted to the same protein concentration. Proteins were separated by SDS-PAGE and analyzed for RhoA and GDI-1 by immunoblot analysis. B, cellular Rho from C3-treated cells is inactive. Recombinant RhoA as well as Rho from lysates were subjected to C21 pull-down as described under "Experimental Procedures." Precipitated Rho was detected using Western blot analysis. Complete ADP-ribosylation of Rho is indicated by gel shift of RhoA to a higher apparent molecular weight.

FIG. 3. ADP-ribosylated RhoA is complexed with GDI-1.
A, fractionation of cytosol using a 30-kDa cut-off membrane filter. Cytosols from control and C3-treated cells were incubated without or with 0.2 mg/ml PIP 2 followed by separation through a 30-kDa cut-off membrane filter. Filtrate and supernatant were analyzed for RhoA and GDI-1 by immunoblot analysis. B, gel permeation chromatography. Cytosols from control and C3-treated cells were chromatographed on Superdex 75 column. Fractions were analyzed for RhoA and GDI-1 by immunoblotting. RhoA and GDI-1 from control cytosol eluted in fractions 6 -9 corresponding to an apparent molecular mass of 50 -40 kDa. A minor amount of monomeric RhoA eluted in fraction 11 (molecular mass 20 -30 kDa). Cytosolic Rho from C3-treated cells eluted also in fractions 6 -9, corresponding to an apparent molecular mass of 40 -50 kDa. C, nucleotide-dependent binding of ADP-ribosylated RhoA to GDI-1. RhoA was ADP-ribosylated and preloaded with GDP or GTP␥S as described under "Experimental Procedures." GST-GDI-1 and GST as control immobilized to glutathione agarose were incubated with RhoA as indicated at 4°C for 1 h. Eluted proteins were probed for RhoA by immunoblot analysis. mation of monomeric RhoA in the cytosol. However, the ADPribosylated Rho-GDI complex in the cytosol persisted in the presence of PIP 2 (Fig. 3A). To directly prove the presence of the Rho-GDI complex in the supernatant, gel permeation chromatography of cytosolic fractions prepared from control and from C3-treated cells was performed. RhoA from control cytosol eluted with an apparent molecular mass of 40 -50 kDa and a minority with 20 -30 kDa (Fig. 3B). RhoA from C3-treated cells eluted completely at 40 -50 kDa indicating that ADP-ribosylated RhoA was complexed with GDI-1 or other proteins but was not monomeric.
Binding of ADP-ribosylated Rho to GDI-1 was also tested in a recombinant system by nucleotide-dependent co-precipitation of Rho with GST-GDI-1 immobilized to Sepharose. As shown in Fig. 3C, unmodified as well as modified RhoA bound to GDI-1 when loaded with GDP but only faintly when bound to GTP␥S. The faint binding of GTP␥S-loaded Rho reflects a low affinity binding of RhoA-GTP␥S to GDI-1. ADP-ribosylated Rho bound to GDI-1 with higher affinity in comparison to non-modified Rho (Fig. 3C). A further indication for higher affinity came from a precipitation experiment of RhoA from cytosolic fractions using immobilized GST-GDI-1. Added GST-GDI-1 competed with endogenous GDI-1 for RhoA binding by complete precipitation of cytosolic Rho. However, GST-GDI-1 was incapable of precipitating modified RhoA from cytosols prepared from C3-treated cells (data not shown). Thus, the ADP-ribosylated RhoA-GDI-1 complex seemed to be more stable than the complex with non-modified Rho.
Activitation of ADP-ribosylated Rho from Lysates-To confirm that the increased stability of the ADP-ribosylated Rho-GDI complex in fact hampers Rho activation, lysates were treated with and without C3 and NAD followed by GTP␥S incubation. Active Rho was determined by rhotekin pull-down. In contrast to the condition in Fig. 2B, the pull-down was carried out in the absence of any detergents. The reason for the latter condition is not to disturb the stability of the Rho-GDI complex. As shown in Fig. 4, unmodified Rho-GTP␥S was active whereas Rho from C3-treated lysates turned out to be inactive. This finding underlines that GTP␥S was capable of releasing unmodified Rho but not ADP-ribosylated Rho from the GDI complex. As a control the sequence of reaction was reversed: lysates were pretreated with GTP␥S followed by ADP-ribosylation. This sequence of reaction guaranteed the presence of monomeric Rho and GDI-1. Monomeric Rho-GTP␥S was then ADP-ribosylated and pulled down by rhotekin (Fig. 4). Thus, entrapment of ADP-ribosylated Rho in the GDI complex effectively prevented Rho activation by GTP␥S. DISCUSSION Because of the widespread application of C3 as a tool in cell biology, the knowledge of the cellular mode of action is essen-tial for the understanding of C3 effects on signal transduction. However, it is still unclear how ADP-ribosylation causes inactivation of the Rho GTPases. Studies on the functional consequences of ADP-ribosylation on the GTPase cycle of Rho do not reveal remarkable changes in nucleotide binding (35,36), intrinsic GTPase cycle (36,37), and GTPase-activating proteinstimulated GTPases activity (36,37). Furthermore, ADP-ribosylated Rho is still capable of binding to effector proteins: nucleotide independently to the lipid kinases phospholipase D1 (38) and phosphatidylinositol 4-phosphate 5-kinase (39) but nucleotide dependently to rhotekin (38), Rho kinase (38), and protein kinase N (36).
