Active, GTP-bound small GTPases need to be attached to membranes by post-translational lipid modifications in order to process and propagate information in cells. However, generating and manipulating lipidated GTPases has remained difficult, which has limited our quantitative understanding of their activation by guanine nucleotide exchange factors (GEFs) and their termination by GTPase-activating proteins. Here, we replaced the lipid modification by a histidine tag in 11 full-length, human small GTPases belonging to the Arf, Rho and Rab families, which allowed to tether them to nickel–lipid-containing membranes and characterize the kinetics of their activation by GEFs. Remarkably, this strategy uncovered large effects of membranes on the efficiency and/or specificity in all systems studied. Notably, it recapitulated the release of autoinhibition of Arf1, Arf3, Arf4, Arf5 and Arf6 GTPases by membranes and revealed that all isoforms are efficiently activated by two GEFs with different regulatory regimes, ARNO and Brag2. It demonstrated that membranes stimulate the GEF activity of Trio toward RhoG by ∼30 fold and Rac1 by ∼10 fold, and uncovered a previously unknown broader specificity toward RhoA and Cdc42 that was undetectable in solution. Finally, it demonstrated that the exceptional affinity of the bacterial RabGEF DrrA for the phosphoinositide PI(4)P delimits the activation of Rab1 to the immediate vicinity of the membrane-bound GEF. Our study thus validates the histidine-tag strategy as a potent and simple means to mimic small GTPase lipidation, which opens a variety of applications to uncover regulations brought about by membranes.

Introduction

A vast majority of small GTPases couple their GDP/GTP structural cycle to cytosol/membrane alternation to function as versatile molecular switches in the cell (reviewed in ref. [1]). Membrane localization of their active, GTP-bound form is pivotal to their ability to propagate information, and this requires their post-translational modification by lipids (reviewed in ref. [2]). Ras, Rho and Rab family small GTPases are modified by isoprenoid lipids attached to their C-terminus and are maintained in a soluble form by GDIs, which shield the lipid from the solvent and dissociate from the GTPases prior to their activation by guanine nucleotide exchange factors (GEFs; reviewed in ref. [3]). Arf GTPases are modified by a myristate attached to their N-terminus, which is shielded by intramolecular interactions in their inactive state [4]. The myristoylated N-terminus is autoinhibitory in solution and is displaced by membranes, which primes Arf GTPases for subsequent activation by their GEFs (reviewed in ref. [5]). Accordingly, the functional cycle of small GTPases, including their activation by GEFs, their interaction with effectors and termination of their activity by GTPase-activating proteins (GAPs), mostly takes place at the membrane interface. Studies of the activation of myristoylated Arf GTPases by their GEFs reconstituted on artificial membranes revealed that membranes are a major determinant of the GEF reaction [610], underlying that actual efficiencies cannot be reliably recapitulated by solution assays.

Methods to generate and manipulate lipidated small GTPases have been devised, yet have remained tedious to implement. Because of their intramolecular GDI mechanism, myristoylated Arf GTPases can be obtained by co-expression with N-myristoyltransferase in bacteria [1113] or by modification of purified proteins by recombinant N-myristoyltransferases [14], but the procedure is tedious and the yield is limited by inefficient cleavage of the N-terminal methionine in the bacteria [14], which has hampered their use as standard biochemical tools. All other small GTPases are modified by insoluble farnesyl or geranylgeranyl isoprenoids that require solubilization by cognate GDIs. Soluble isoprenylated GTPases–GDI complexes have been produced using eukaryotically expressed lipidated GTPases either by co-expression with their cognate GDI or by subsequent membrane extraction by bacterially expressed GDI [1517]. This approach was used to highlight that membrane translocation of Rac favors its activation by the RacGEF Tiam [18]. Alternatively, Rab GTPases could be prenylated in vitro in complex with escort protein using immobilized geranylgeranyl transferase [19]. An elegant alternative has been the semisynthesis of lipidated GTPases, involving bacterial expression of truncated non-lipidated proteins and their covalent coupling to chemically synthesized lipidated peptides [20], which found various applications in investigating the biology of Ras and other GTPases on membranes (reviewed in ref. [21]).

However, the above approaches have remained difficult to implement, which has limited the quantitative characterization of their reactions on membranes, begging for alternatives based on soluble GTPases. Two main approaches have been devised: covalent cross-linking of a cysteine residue to liposomes containing maleimide-derivatized lipids and noncovalent tethering by a polyhistidine tag to liposomes containing Ni2+-NTA lipids (Ni-lipids). Both strategies uncovered pivotal contributions of membranes to the regulation of small GTPases. Thiol-maleimide cross-linking of Ras to membranes showed that membranes induce a large enhancement of the exchange rate of the RasGEF SOS [22]. Likewise, reconstitution of His-tagged RhoA-GTP in liposomes enhanced activation of RhoA by PDZ-RhoGEF, highlighting a positive feedback effect that was undetectable in solution [23]. Membrane tethering by a histidine tag of yeast Ypt7 and Ypt21/32, both members of the Rab family, also uncovered a potent GEF activity of their respective GEFs, Mon/Ccz1 [24] and TRAPPII [25], which had remained uncertain in solution. These studies point to the considerable potential of artificially tethered GTPases as standard tools to monitor the activity of small GTPases on membranes in a quantitative manner; however, their use has remained limited and comparison with lipidated GTPases has not been done.

Here, we designed His-tagged versions of 11 members of the Arf, Rho and Rab GTPases and characterized their kinetics of activation by GEFs on Ni-lipid-containing artificial membranes. We first used Arf1 and Arf6 GTPases, for which extensive characterization of the lipidated proteins is available, to demonstrate that the His-tagged GTPases quantitatively recapitulate the contribution of membranes to their activation by GEFs. We then extented the approach to all five Arf isoforms, to four major representatives of the Rho GTPases family and to two Rab GTPases that are hijacked by bacterial effectors. In all cases, the approach uncovered important contributions of membranes to the specificity and efficiency of the reactions. We conclude that tethering small GTPases to membranes by a histidine tag provides a general and easy-to-implement means to investigate their activities on membranes, which should be of broad application for understanding how GEFs, GAPs and effectors work at the membrane interface.

Results

Reconstitution of membrane-dependent activation of Arf1 and Arf6 by histidine-tag tethering

The regulation, efficiency and specificity of Arf GTPases and their GEFs cannot be reliably assessed in solution, notably because of their autoinhibition by their N-terminal helix (reviewed in ref. [26]). To assess whether tethering small GTPases to membranes by a histidine tag recapitulates the properties of the lipidated proteins, we focused on Arf1, a major regulator of membrane traffic throughout the cell, and Arf6, which functions at the crossroads of traffic and cytoskeletal dynamics at the plasma membrane (reviewed in ref. [27]), because the kinetics of activation of their myristoylated forms have been extensively characterized. We fused Arf1 and Arf6 with a 6-histidine tag in the N-terminus followed by a 2-glycine linker (Table 1). Both proteins were expressed in Escherichia coli and purified to homogeneity (Figure 1A, left panel), and the presence of the N-terminal 6xHis-tag was confirmed by western blot (Figure 1A, right panel). His-Arf1 and His-Arf6 bound efficiently to liposomes containing a small percentage of Ni-lipids (5%) as shown by a stringent liposome flotation assay (Figure 1B), indicating that the location of the tag on a switch element is compatible with its binding to membranes.

Comparison of His-tagged and myristoylated Arf1 and Arf6.

Figure 1.
Comparison of His-tagged and myristoylated Arf1 and Arf6.

(A) Purity of the Arf GTPases and their GEF and the presence of the histidine tag were checked by SDS–PAGE (left) and anti-His-tag western blot (right). (B) Binding of His-Arf GTPases to Ni-liposomes was analyzed by liposome flotation. Proteins bound to liposomes are found in the top fraction (lanes ‘T’); unbound proteins are in the bottom fraction (lanes ‘B’). (C) Activation of liposome-bound His-Arf1 and His-Arf6 (500 nM) by ARNO (10 nM) was measured by tryptophan fluorescence kinetics; a.u., arbitrary unit. (D) Representative liposome size distribution measured by DLS, showing the absence of physical alterations. (E) Determination of kcat/Km of ARNO for His-tagged and myristoyled Arf1 and Arf6 in the presence of liposomes. All liposomes contained Ni-lipids, except for myrArf6 which has an uncleavable C-terminal His-tag that may interfere with the reaction.

Figure 1.
Comparison of His-tagged and myristoylated Arf1 and Arf6.

(A) Purity of the Arf GTPases and their GEF and the presence of the histidine tag were checked by SDS–PAGE (left) and anti-His-tag western blot (right). (B) Binding of His-Arf GTPases to Ni-liposomes was analyzed by liposome flotation. Proteins bound to liposomes are found in the top fraction (lanes ‘T’); unbound proteins are in the bottom fraction (lanes ‘B’). (C) Activation of liposome-bound His-Arf1 and His-Arf6 (500 nM) by ARNO (10 nM) was measured by tryptophan fluorescence kinetics; a.u., arbitrary unit. (D) Representative liposome size distribution measured by DLS, showing the absence of physical alterations. (E) Determination of kcat/Km of ARNO for His-tagged and myristoyled Arf1 and Arf6 in the presence of liposomes. All liposomes contained Ni-lipids, except for myrArf6 which has an uncleavable C-terminal His-tag that may interfere with the reaction.

Table 1
Sequences of the N-terminal helix (Arf) and C-terminal extensions (Rho and Rab) of the His-tagged GTPase constructs used in the present study

The residue carrying the lipid post-translational modification is indicated in bold. The sequences of the tags are underlined.

GTPase Natural lipid Sequence 
His-Arf1 Myristate GHHHHHHGGGNIFANLFKGLFGKK EMRI … 
His-Arf3 Myristate GHHHHHHGGGNIFGNLLKSLIGKK EMRI … 
His-Arf4 Myristate GHHHHHHGGGLTISSLFSRLFGKK QMRI … 
His-Arf5 Myristate GHHHHHHGGGLTVSALFSRIFGKK QMRI … 
His-Arf6 Myristate GHHHHHHGGGKVLSKIFGNK EMRI … 
RhoG-His Geranylgeranyl  … AVRAVL NPTPIKRGRSCILL HHHHHH 
Rac1-His Geranylgeranyl  … AIRAVL CPPPVKKRKRKCLLL HHHHHH 
RhoA-His Geranylgeranyl  … ATRAAL QARRGKKKSGCLVL HHHHHH 
Cdc42-His Geranylgeranyl  … AILAAL EPPEPTKKRKCKFL HHHHHH 
Rab1-His Geranylgeranyl … EIKKR MGCPGATAGGAEKSNVKIQSTPVKQSGGGCCHHHHHH 
Rab35-His Geranylgeranyl  … AKKDN LAKQQQQQQNDVVKLTKNSKRKKRCCHHHHHH 
GTPase Natural lipid Sequence 
His-Arf1 Myristate GHHHHHHGGGNIFANLFKGLFGKK EMRI … 
His-Arf3 Myristate GHHHHHHGGGNIFGNLLKSLIGKK EMRI … 
His-Arf4 Myristate GHHHHHHGGGLTISSLFSRLFGKK QMRI … 
His-Arf5 Myristate GHHHHHHGGGLTVSALFSRIFGKK QMRI … 
His-Arf6 Myristate GHHHHHHGGGKVLSKIFGNK EMRI … 
RhoG-His Geranylgeranyl  … AVRAVL NPTPIKRGRSCILL HHHHHH 
Rac1-His Geranylgeranyl  … AIRAVL CPPPVKKRKRKCLLL HHHHHH 
RhoA-His Geranylgeranyl  … ATRAAL QARRGKKKSGCLVL HHHHHH 
Cdc42-His Geranylgeranyl  … AILAAL EPPEPTKKRKCKFL HHHHHH 
Rab1-His Geranylgeranyl … EIKKR MGCPGATAGGAEKSNVKIQSTPVKQSGGGCCHHHHHH 
Rab35-His Geranylgeranyl  … AKKDN LAKQQQQQQNDVVKLTKNSKRKKRCCHHHHHH 

Next, we determined the activation rates of His-Arf1 and His-Arf6 by ARNO, an ArfGEF of the cytohesin subfamily involved in signal transduction which consists of a Sec7 domain carrying the GEF activity and a pleckstrin homology (PH) domain that binds PI(4,5)P2 (PIP2) and PI(3,4,5)P3 phosphoinositides [6,28]. Spontaneous and ARNO-stimulated nucleotide exchange kinetics were monitored by tryptophan fluorescence in solution and in the presence of liposomes containing Ni-lipids and PIP2 (Figure 1C). We checked by dynamic light scattering (DLS) that no aggregation or physical alteration of liposomes occurred in the course of the reaction (Figure 1D). His-Arf1 and His-Arf6 were resistant to activation in solution, indicating that the replacement of the myristate by a 6xHis-tag preserves autoinhibition. The addition of liposomes resulted in slow spontaneous nucleotide exchange, probably reflecting the fact that membranes unlock the N-terminal helix to facilitate subsequent activation by GTP (reviewed in ref. [5]). Remarkably, nucleotide exchange in the presence of liposomes was strongly increased by catalytic amounts of ARNO (1:50 ratio), and this increase was suppressed by displacing the His-tagged GTPases from membranes by imidazole.

We then compared the efficiency of the activation of myristoylated and His-tagged Arf1 and Arf6 by determining kcat/Km on liposomes accurately using a range of ARNO concentrations. ARNO activated both His-Arf1 and His-Arf6 with high efficiencies, and these were in the same range as those of the myristoylated proteins (Figure 1E and Table 2). Altogether, these experiments show that the contribution of membranes to the efficiency of activation of myristoylated Arf GTPases by their GEFs is efficiently recapitulated by their attachment to membranes by a 6xHis anchor.

Table 2
kcat/Km values determined in the present study
GTPase/GEF Liposomes kcat/Km (×105 M−1 s−1
myrArf1/ARNO 130.1 
His-Arf1/ARNO 52.1 
myrArf6/ARNO 65.6 
His-Arf6/ARNO 13.5 
RhoG-His/Trio − 1.9 
RhoG-His/Trio 59.1 
Rac1-His/Trio − 0.9 
Rac1-His/Trio 8.2 
Rab1-His/DrrA − 8.5 
Rab1-His/DrrA 3.4 
GTPase/GEF Liposomes kcat/Km (×105 M−1 s−1
myrArf1/ARNO 130.1 
His-Arf1/ARNO 52.1 
myrArf6/ARNO 65.6 
His-Arf6/ARNO 13.5 
RhoG-His/Trio − 1.9 
RhoG-His/Trio 59.1 
Rac1-His/Trio − 0.9 
Rac1-His/Trio 8.2 
Rab1-His/DrrA − 8.5 
Rab1-His/DrrA 3.4 

Family-wide analysis of the efficiency and specificity of ArfGEFs using His-tagged Arf GTPases

The Arf GTPase family consists of five members, of which the biochemistry and function of class I Arf3 and class II Arf4 and Arf5 have remained poorly understood. Notably, these isoforms have remained difficult to purify in their myristoylated form, which has limited the investigation of their activation by GEFs. All Arf isoforms share >60% identity with Arf1, the highest divergence being in their N-terminal helix. We designed His-tagged versions of human Arf3, Arf4 and Arf5 (Table 1), which were purified to homogeneity (Figure 1A, left panel) and assessed for the presence of the His-tag (Figure 1A, right panel) and binding to Ni-lipid-containing liposomes (Figure 1B). All three Arf isoforms were resistant to the activation by ARNO in solution, and their spontaneous nucleotide exchange in the presence of liposomes was very slow (Figure 2A). In contrast, all isoforms were efficiently activated by ARNO in the presence of liposomes, revealing that ARNO does not display significant specificity within the Arf family (Figure 2A,B).