C3-catalyzed ADP-ribosylation is thought to inactivate the cellular functions of Rho by blocking Rho-effector coupling, the crucial step in downstream signaling (40 -42). ROK belongs to a class of Rho effector proteins that are characterized by an autoinhibitory intramolecular interaction released by binding to RhoA (43,44). This regulatory concept is based on the findings that the Rho binding domain (RBD)-deficient ROK is constitutively active and that an antibody toward the RBD of ROK is sufficient for activating ROK, thus mimicking Rho (44). Based on the latter finding, mere binding of ADP-ribosylated RhoA should be sufficient for ROK activation as presented above in a recombinant system. However, direct inhibition of Rho-effector coupling is obviously not the mode of action for how Rho is inactivated by ADP-ribosylation in intact cells. The notion that ADP-ribosylated Rho is biologically active is further supported by the observation that sequential treatment of cells with C3 followed by CNF1 results in typical features of the CNF1 morphology (22,24). ADP-ribosylated RhoA Q63E , likely loaded with cellular GTP, is obviously active, i.e. capable of downstream signaling. The ADP-ribose appears to be inert.
The prominent finding after treatment of cells with C3 was the altered subcellular distribution of Rho, so that ADP-ribosylated cellular Rho was completely localized to the cytosolic fraction. From this finding arises the question whether cytosolic Rho is monomeric or bound to GDI-1. From the gel filtration and the ultrafiltration experiments, it became clear that ADPribosylated Rho is bound to GDI. Comparable data, the complexation of ADP-ribosylated Rho with GDI, have been shown in smooth muscle strips using immunoprecipitation techniques (45). We confirmed this finding directly, by the binding of recombinant ADP-ribosylated RhoA to recombinant GDI-1 thereby excluding the involvement of any additional factors. From the structure of the RhoA-GDI-1 complex can be deduced that the ADP-ribose at Asn-41 is directed to the solvent and may not hamper the formation of the complex (46). ADP-ribosylation, however, changes the binding properties of Rho to GDI-1. PIP 2 and GTP␥S, which are reported to release Rho from the complex (8,9,45), were unable to do so when GDI-1 was complexed with ADP-ribosylated RhoA. Our findings on the stability of the ADP-ribosylated Rho-GDI complex are an indication that the modified complex escapes the release mechanisms. This notion is further supported by the finding that recombinant GDI can compete with cellular GDI for Rho binding but not when cellular GDI is bound to ADP-ribosylated Rho. Thus, modified Rho seems to be trapped in the cytosolic GDI complex thereby escaping the regulatory cycle.
It is generally accepted that the activation of Rho is accompanied by a translocation to the membranes and that membrane-bound Rho reflects the active one (47,48). The ERM proteins are considered as displacement factors that release Rho from the GDI complex to allow subsequent interaction with the exchange factors (GEF) (11,12). The GEFs are located at the plasma membrane as shown for the Rho-specific Lbc (49). Thus, ADP-ribosylation inhibits Rho activation by GEFs   FIG. 4. Activation of ADP-ribosylated Rho from lysates. Lysates were incubated with GTP␥S and C3 in the sequence as indicated. Rho pull-down was performed using rhotekin RBD (C21). RhoA was analyzed by immunoblot analysis. ADP-ribosylated Rho from lysates first treated with GTP␥S followed by ADP-ribosylation as well as unmodified Rho-GTP␥S turned out to be active. through prevention of its membrane binding. In a recombinant system, ADP-ribosylation reduced Lbc-catalyzed GTP loading of Rho (22). Both effects together may prevent GTP loading of ADP-ribosylated Rho. This notion is supported by our finding on the absence of any active ADP-ribosylated Rho from C3treated cells as determined by rhotekin pull-down.
In terms of the cytosol-membrane model of Rho GTPase cycle a functional effector protein is characterized by membrane association. Because ADP-ribosylation entraps Rho in the cytosolic GDI complex, ADP-ribosylation indirectly blocks coupling of Rho to a membrane-bound effector protein. This notion meets the former hypothesis of blocked Rho-effector coupling by ADP-ribosylation; however, under special conditions such as the sequential intoxication with C3 and CNF1, ADP-ribosylated RhoA Q63E is likely to escape GDI-dependent inactivation due to the low affinity of modified Rho-GTP to GDI-1 (22,24,50).
Based on our current findings we propose a model of how ADP-ribosylation results in inactive Rho: ADP-ribosylated Rho-GDP binds strongly to GDI-1 and is trapped in this complex. The modified complex possesses altered properties so that ADP-ribosylated Rho is not released by activating signal inputs. ADP-ribosylated Rho, incapable of being translocated to the membranes, fails to meet membrane-associated players of the GTPase cycle such as exchange factors and effector proteins. ADP-ribosylation at Asn-41 captures Rho in the inactive form.