Family-wide analysis of the activation of His-Arf GTPases in the presence of Ni-liposomes.

Figure 2.
Family-wide analysis of the activation of His-Arf GTPases in the presence of Ni-liposomes.

(A) Activation of liposome-bound His-Arf GTPases (500 nM) by ARNO (10 nM) measured by tryptophan fluorescence kinetics. (B) Activation rates (kobs) of His-Arf GTPases by ARNO on liposomes. (C) Activation kinetics of liposome-bound His-Arf GTPases (500 nM) by BRAG2 (10 nM) measured as in (A). (D) Activation rates (kobs) of His-Arf GTPases by BRAG2 on liposomes.

Figure 2.
Family-wide analysis of the activation of His-Arf GTPases in the presence of Ni-liposomes.

(A) Activation of liposome-bound His-Arf GTPases (500 nM) by ARNO (10 nM) measured by tryptophan fluorescence kinetics. (B) Activation rates (kobs) of His-Arf GTPases by ARNO on liposomes. (C) Activation kinetics of liposome-bound His-Arf GTPases (500 nM) by BRAG2 (10 nM) measured as in (A). (D) Activation rates (kobs) of His-Arf GTPases by BRAG2 on liposomes.

A unique feature of ARNO is that its Sec7 domain is autoinhibited by its PH domain and autoinhibition is released by interaction of the PH domain with Arf-GTP [6,28,29]. In contrast, BRAG2, which is involved in endosomal signaling, is constitutively active and does not feature feedback regulation (reviewed in ref. [30]). To investigate the response of the His-tagged Arf isoforms to GEFs with different regulatory regimes, we measured their activation rates by BRAG2 on membranes (Figure 2C,D). As for ARNO, all Arf isoforms were efficiently activated by BRAG2, revealing that this GEF, originally described as being specific for Arf6 and later shown to activate Arf1, is in fact a potent activator for all members of the Arf family. Altogether, these results indicate that the His-tagged Arf isoforms recapitulated the mechanistic properties of the myristoylated GTPases, and uncover that the GEFs ARNO and BRAG2 activate all Arf isoforms.

His-tagged Rho family GTPases highlight activity enhancement and broader specificity of the GEF Trio on membranes

Small GTPases of the Rho/Rac/Cdc42 family regulate many aspects of actin cytoskeleton dynamics whereby they control cell shape and motility (reviewed in ref. [31]). They are modified by a geranylgeranyl lipid attached to a cysteine in their variable C-terminus, which binds strongly to membranes and renders them insoluble in the absence of a GDI (reviewed in ref. [32]). Unlike Arf GTPases, they do not have autoinhibitory elements and are readily activated in solution. However, because geranylgeranylated proteins are difficult to purify and manipulate, our understanding of the contribution of membranes to the regulation of these GTPases has remained fragmentary. We designed His-tagged versions of human Rac1, RhoA, Cdc42 and RhoG, in which six histidines were added immediately after the C-terminal residue (Table 1). All proteins were purified to homogeneity and assessed for the presence of the His-tag and for their ability to bind to Ni-lipid-containing liposomes (Figure 3A,B). To analyze their activation on membranes, we selected the N-terminal DH-PH tandem (DH1-PH1) of Trio, a RhoGEF with a prominent role in neuronal development, cell adhesion and G-protein-coupled receptor signaling (reviewed in ref. [33]). The DH1-PH1 domain has been reported to bind pure phospholipids detectably only in the presence of RhoG in a dot-blot assay, and this required its unlipidated C-terminus, yet phospholipid headgroups had no effect on its GEF efficiency in solution [34].

Membranes contribute to the efficiency and specificity of the RhoGEF Trio toward Rho family GTPases.

Figure 3.
Membranes contribute to the efficiency and specificity of the RhoGEF Trio toward Rho family GTPases.

(A) Purity of Rho-His GTPases and of the GEF Trio and the presence of the histidine tag were checked by SDS–PAGE (left) and anti-His-tag western blot (right). (B) Binding of the Rho-His GTPases to Ni-liposomes was analyzed by liposome flotation. See Figure 1B for B and T labels. (C) Recruitment of Trio by liposome-bound RhoG-His, analyzed by liposome flotation. See Figure 1B for B and T labels. (D) Kinetics of activation of liposome-bound RhoG-His and Rac1-His (500 nM) by Trio (10 nM) measured by mant-GDP fluorescence. (E) Kinetics of activation of liposome-bound Cdc42-His and RhoA-His by Trio, measured as in (D). (F) Determination of kcat/Km of Trio for RhoG-His and Rac1-His in the presence and absence of liposomes. (G) Comparison of the activation rates (kobs) of Rho-His GTPases by Trio in the presence and absence of liposomes.

Figure 3.
Membranes contribute to the efficiency and specificity of the RhoGEF Trio toward Rho family GTPases.

(A) Purity of Rho-His GTPases and of the GEF Trio and the presence of the histidine tag were checked by SDS–PAGE (left) and anti-His-tag western blot (right). (B) Binding of the Rho-His GTPases to Ni-liposomes was analyzed by liposome flotation. See Figure 1B for B and T labels. (C) Recruitment of Trio by liposome-bound RhoG-His, analyzed by liposome flotation. See Figure 1B for B and T labels. (D) Kinetics of activation of liposome-bound RhoG-His and Rac1-His (500 nM) by Trio (10 nM) measured by mant-GDP fluorescence. (E) Kinetics of activation of liposome-bound Cdc42-His and RhoA-His by Trio, measured as in (D). (F) Determination of kcat/Km of Trio for RhoG-His and Rac1-His in the presence and absence of liposomes. (G) Comparison of the activation rates (kobs) of Rho-His GTPases by Trio in the presence and absence of liposomes.

These intriguing features call for the analysis of Trio interactions and GEF activity in the presence of membranes and membrane-attached GTPases. First, we used a liposome flotation assay to characterize the recruitment of Trio to membranes. We find that Trio is poorly recruited to membranes on its own, but it is entirely bound to liposomes containing RhoG-His and this effect is lost upon dissociation of RhoG-His from liposomes by imidazole (Figure 3C). Next, we investigated the effect of membranes on the efficiency of Trio. His-tagged GTPases were loaded with mant-GDP and the kinetics of nucleotide exchange were monitored by the decrease in fluorescence following mant-GDP dissociation. In solution, Trio activated both RhoG-His and Rac1-His efficiently (Figure 3D) and was inactive toward RhoA-His and Cdc42-His (Figure 3E), which is consistent with previous studies using unlipidated GTPases [3436]. Remarkably, membranes increased kcat/Km by 30-fold for RhoG-His and 8.5-fold for Rac1-His (Figure 3F and Table 2). This enhancing effect was lost when RhoG-His and Rac1-His were dissociated from liposomes by imidazole, indicating that it originates from the membrane association of the GTPase. We note that the effect of imidazole on Rac1-His is incomplete, possibly due to an interaction of its C-terminus with anionic phospholipids. Thus, the reconstitution of His-tagged RhoG and Rac1 in liposomes uncovered a strong contribution of membranes to the efficiency of Trio that was not detected using isolated phospholipids.

Finally, we analyzed the specificity of the DH1-PH1 tandem of Trio by comparing its efficiency on RhoG and Rac1, its known substrates, with that on RhoA, the substrate of the second DH-PH tandem, and on a related GTPase, Cdc42. To our surprise, we observed that Trio potently activated RhoA-His and Cdc42-His on membranes, although with a lower exchange rate than RhoG-His and Rac1-His (Figure 3E). This is in striking contrast with the situation in solution where RhoA and Cdc42 were completely resistant to activation. A measure of the discrepancy between the solution and liposome assays is provided by the comparison of kobs values determined at identical GTPase and GEF concentrations, which shows that Trio activates its poorer substrate, Cdc42, on membranes more efficiently than it does its favored substrate, RhoG, in solution (Figure 3G). Altogether, tethering Rho GTPases to membranes by a histidine tag provided quantitative information on the contribution of membranes to their regulation and uncovered a previously overlooked broader specificity.

Membrane-delimited activation of Rab1 by a bacterial RabGEF

Rab GTPases are master organizers of intracellular membrane traffic and organelle identity (reviewed in ref. [37]). As Rho GTPases, they are recruited to membranes by geranylgeranyl lipids covalently attached to their C-terminal extension, the sequence of which is highly variable among Rab GTPases. Many Rab regulators and effectors carry membrane-binding domains that are highly specific for given phosphoinositides, but the contribution of phosphoinositide-containing membranes to the efficiency of the activation/inactivation cycle has remained poorly studied due to the difficulty in generating and manipulating lipidated Rab GTPases. To investigate our His-tag tethering strategy in the Rab family, we selected human Rab1, a major regulator of membrane traffic at the ER (reviewed in ref. [38]), and Legionella pneumophila DrrA, a RabGEF that activates cellular Rab1 on the Legionella-containing vacuole during infection [39]. An important characteristic of DrrA is that it binds PI(4)P with exceptional affinity [40,41]; however, the contribution of PI(4)P-containing membranes to its GEF activity is not known. Rab1 fused to a 6-histidine tag in the C-terminus (Table 1) was purified to homogeneity and assessed for the presence of the His-tag and its ability to bind exclusively to liposomes containing Ni-lipids (Figure 4A–C). Likewise, DrrA bound strongly to liposomes containing PI(4)P and Ni-lipids but not to liposomes containing only Ni-lipids (Figure 4C). DrrA strongly activated Rab1-His both in solution and in the presence of liposomes containing both PI(4)P to recruit DrrA and Ni-lipids to tether Rab1-His (Figure 4D), although with different efficiencies (see below).

Membrane-delimited activation of human Rab1 by bacterial DrrA.

Figure 4.
Membrane-delimited activation of human Rab1 by bacterial DrrA.

(A) Purity of Rab-His GTPases and DrrA and the presence of the histidine tag were checked by SDS–PAGE (left) and anti-His-tag western blot (right). (B) Binding of the Rab-His GTPases to Ni-liposomes was analyzed by liposome flotation. Liposomes contained 91% PC, 4% PI4P, 5% DGS-NTA-Ni and 0.2% NBD-PE. See Figure 1B for B and T labels. (C) Selective binding of DrrA to PI(4)P-containing liposomes and of Rab1-His to Ni-lipid-containing liposomes. Liposome compositions were as follows: Ni-liposomes: 95% PC, 5% DGS-Ni-NTA, 0.2% NBD-PE; PI(4)P-liposomes: 96% PC, 4% PI(4)P, 0.2% NBD-PE; Ni + PI(4)P-liposomes: 91% PC, 4% PI(4)P, 5% DGS-NTA-Ni, 0.2% NBD-PE. See Figure 1B for B and T labels. (D) Kinetics of activation of liposome-bound Rab1-His (500 nM) by DrrA (25 nM), measured by mant-GDP fluorescence using Ni + PI(4)P liposomes. (E) Kinetics of activation of liposome-bound Rab35-His by DrrA, measured as in (D). (F) Kinetics of activation of liposome-bound Rab1-His by DrrA measured at increasing Ni + PI(4)P liposome concentrations. (G) Analysis of membrane-delimited activation of Rab1-His by DrrA. The set-up of the kinetics assays was varied as follows: Drra + Rab1: no liposomes; DrrA + Rab1–liposomes: liposomes containing only Ni-lipids; Rab1 + Drra–liposomes: liposomes containing only PI(4)P; Rab1–DrrA–liposomes: liposomes containing Ni-lipids and PI(4)P; DrrA–liposomes and Rab1–liposomes: Rab1 and DrrA were reconstituted separately on Ni+PI(4)P-liposomes, and the two pools of proteo-liposomes were mixed just before starting the exchange reaction; DrrA–liposomes and Rab1–liposomes + imidazole: as previous one in the presence of imidazole to dissociate Rab1-His from the liposomes.

Figure 4.
Membrane-delimited activation of human Rab1 by bacterial DrrA.

(A) Purity of Rab-His GTPases and DrrA and the presence of the histidine tag were checked by SDS–PAGE (left) and anti-His-tag western blot (right). (B) Binding of the Rab-His GTPases to Ni-liposomes was analyzed by liposome flotation. Liposomes contained 91% PC, 4% PI4P, 5% DGS-NTA-Ni and 0.2% NBD-PE. See Figure 1B for B and T labels. (C) Selective binding of DrrA to PI(4)P-containing liposomes and of Rab1-His to Ni-lipid-containing liposomes. Liposome compositions were as follows: Ni-liposomes: 95% PC, 5% DGS-Ni-NTA, 0.2% NBD-PE; PI(4)P-liposomes: 96% PC, 4% PI(4)P, 0.2% NBD-PE; Ni + PI(4)P-liposomes: 91% PC, 4% PI(4)P, 5% DGS-NTA-Ni, 0.2% NBD-PE. See Figure 1B for B and T labels. (D) Kinetics of activation of liposome-bound Rab1-His (500 nM) by DrrA (25 nM), measured by mant-GDP fluorescence using Ni + PI(4)P liposomes. (E) Kinetics of activation of liposome-bound Rab35-His by DrrA, measured as in (D). (F) Kinetics of activation of liposome-bound Rab1-His by DrrA measured at increasing Ni + PI(4)P liposome concentrations. (G) Analysis of membrane-delimited activation of Rab1-His by DrrA. The set-up of the kinetics assays was varied as follows: Drra + Rab1: no liposomes; DrrA + Rab1–liposomes: liposomes containing only Ni-lipids; Rab1 + Drra–liposomes: liposomes containing only PI(4)P; Rab1–DrrA–liposomes: liposomes containing Ni-lipids and PI(4)P; DrrA–liposomes and Rab1–liposomes: Rab1 and DrrA were reconstituted separately on Ni+PI(4)P-liposomes, and the two pools of proteo-liposomes were mixed just before starting the exchange reaction; DrrA–liposomes and Rab1–liposomes + imidazole: as previous one in the presence of imidazole to dissociate Rab1-His from the liposomes.

Next, we assessed whether membranes would broaden the specificity of DrrA as observed for Trio. We chose human Rab35, which regulates endocytic recycling (reviewed in ref. [42]) and bears resemblances to Rab1 that allow it to be the substrate of another Legionella effector, AnkX [43]. We fused full-length Rab35 to a C-terminal 6-histidine tag, purified it to homogeneity (Figure 4A) and controlled that it is recruited to Ni-lipid-containing liposomes (Figure 4B) and that it is efficiently activated by its specific GEF DennD in solution (Figure 4E). Remarkably, DrrA did not activate Rab35-His on liposomes (Figure 4E), highlighting a strict specificity that contrasts with the broadened specificity observed for the RacGEF Trio on membranes.

Finally, we quantified the contribution of membranes to the efficiency of DrrA by determining kcat/Km. To our surprise, the efficiency of DrrA on liposomes was reduced by 2.5 times compared with the efficiency determined in solution (Figure 4D and Table 2). Varying the concentration of liposomes did not modify the exchange rates, but the plateau of the reaction decreased with liposome concentration (Figure 4F), indicating that the fraction of activated Rab1-His decreases as the concentration of liposomes increases. To analyze whether this unexpected effect could result from the very high affinity of DrrA for PI(4)P leading to its slow dissociation, we compared the activation of liposome-bound Rab1-His by soluble DrrA and of soluble Rab1-His by liposome-bound DrrA (Figure 4G). In both cases, exchange rates were intermediate between the solution-only and the liposome-only rates. In contrast, the activation rate was further decreased when Rab1–liposomes and DrrA–liposomes were prepared separately, and this effect was reversed when Rab1-His was dissociated from liposomes by imidazole (Figure 4G). We conclude from this ensemble of experiments that DrrA is less active in the presence of membranes because it does not significantly dissociate from PI(4)P, such that it activates only Rab1 proteins that are in its immediate vicinity.

Discussion

In this work, we designed His-tagged versions of 11 small GTPases, including autoinhibited small GTPases of the Arf family and classical Rho and Rab GTPases, and we reconstituted their activation on membranes by membrane-binding GEFs. Using Arf1 and Arf6 for the validation of the approach, we found that the His-tagged GTPases recapitulated the activation of their myristoylated counterpart, thus validating the approach for quantitative biochemical assays. Importantly, we did not detect significant interference of the histidine tag with the reaction despite its location on a critical switch element, in contrast with larger tags that have been shown to impair biological functions of Arf GTPases in cells [44]. Next, we showed for a variety of systems that His-tagged GTPases provide valuable information on properties that were undetectable in solution. These systems included two ArfGEFs with opposite regulatory regimes: a RhoGEF that binds weakly to membranes and a RabGEF with exceptional affinity for phosphoinositides. Notably, we identified robust enhancement of ArfGEFs and a RhoGEF by membranes, a specificity broadening of a RacGEF that was not detectable in solution, and membrane-restricted activation of Rab1 by a bacterial RabGEF.

These findings have important biological consequences and open new perspectives for investigations. First, they reveal that all Arf isoforms are excellent substrates on membranes for two major endosomal ArfGEFs, ARNO and Brag2, a remarkable pan-family specificity that probably extends to the other members of these ArfGEF subfamilies. Thus, endosomal ArfGEFs do not select their substrates by a simple specificity rule, which suggests that several isoforms may be activated and mobilized in parallel in cells. This is consistent with previous observations that more than one Arf isoform has to be silenced to yield observable changes in organelle phenotypes [45].

In the case of Rho family GTPases, membrane tethering of RhoG showed that Trio is recruited to membranes in a RhoG-dependent manner, possibly through the engagement of the GTPase with the PH domain as seen in the Rac1–Trio co-crystal structure [36], and that this is accompanied by a large enhancement of its GEF activity. Our findings thus reconcile the conflicting observations that Trio bound isolated phospholipids only in the presence of unlipidated RhoG, yet phospholipid headgroups had no effect on its GEF efficiency [34]. Another unexpected observation is the broader specificity of Trio toward RhoA and Cdc42, which is undetectable in solution. Specificity broadening by membranes was observed before for EFA6, an ArfGEF that is strictly specific for Arf6 in solution (using truncated versions of Arf proteins to get around autoinhibition), but activates both myrAr1 and myrArf6 on membranes [10]. The presence in Trio of a second DH-PH tandem specific for the RhoA pathway makes it conceivable that the DH1-PH1 tandem generates RhoA-GTP on cellular membranes in small amounts due to its lesser efficiency when compared with RhoG and Rac1, possibly as part of a feedback mechanism which remains to be investigated. Finally, the apparent antagonistic effect of membranes on a bacterial RabGEF toward a cellular Rab GTPase highlights that the efficiency of GEFs is balanced by the strength with which they bind to the membrane. The extremely high affinity of DrrA for PI(4)P implies that its dissociation rate is very slow, effectively restricting its activity to Rab GTPases already bound in its immediate vicinity at the membrane surface. These characteristics endow the Legionella pathogen with a precise spatiotemporal control of Rab activation and its subsequent AMPylation on the Legionella-containing vacuole during infection.

Our comparison of myristoylated and His-tagged Arf GTPases highlights one limitation to keep in mind when using membrane-tethered GTPases. In the case of His-Arf1 and His-Arf6, we observed a slight increase in spontaneous nucleotide exchange, which we ascribe to the fact that the displacement of the autoinhibitory helix by membranes is facilitated in membrane-tethered Arf GTPases when compared with myristoylated GTPases which are in solution/membrane equilibrium. Accordingly, GEF efficiencies may appear lower than they actually are. Rho, Rab and Ras GTPases do not have autoinhibitory elements that maintain them in the cytosol, but they have to dissociate from their GDIs; the energetic cost of the membrane translocation step cannot be recapitulated in this set-up. Based on the case of the Arf GTPases, we believe that these effects remain small and do not compromise quantitative biochemical analysis.

In conclusion, we anticipate that the His-tagged GTPases generated in this work will be of important use to reconstitute and characterize other GEFs, effectors and GAPs in the context of membranes. More generally, we propose fusion of histidine tags adjacent to lipidation sites is an easy implementary means to tether small GTPases to membranes, which allows quantitative biochemical analysis and opens broad applications to investigate how membranes co-operate with small GTPases and their partners to monitor and process cellular signals.

Materials and methods

Protein cloning, expression and purification

The sequences coding for full-length human Arf1, Arf3, Arf4, Arf5 and Arf6 were PCR-amplified with an additional sequence in 5′ encoding an N-terminal 6-histidine tag followed by a linker of two glycines. PCR products were cloned into the pET15b expression vector. The sequence coding for full-length human RhoG, Rab1 and Rab35 was PCR-amplified with an additional sequence in 3′ encoding a 6-histidine tag immediately following their C-terminal residue and cloned into the Gateway destination vector pDEST14. Full-length human Rac1, Cdc42 and RhoA with a 6-histidine tag immediately after their C-terminal residue have been described in ref. [46]. Untagged human ARNO containing the Sec7 and PH domains (residues 50–399) cloned into pET8c expression vector is a kind gift of Bruno Antonny (CNRS, Sophia-Antipolis, France). Human Brag2 (Sec7 and PH domains, residues 390–763) carrying an N-terminal 6-histidine tag followed by a TEV cleavage site has been described in ref. [7]. Human Trio containing the DH1 and PH1 domains (residues 1232–1550) is a kind gift from Anne Blangy (CNRS, Montpellier, France). The sequence was amplified by PCR and cloned into the Gateway destination vector pETG-20A to introduce a TEV protease cleavage site. L. pneumophila DrrA [GEF and PI(4)P-binding domains, residues 340–647] carrying a 6-His tag cleavable by the TEV protease is a kind gift from Lena Osterlein and Aymelt Itzen (MPI Dortmund, Germany) and has been described in ref. [40]. Human DennD1B (a construct lacking the 150–169 loop) is a kind gift from Karin Reinisch (Yale School of Medicine, Yale University, U.S.A.) and has been described in ref. [47]. All plasmids were confirmed by sequencing (GATC Biotech).

All GTPase constructs were expressed in E. coli Rosetta (DE3) pLysS in LB medium. Overexpression was induced overnight with 0.5 mM IPTG at 20°C. Bacterial pellets were lysed by lysosyme complemented with benzonase and an antiprotease inhibitor cocktail and disrupted at 1250 psi using a pressure cell homogenizer. The cleared lysate supernatant was loaded onto a Ni-NTA affinity chromatography column (HisTrap FF, GE Healthcare) and eluted with 250 mM imidazole. Purification was perfected by gel filtration on a Superdex 75 16/600 column (GE Healthcare) equilibrated with storage buffer [50 mM Tris, 300 mM NaCl, and 2 mM MgCl2 (pH 8), supplemented with 2 mM β-mercaptoethanol and 5% glycerol for Arf and Rho GTPases]. Myristoylated Arf1 was obtained by co-expression of full-length human Arf1 and yeast N-myristoyltransferase as described in ref. [7]. Myristoylated Arf6 was obtained by in vitro myristoylation of full-length Arf6 carrying a C-terminal His-tag by recombinant myristoyltransferase as described in ref. [14].

ARNO was expressed in E. coli BL21 (DE3) Gold in LB medium. Overexpression was induced for 3 h by 0.5 mM IPTG at 37°C. Bacterial pellets were lysed as above and disrupted by sonication. The cleared lysate supernatant was loaded onto a Q-Sepharose anion exchange chromatography column and eluted with a 0–1 M NaCl gradient. Purification was perfected by gel filtration on a Superdex 200 16/600 equilibrated with storage buffer [50 mM Tris, 150 mM NaCl, 1 mM MgCl2, and 1 mM DTT (pH 8)]. Brag2 was expressed and purified as described in ref. [7]. Trio was expressed and purified as described above for the small GTPases. The 6-His tag was cleaved by incubation with the TEV protease (1/4 w/w ratio) overnight at 4°C at low stirring. The cleaved tag was eliminated by passage over a Ni-NTA affinity chromatography column (Histrap FF, GE Healthcare) equilibrated with storage buffer [50 mM Tris, 500 mM NaCl, 10% glycerol, 2 mM β-mercaptoethanol and 2 mM MgCl2 (pH 8)], and the protein was retrieved in the flowthrough. DrrA was expressed as in ref. [40], except that we used the E. coli BL21 (DE3) pG-KJE8 strain that expresses chaperones, in LB medium with 0.5 g/l l-arabinose and 2.5 mg/l tetracycline. The 6-His tag was cleaved by the TEV protease (1/10 w/w ratio) during overnight dialysis in storage buffer [20 mM HEPES, 100 mM NaCl and 2 mM β-mercaptoethanol (pH 7.4)] at 4°C. The cleaved tag was eliminated by a second Ni-NTA affinity chromatography step and the purity of cleaved protein was perfected on a Superdex 75 16/600 equilibrated with storage buffer. DennD was expressed and purified as described in ref. [47]. The presence of the His-tag on the small GTPases and the absence of the His-tag on their associated GEFs were checked by western blot with monoclonal antipolyhistidine peroxidase-conjugated antibody (Sigma; 1:5000).

Liposome synthesis and binding assays

Liposomes were prepared as described in refs [7,29], in a buffer containing 50 mM HEPES and 120 mM potassium acetate (pH 7.4). Except when indicated otherwise, experiments were carried out with liposomes containing 43% PC, 20% PE, 10% PS, 20% cholesterol, 2% PI(4,5)P2 completed with 5% DGS-NTA-Ni lipids for the attachment of His-tagged GTPases and 0.2% fluorescent NDB-PE lipids to facilitate detection. Liposomes were extruded through a 0.2 µm pore size filter before use, and size distributions were checked by DLS on a DynaPro instrument (Wyatt).

For the liposome flotation assays, 2 µM proteins and 1 mM liposomes (lipids) were incubated in HKM buffer [50 mM HEPES, 120 mM potassium acetate and 1 mM MgCl2 (pH 7.4)] at room temperature for 20 min in a total volume of 150 µl. The suspension was adjusted to 30% sucrose, overlaid with 200 µl of HKM containing 25% w/v sucrose and 50 µl of HKM containing no sucrose. The sample was centrifuged at 55 000 r.p.m. (240 000×g) in a Beckman swinging rotor for 1 h. The bottom (250 µl), middle (150 µl) and top (100 µl) fractions were manually collected from the bottom. The bottom and top fractions were analyzed by SDS–PAGE using Coomassie staining. The flotation of liposomes on the sucrose gradient after centrifugation and their separation in the collected fractions was checked by following the NBD-PE fluorescence using a ChemiDoc Imaging System (Biorad). All experiments were done in duplicate.

Nucleotide exchange assays

Nucleotide exchange kinetics were monitored by fluorescence on a Cary Eclipse fluorimeter (Varian) under stirring. In all experiments, proteins and liposomes were preincubated for 1 min under stirring before the exchange reaction was initiated by the addition of 100 µM GTP. When indicated, His-tagged GTPases were dissociated from Ni-lipid-containing liposomes by 250 mM imidazole. kobs values were determined from a mono-exponential fit. Values of kcat/Km were determined by linear regression from kobs values measured over a range of GEF concentrations following the Michaelis–Menten formalism as described previously [48]. For Arf GTPases, we monitored the increase in tryptophan fluorescence associated with the conformational changes induced by GTP binding, using excitation and emission wavelengths of 290 and 340 nm, as described in refs [7,26]. The reaction was performed at 37°C in HKM buffer containing 500 nM Arf GTPases and 10 nM GEF with or without 100 µM liposomes (lipids). For Rho and Rab GTPases, nucleotide exchange kinetics were monitored by the decay of mant-GDP fluorescence following its replacement by GTP, using excitation and emission wavelengths of 360 and 440 nm, respectively. Small GTPases were loaded with mant-GDP prior to nucleotide exchange by incubation of 250 µM GTPase with 1.5 mM mant-GDP and 10 mM EDTA for 30 min at room temperature. Nucleotide exchange was stopped by the addition of 75 mM MgCl2. Removal of excess nucleotides and buffer exchange was done on a PD SpinTrap G-25 (GE HealthCare Life Science). Nucleotide exchange was carried out at 30°C (Rho GTPases) or 37°C (Rab GTPases) in HKM buffer using 500 nM Rho or Rab GTPases loaded with mant-GDP and either 10 nM Trio, 15 nM DrrA or 15 nM DennD, with or without 100 µM liposomes. All experiments were done in triplicate.

Abbreviations

     
  • BRAG2

    Brefeldin-resistant ArfGEF 2

  •  
  • DH

    Dbl-homology

  •  
  • DLS

    dynamic light scattering

  •  
  • EDTA

    ethylenediaminetetraacetic acid

  •  
  • EFA6

    Exchange factor for Arf6

  •  
  • ER

    endoplasmic reticulum

  •  
  • GAPs

    GTPase-activating proteins

  •  
  • GDI

    guanosine nucleotide dissociation inhibitor

  •  
  • GEFs

    guanine nucleotide exchange factors

  •  
  • His-tag

    histidine tag

  •  
  • IPTG

    isopropyl β-D-1-thiogalactopyranoside

  •  
  • mant-GDP

    N-methylanthraniloyl-GDP

  •  
  • NBD-PE

    N-(7-nitrobenz-2-Oxa-1,3-Diazol-4-yl)-1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine

  •  
  • Ni-lipids

    DGS-NTA-Ni, 1,2-dioleoyl-sn-glycero-3-[(N-(5-amino-1-carboxypentyl)iminodiacetic acid)succinyl] -nickel salt

  •  
  • PC

    phosphatidylcholine

  •  
  • PE

    phosphatidylethanolamine

  •  
  • PH

    pleckstrin homology

  •  
  • PI(4)P

    phosphatidylinositol 4-phosphate

  •  
  • PI(4,5)P2

    PIP2, phosphatidylinositol 4,5 bisphosphate

  •  
  • PS

    phosphatidylserine

  •  
  • TEV

    tobacco etch virus.

Author Contribution

F.P. designed and performed experiments, analyzed data and drafted the manuscript. S.V. designed and performed experiments and analyzed data. Y.F. designed and performed experiments and analyzed data. I.L. performed experiments. S.B. performed experiments. M.Z. discussed data. G.P. analyzed data and supervised experiments. J.C. conceived and supervised the study and wrote the manuscript with input from the other authors.

Funding

This work was supported by grants to J.C. from Institut National du Cancer [INCA_7886], Fondation pour la Recherche Médicale [DEQ20150331694] and Agence Nationale de la Recherche [ANR-11-BSV1-0006 and ANR-14-CE09-0028] to F.P. from Ecole Normale Supérieure Paris-Saclay and to S.V. by Région Ile-de-France/DIM MALINF.

Acknowledgments

We thank Lionel Duarte, Anne-Gaëlle Planson, Théo Desbordes, David Rémy and Raphaëlle Servant (CNRS and ENS Paris-Saclay) for preliminary experiments.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Cherfils
,
J.
and
Zeghouf
,
M.
(
2011
)
Chronicles of the GTPase switch
.
Nat. Chem. Biol.
7
,
493
495
doi:
2
Wang
,
M.
and
Casey
,
P.J.
(
2016
)
Protein prenylation: unique fats make their mark on biology
.
Nat. Rev. Mol. Cell Biol.
17
,
110
122
doi:
3
Cherfils
,
J.
and
Zeghouf
,
M.
(
2013
)
Regulation of small GTPases by GEFs, GAPs, and GDIs
.
Physiol. Rev.
93
,
269
309
doi:
4
Liu
,
Y.
,
Kahn
,
R.A.
and
Prestegard
,
J.H.
(
2009
)
Structure and membrane interaction of myristoylated ARF1
.
Structure
17
,
79
87
doi:
5
Pasqualato
,
S.
,
Renault
,
L.
and
Cherfils
,
J.
(
2002
)
Arf, Arl, Arp and Sar proteins: a family of GTP-binding proteins with a structural device for ‘front-back’ communication
.
EMBO Rep.
3
,
1035
1041
doi:
6
Chardin
,
P.
,
Paris
,
S.
,
Antonny
,
B.
,
Robineau
,
S.
,
Béraud-Dufour
,
S.
,
Jackson
,
C.L.
et al. 
(
1996
)
A human exchange factor for ARF contains Sec7- and pleckstrin-homology domains
.
Nature
384
,
481
484
doi:
7
Aizel
,
K.
,
Biou
,
V.
,
Navaza
,
J.
,
Duarte
,
L.V.
,
Campanacci
,
V.
,
Cherfils
,
J.
et al. 
(
2013
)
Integrated conformational and lipid-sensing regulation of endosomal ArfGEF BRAG2
.
PLoS Biol.
11
,
e1001652
doi:
8
Jian
,
X.
,
Gruschus
,
J.M.
,
Sztul
,
E.
and
Randazzo
,
P.A.
(
2012
)
The pleckstrin homology (PH) domain of the Arf exchange factor Brag2 is an allosteric binding site
.
J. Biol. Chem.
287
,
24273
24283
doi:
9
Folly-Klan
,
M.
,
Alix
,
E.
,
Stalder
,
D.
,
Ray
,
P.
,
Duarte
,
L.V.
,
Delprato
,
A.
et al. 
(
2013
)
A novel membrane sensor controls the localization and ArfGEF activity of bacterial RalF
.
PLoS Pathog.
9
,
e1003747
doi:
10
Padovani
,
D.
,
Folly-Klan
,
M.
,
Labarde
,
A.
,
Boulakirba
,
S.
,
Campanacci
,
V.
,
Franco
,
M.
et al. 
(
2014
)
EFA6 controls Arf1 and Arf6 activation through a negative feedback loop
.
Proc. Natl Acad. Sci. U.S.A.
111
,
12378
12383
doi:
11
Franco
,
M.
,
Chardin
,
P.
,
Chabre
,
M.
and
Paris
,
S.
(
1996
)
Myristoylation-facilitated binding of the G protein ARF1GDP to membrane phospholipids is required for its activation by a soluble nucleotide exchange factor
.
J. Biol. Chem.
271
,
1573
1578
doi:
12
Ha
,
V.L.
,
Thomas
,
G.M.H.
,
Stauffer
,
S.
and
Randazzo
,
P.A.
(
2005
)
Preparation of myristoylated Arf1 and Arf6
.
Methods Enzymol.
404
,
164
174
doi:
13
Glück
,
J.M.
,
Hoffmann
,
S.
,
Koenig
,
B.W.
and
Willbold
,
D.
(
2010
)
Single vector system for efficient N-myristoylation of recombinant proteins in E. coli
.
PLoS ONE
5
,
e10081
doi:
14
Padovani
,
D.
,
Zeghouf
,
M.
,
Traverso
,
J.A.
,
Giglione
,
C.
and
Cherfils
,
J.
(
2013
)
High yield production of myristoylated Arf6 small GTPase by recombinant N-myristoyl transferase
.
Small GTPases
4
,
3
8
doi:
15
Grizot
,
S.
,
Fauré
,
J.
,
Fieschi
,
F.
,
Vignais
,
P.V.
,
Dagher
,
M.-C.
and
Pebay-Peyroula
,
E.
(
2001
)
Crystal structure of the Rac1–RhoGDI complex involved in NADPH oxidase activation
.
Biochemistry
40
,
10007
10013
doi:
16
Scheffzek
,
K.
,
Stephan
,
I.
,
Jensen
,
O.N.
,
Illenberger
,
D.
and
Gierschik
,
P.
(
2000
)
The Rac-RhoGDI complex and the structural basis for the regulation of Rho proteins by RhoGDI
.
Nat. Struct. Biol.
7
,
122
126
doi:
17
Hoffman
,
G.R.
,
Nassar
,
N.
and
Cerione
,
R.A.
(
2000
)
Structure of the Rho family GTP-binding protein Cdc42 in complex with the multifunctional regulator RhoGDI
.
Cell
100
,
345
356
doi:
18
Robbe
,
K.
,
Otto-Bruc
,
A.
,
Chardin
,
P.
and
Antonny
,
B.
(
2003
)
Dissociation of GDP dissociation inhibitor and membrane translocation are required for efficient activation of Rac by the Dbl homology-pleckstrin homology region of Tiam
.
J. Biol. Chem.
278
,
4756
4762
doi:
19
Rak
,
A.
,
Pylypenko
,
O.
,
Niculae
,
A.
,
Goody
,
R.S.
and
Alexandrov
,
K.
(
2003
)
Crystallization and preliminary X-ray diffraction analysis of monoprenylated Rab7 GTPase in complex with Rab escort protein 1
.
J. Struct. Biol.
141
,
93
95
doi:
20
Bader
,
B.
,
Kuhn
,
K.
,
Owen
,
D.J.
,
Waldmann
,
H.
,
Wittinghofer
,
A.
and
Kuhlmann
,
J.
(
2000
)
Bioorganic synthesis of lipid-modified proteins for the study of signal transduction
.
Nature
403
,
223
226
doi:
21
Mejuch
,
T.
and
Waldmann
,
H.
(
2016
)
Synthesis of lipidated proteins
.
Bioconjug. Chem.
27
,
1771
1783
doi:
22
Gureasko
,
J.
,
Galush
,
W.J.
,
Boykevisch
,
S.
,
Sondermann
,
H.
,
Bar-Sagi
,
D.
,
Groves
,
J.T.
et al. 
(
2008
)
Membrane-dependent signal integration by the Ras activator Son of sevenless
.
Nat. Struct. Mol. Biol.
15
,
452
461
doi:
23
Medina
,
F.
,
Carter
,
A.M.
,
Dada
,
O.
,
Gutowski
,
S.
,
Hadas
,
J.
,
Chen
,
Z.
et al. 
(
2013
)
Activated RhoA is a positive feedback regulator of the Lbc family of Rho guanine nucleotide exchange factor proteins
.
J. Biol. Chem.
288
,
11325
11333
doi:
24
Cabrera
,
M.
,
Nordmann
,
M.
,
Perz
,
A.
,
Schmedt
,
D.
,
Gerondopoulos
,
A.
,
Barr
,
F.
et al. 
(
2014
)
The Mon1-Ccz1 GEF activates the Rab7 GTPase Ypt7 via a longin-fold-Rab interface and association with PI3P-positive membranes
.
J. Cell. Sci.
127
(
Pt 5
),
1043
1051
doi:
25
Thomas
,
L.L.
and
Fromme
,
J.C.
(
2016
)
GTPase cross talk regulates TRAPPII activation of Rab11 homologues during vesicle biogenesis
.
J. Cell Biol.
215
,
499
513
doi:
26
Nawrotek
,
A.
,
Zeghouf
,
M.
and
Cherfils
,
J.
(
2016
)
Allosteric regulation of Arf GTPases and their GEFs at the membrane interface
.
Small GTPases
7
,
283
296
doi:
27
Jackson
,
C.L.
and
Bouvet
,
S.
(
2014
)
Arfs at a glance
.
J. Cell Sci.
127
(
Pt 19
),
4103
4109
doi:
28
DiNitto
,
J.P.
,
Delprato
,
A.
,
Gabe Lee
,
M.-T.
,
Cronin
,
T.C.
,
Huang
,
S.
,
Guilherme
,
A.
et al. 
(
2007
)
Structural basis and mechanism of autoregulation in 3-phosphoinositide-dependent Grp1 family Arf GTPase exchange factors
.
Mol. Cell
28
,
569
583
doi:
29
Stalder
,
D.
,
Barelli
,
H.
,
Gautier
,
R.
,
Macia
,
E.
,
Jackson
,
C.L.
and
Antonny
,
B.
(
2011
)
Kinetic studies of the Arf activator Arno on model membranes in the presence of Arf effectors suggest control by a positive feedback loop
.
J. Biol. Chem.
286
,
3873
3883
doi:
30
Casanova
,
J.E.
(
2007
)
Regulation of Arf activation: the Sec7 family of guanine nucleotide exchange factors
.
Traffic
8
,
1476
1485
doi:
31
Heasman
,
S.J.
and
Ridley
,
A.J.
(
2008
)
Mammalian Rho GTPases: new insights into their functions from in vivo studies
.
Nat. Rev. Mol. Cell Biol.
9
,
690
701
doi:
32
Dransart
,
E.
,
Olofsson
,
B.
and
Cherfils
,
J.
(
2005
)
RhoGDIs revisited: novel roles in Rho regulation
.
Traffic
6
,
957
966
doi:
33
Schmidt
,
S.
and
Debant
,
A.
(
2014
)
Function and regulation of the Rho guanine nucleotide exchange factor Trio
.
Small GTPases
5
,
e983880
doi:
34
Skowronek
,
K.R.
,
Guo
,
F.
,
Zheng
,
Y.
and
Nassar
,
N.
(
2004
)
The C-terminal basic tail of RhoG assists the guanine nucleotide exchange factor trio in binding to phospholipids
.
J. Biol. Chem.
279
,
37895
37907
doi:
35
Blangy
,
A.
,
Vignal
,
E.
,
Schmidt
,
S.
,
Debant
,
A.
,
Gauthier-Rouviere
,
C.
and
Fort
,
P.
(
2000
)
TrioGEF1 controls Rac- and Cdc42-dependent cell structures through the direct activation of rhoG
.
J. Cell Sci.
113
(
Pt 4
),
729
739
PMID:
[PubMed]
36
Chhatriwala
,
M.K.
,
Betts
,
L.
,
Worthylake
,
D.K.
and
Sondek
,
J.
(
2007
)
The DH and PH domains of Trio coordinately engage Rho GTPases for their efficient activation
.
J. Mol. Biol.
368
,
1307
1320
doi:
37
Zhen
,
Y.
and
Stenmark
,
H.
(
2015
)
Cellular functions of Rab GTPases at a glance
.
J. Cell Sci.
128
,
3171
3176
doi:
38
Yang
,
X.-Z.
,
Li
,
X.-X.
,
Zhang
,
Y.-J.
,
Rodriguez-Rodriguez
,
L.
,
Xiang
,
M.-Q.
,
Wang
,
H.-Y.
et al. 
(
2016
)
Rab1 in cell signaling, cancer and other diseases
.
Oncogene
35
,
5699
5704
doi:
39
Murata
,
T.
,
Delprato
,
A.
,
Ingmundson
,
A.
,
Toomre
,
D.K.
,
Lambright
,
D.G.
and
Roy
,
C.R.
(
2006
)
The Legionella pneumophila effector protein DrrA is a Rab1 guanine nucleotide-exchange factor
.
Nat. Cell Biol.
8
,
971
977
doi:
40
Schoebel
,
S.
,
Blankenfeldt
,
W.
,
Goody
,
R.S.
and
Itzen
,
A.
(
2010
)
High-affinity binding of phosphatidylinositol 4-phosphate by Legionella pneumophila DrrA
.
EMBO Rep.
11
,
598
604
doi:
41
Del Campo
,
C.M.
,
Mishra
,
A.K.
,
Wang
,
Y.-H.
,
Roy
,
C.R.
,
Janmey
,
P.A.
and
Lambright
,
D.G.
(
2014
)
Structural basis for PI(4)P-specific membrane recruitment of the Legionella pneumophila effector DrrA/SidM
.
Structure
22
,
397
408
doi:
42
Klinkert
,
K.
and
Echard
,
A.
(
2016
)
Rab35 GTPase: a central regulator of phosphoinositides and F-actin in endocytic recycling and beyond
.
Traffic
17
,
1063
1077
doi:
43
Mukherjee
,
S.
,
Liu
,
X.
,
Arasaki
,
K.
,
McDonough
,
J.
,
Galán
,
J.E.
and
Roy
,
C.R.
(
2011
)
Modulation of Rab GTPase function by a protein phosphocholine transferase
.
Nature
477
,
103
106
doi:
44
Jian
,
X.
,
Cavenagh
,
M.
,
Gruschus
,
J.M.
,
Randazzo
,
P.A.
and
Kahn
,
R.A.
(
2010
)
Modifications to the C-terminus of Arf1 alter cell functions and protein interactions
.
Traffic
11
,
732
742
doi:
45
Volpicelli-Daley
,
L.A.
,
Li
,
Y.
,
Zhang
,
C.J.
and
Kahn
,
R.A.
(
2005
)
Isoform-selective effects of the depletion of ADP-ribosylation factors 1-5 on membrane traffic
.
Mol. Biol. Cell
16
,
4495
4508
doi:
46
Vives
,
V.
,
Cres
,
G.
,
Richard
,
C.
,
Busson
,
M.
,
Ferrandez
,
Y.
,
Planson
,
A.-G.
et al. 
(
2015
)
Pharmacological inhibition of Dock5 prevents osteolysis by affecting osteoclast podosome organization while preserving bone formation
.
Nat. Commun.
6
,
6218
doi:
47
Wu
,
X.
,
Bradley
,
M.J.
,
Cai
,
Y.
,
Kummel
,
D.
,
De La Cruz
,
E.M.
,
Barr
,
F.A.
et al. 
(
2011
)
Insights regarding guanine nucleotide exchange from the structure of a DENN-domain protein complexed with its Rab GTPase substrate
.
Proc. Natl Acad. Sci. U.S.A.
108
,
18672
18677
doi:
48
Beraud-Dufour
,
S.
,
Robineau
,
S.
,
Chardin
,
P.
,
Paris
,
S.
,
Chabre
,
M.
,
Cherfils
,
J.
et al. 
(
1998
)
A glutamic finger in the guanine nucleotide exchange factor ARNO displaces Mg2+ and the beta-phosphate to destabilize GDP on ARF1
.
EMBO J.
17
,
3651
3659
doi: