Role of Protein-Phospholipid Interactions in the Activation of ARF1 by the Guanine Nucleotide Exchange Factor Arno

Abstract

Arno is a 47-kDa human protein recently identified as a guanine nucleotide exchange factor for ADP ribosylation factor 1 (ARF1) with a central Sec7 domain responsible for the exchange activity and a carboxyl-terminal pleckstrin homology (PH) domain (Chardin, P., Paris, S., Antonny, B., Robineau, S., Béraud-Dufour, S., Jackson, C. L., and Chabre, M. (1996)Nature 384, 481–484). Binding of the PH domain to phosphatidylinositol 4,5-bisphosphate (PIP2) greatly enhances Arno-mediated activation of myristoylated ARF1. We show here that in the absence of phospholipids, Arno promotes nucleotide exchange on [Δ17]ARF1, a soluble mutant of ARF1 lacking the first 17 amino acids. This reaction is unaffected by PIP2, which suggests that the PIP2-PH domain interaction does not directly regulate the catalytic activity of Arno but rather serves to recruit Arno to membranes. Arno catalyzes the release of GDP more efficiently than that of GTP from [Δ17]ARF1, and a stable complex between Arno Sec7 domain and nucleotide-free [Δ17]ARF1 can be isolated. In contrast to [Δ17]ARF1, full-length unmyristoylated ARF1 is not readily activated by Arno in solution. Its activation requires the presence of phospholipids and a reduction of ionic strength and Mg2+ concentration. PIP2 is strongly stimulatory, indicating that binding of Arno to phospholipids is involved, but in addition, electrostatic interactions between phospholipids and the amino-terminal portion of unmyristoylated ARF1GDP seem to be important.We conclude that efficient activation of full-length ARF1 by Arno requires a membrane surface and two distinct protein-phospholipid interactions: one between the PH domain of Arno and PIP2, and the other between amino-terminal cationic residues of ARF1 and anionic phospholipids. The latter interaction is normally induced by insertion of the amino-terminal myristate into the bilayer but can also be artificially facilitated by decreasing Mg2+ and salt concentrations. Arno is a 47-kDa human protein recently identified as a guanine nucleotide exchange factor for ADP ribosylation factor 1 (ARF1) with a central Sec7 domain responsible for the exchange activity and a carboxyl-terminal pleckstrin homology (PH) domain (Chardin, P., Paris, S., Antonny, B., Robineau, S., Béraud-Dufour, S., Jackson, C. L., and Chabre, M. (1996)Nature 384, 481–484). Binding of the PH domain to phosphatidylinositol 4,5-bisphosphate (PIP2) greatly enhances Arno-mediated activation of myristoylated ARF1. We show here that in the absence of phospholipids, Arno promotes nucleotide exchange on [Δ17]ARF1, a soluble mutant of ARF1 lacking the first 17 amino acids. This reaction is unaffected by PIP2, which suggests that the PIP2-PH domain interaction does not directly regulate the catalytic activity of Arno but rather serves to recruit Arno to membranes. Arno catalyzes the release of GDP more efficiently than that of GTP from [Δ17]ARF1, and a stable complex between Arno Sec7 domain and nucleotide-free [Δ17]ARF1 can be isolated. In contrast to [Δ17]ARF1, full-length unmyristoylated ARF1 is not readily activated by Arno in solution. Its activation requires the presence of phospholipids and a reduction of ionic strength and Mg2+ concentration. PIP2 is strongly stimulatory, indicating that binding of Arno to phospholipids is involved, but in addition, electrostatic interactions between phospholipids and the amino-terminal portion of unmyristoylated ARF1GDP seem to be important. We conclude that efficient activation of full-length ARF1 by Arno requires a membrane surface and two distinct protein-phospholipid interactions: one between the PH domain of Arno and PIP2, and the other between amino-terminal cationic residues of ARF1 and anionic phospholipids. The latter interaction is normally induced by insertion of the amino-terminal myristate into the bilayer but can also be artificially facilitated by decreasing Mg2+ and salt concentrations. ADP-ribosylation factors (ARFs), 1The abbreviations used are: ARF, ADP ribosylation factor; PH, pleckstrin homology; myrARF1, myristoylated ARF1; unmyrARF1, unmyristoylated ARF1; PIP2, phosphatidylinositol 4,5-bisphosphate; GTPγS, guanosine 5′-3-O-(thio)triphosphate; PC, phosphatidylcholine; PG, phosphatidylglycerol. which were originally identified as cofactors for cholera toxin, are small GTP-binding proteins involved in intracellular vesicular transport (1Donaldson J.G. Klausner R.D. Curr. Opin. Cell Biol. 1994; 6: 527-532Crossref PubMed Scopus (232) Google Scholar,2Moss J. Vaughan M. J. Biol. Chem. 1995; 270: 12327-12330Abstract Full Text Full Text PDF PubMed Scopus (272) Google Scholar). Their functions include acting as regulators of the binding of coat proteins and adaptins to intracellular membranes (3Rothman J.E. Nature. 1994; 372: 55-62Crossref PubMed Scopus (2011) Google Scholar, 4Robinson M.S. Trends Cell Biol. 1997; 7: 99-102Abstract Full Text PDF PubMed Scopus (123) Google Scholar) and as activators of phospholipase D (5Brown H.A. Gutowski S. Moomaw C.R. Slaughter C. Sternweis P.C. Cell. 1993; 75: 1137-1144Abstract Full Text PDF PubMed Scopus (823) Google Scholar, 6Cockroft S. Thomas G.M.H. Fensome A. Geny B. Cunningham E. Gout I. Hiles I. Totty N.F. Truong O. Hsuan J.J. Science. 1994; 263: 523-526Crossref PubMed Scopus (586) Google Scholar, 7Frohman M.A. Morris A.J. Curr. Biol. 1996; 6: 945-947Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar). It is still unclear whether phospholipase D mediates ARF signals to initiate coated vesicle formation (8Ktistakis N.T. Brown H.A. Waters M.G. Sternweis P.C. Roth M.G. J. Cell Biol. 1996; 134: 295-306Crossref PubMed Scopus (329) Google Scholar, 9Zhao L. Helms J.B. Brügger B. Harter C. Martoglio B. Graf R. Brunner J. Wieland F.T. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4418-4423Crossref PubMed Scopus (122) Google Scholar). Like other G proteins, ARFs cycle between inactive GDP-bound and active GTP-bound conformations, and they are thought to shuttle during the activation cycle between cytosol and membranes, with the possible exception of ARF6 (10Cavenagh M.M. Whitney J.A. Carroll K. Zhang C. Boman A.L. Rosenwald A.G. Mellman I. Kahn R.A. J. Biol. Chem. 1996; 271: 21767-21774Abstract Full Text Full Text PDF PubMed Scopus (206) Google Scholar). Taking ARF1, the most abundant ARF family member, as a model, we previously assessed the importance ofN-myristoylation for GTP-dependent binding to phospholipid vesicles in vitro (11Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1993; 268: 24531-24534Abstract Full Text PDF PubMed Google Scholar, 12Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1995; 270: 1337-1341Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar). We concluded that the increased affinity of ARFGTP for membranes cannot be ascribed to a myristoyl switch, i.e. a nucleotide exchange-dependent exposure of the myristate, as frequently proposed in reviews (3Rothman J.E. Nature. 1994; 372: 55-62Crossref PubMed Scopus (2011) Google Scholar, 13Harter C. Wieland F. Biochim. Biophys. Acta. 1996; 1286: 75-93Crossref PubMed Scopus (49) Google Scholar). We believe that in ARFGDP, the myristate is accessible for interacting with phospholipids and that this interaction mostly accounts for the weak binding of ARFGDP to membranes. The tight binding of ARFGTP to phospholipids can be explained by a GTP-dependent release of the amino-terminal helix from the protein core, leading to the exposure of several hydrophobic residues that insert into the lipids in addition to the myristate (14Antonny B. Béraud-Dufour S. Chardin P. Chabre M. Biochemistry. 1997; 36: 4675-4685Crossref PubMed Scopus (281) Google Scholar). Conversion of ARFGDP to ARFGTP is promotedin vivo by a guanine nucleotide exchange factor. Complementation studies in yeast recently led to the discovery of a family of ARF nucleotide exchange factors that share a region of sequence homology with Sec7, a yeast protein necessary for intra-Golgi transport (15Peyroche A. Paris S. Jackson C.L. Nature. 1996; 384: 479-481Crossref PubMed Scopus (236) Google Scholar). We identified as a member of this family an ubiquitously expressed ∼47-kDa human protein termed Arno (forARF nucleotide-binding site opener) (16Chardin P. Paris S. Antonny B. Robineau S. Béraud-Dufour S. Jackson C.L. Chabre M. Nature. 1996; 384: 481-484Crossref PubMed Scopus (409) Google Scholar). Another protein of very similar sequence but mostly expressed in hematopoietic cells (17Liu L. Pohajdak B. Biochim. Biophys. Acta. 1992; 1132: 75-78Crossref PubMed Scopus (64) Google Scholar) turned out to be identical to cytohesin-1, a protein described as a regulator of β2 integrin in lymphocytes (18Kolanus W. Nagel W. Schiller B. Zeitlmann L. Godar S. Stockinger H. Seed B. Cell. 1996; 86: 233-242Abstract Full Text Full Text PDF PubMed Scopus (401) Google Scholar). Cytohesin-1 also promotes nucleotide exchange on ARF1 (16Chardin P. Paris S. Antonny B. Robineau S. Béraud-Dufour S. Jackson C.L. Chabre M. Nature. 1996; 384: 481-484Crossref PubMed Scopus (409) Google Scholar) as well as on ARF3 (19Meacci E. Tsai S.-C. Adamik R. Moss J. Vaughan M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 1745-1748Crossref PubMed Scopus (136) Google Scholar). Arno and cytohesin-1 contain an amino-terminal coiled-coil motif, a central Sec7 domain, and a pleckstrin homology (PH) domain followed by a cluster of positively charged residues at the carboxyl terminus. Optimal exchange activity on myrARF requires the presence of negatively charged phospholipid vesicles supplemented with PIP2. Studies with truncated mutants of Arno have demonstrated that the Sec7 domain is responsible for the exchange activity, whereas the PH domain binds to PIP2 (16Chardin P. Paris S. Antonny B. Robineau S. Béraud-Dufour S. Jackson C.L. Chabre M. Nature. 1996; 384: 481-484Crossref PubMed Scopus (409) Google Scholar). In the present study, we have addressed two questions concerning the activation of ARF1 by Arno. (a) What is the role of the binding of the PH domain to PIP2: does it serve only to recruit Arno to membranes, or is it required to activate the exchange factor? (b) What is the role of the amino-terminal helix of ARF1 in the interaction ARF-Arno? Egg phosphatidylcholine, egg phosphatidylglycerol, and bovine brain PIP2 were purchased from Sigma. [35S]GTPγS was from NEN Life Science Products, [3H]GDP was from Amersham Corp., and unlabeled nucleotides were from Boehringer Mannheim. Bovine recombinant ARF1 was expressed in Escherichia coli and purified by a single QAE-Sepharose chromatography as described previously (11Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1993; 268: 24531-24534Abstract Full Text PDF PubMed Google Scholar). Myristoylated ARF1 was made by co-expression with a yeast N-myristoyltransferase (20Duronio R.J. Jackson-Machelski E. Heuckeroth R.O. Olins P.O. Devine C.S. Yonemoto W. Slice L.W. Taylor S.S. Gordon J.I. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 1506-1510Crossref PubMed Scopus (201) Google Scholar) and separated from the contaminating nonmyristoylated protein by a precipitation at 35% saturation of ammonium sulfate followed by sequential chromatography on DEAE-Sepharose and Mono-S columns (12Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1995; 270: 1337-1341Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar). The mutant [Δ17]ARF1, which lacks the first 17 amino acids, was purified by gel filtration with Ultrogel AcA44 (14Antonny B. Béraud-Dufour S. Chardin P. Chabre M. Biochemistry. 1997; 36: 4675-4685Crossref PubMed Scopus (281) Google Scholar). Wild-type Arno, ΔPH Arno (a mutant lacking the PH and Arno Sec7 domain were by exchange on QAE-Sepharose and gel filtration on as described previously (16Chardin P. Paris S. Antonny B. Robineau S. Béraud-Dufour S. Jackson C.L. Chabre M. Nature. 1996; 384: 481-484Crossref PubMed Scopus (409) Google Scholar). by gel was for ARF proteins and was between and for Arno phospholipid vesicles were by the of Biochim. Biophys. Acta. PubMed Scopus Google Scholar). of phospholipids of the in and PIP2 was in a and at in with or as The was for and vesicles were by (14Antonny B. Béraud-Dufour S. Chardin P. Chabre M. Biochemistry. 1997; 36: 4675-4685Crossref PubMed Scopus (281) Google Scholar). or truncated ARF was at with [35S]GTPγS or [3H]GDP in 1 1 with or in the or 1 and 1 phospholipid the of of were into of and and on were with of the and [Δ17]ARF1 and Arno Sec7 domain were or for at in and 1 with or or 1 of was to a and with the at a of of were and were and by gel previously (16Chardin P. Paris S. Antonny B. Robineau S. Béraud-Dufour S. Jackson C.L. Chabre M. Nature. 1996; 384: 481-484Crossref PubMed Scopus (409) Google nucleotide exchange activity of Arno on requires the presence of vesicles in the 1 The of PIP2 is most to an interaction between this phospholipid and the PH domain of Arno (16Chardin P. Paris S. Antonny B. Robineau S. Béraud-Dufour S. Jackson C.L. Chabre M. Nature. 1996; 384: 481-484Crossref PubMed Scopus (409) Google but two possible can be The binding of the PH domain to PIP2 Arno on the surface of vesicles and interaction with the of binding to PIP2 a of Arno, leading to an increased catalytic activity, for by a between we the of Arno on [Δ17]ARF1, a soluble this mutant of ARF1 lacking the 17 amino-terminal amino the for phospholipids in the nucleotide exchange reaction R.A. P. O. C. J. M. P. L. J.E. J. Biol. Chem. 1992; Full Text PDF PubMed Google Scholar) and soluble in the GTP-bound as demonstrated by (14Antonny B. Béraud-Dufour S. Chardin P. Chabre M. Biochemistry. 1997; 36: 4675-4685Crossref PubMed Scopus (281) Google Scholar) and by a with phospholipid vesicles not This of the exchange in the absence of phospholipids. 1 that Arno is active on [Δ17]ARF1, which first that the amino-terminal of ARF1 is not for the interaction the exchange activity of Arno on [Δ17]ARF1 was the in the absence of phospholipids and in the presence of a similar activity was with ΔPH Arno, a mutant lacking the PH domain, or Arno Sec7 domain, a mutant to the central catalytic domain 1 that in full-length Arno, is on the catalytic site that be by the binding of the PH domain to the interaction of the PH domain with not regulate the activity of Arno but mediates membrane The that Arno Sec7 domain catalyzes nucleotide exchange on [Δ17]ARF1 1 to whether the interaction between two soluble proteins be demonstrated by gel filtration that this is the a of Arno Sec7 domain and [Δ17]ARF1 was to and from a the two proteins were to be at 1 Mg2+ and at 1 Mg2+ of GDP to the at 1 Mg2+ the interaction between the two proteins which that Arno Sec7 domain a stable complex with the nucleotide-free of We previously that a soluble nucleotide exchange factor present in a can release as well as GDP release from ARF1 M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1996; 271: Full Text Full Text PDF PubMed Scopus Google Scholar). We whether purified Arno is also to the exchange reaction in This was first in the absence of phospholipids with [Δ17]ARF1 as a in Arno GDP and from [Δ17]ARF1, but the is more on GDP of Arno to 1 [Δ17]ARF1 increased the of GDP and increased the of only and this was not by the of the nucleotide Arno to have in a affinity for ARFGDP than it does for This is the nucleotide exchange is with in the presence of vesicles GDP and seem to be to a similar by Arno, but this is to the that the reaction on the membrane with phospholipids are from the the of [3H]GDP release from by Arno only of the exchange factor not This that only the of is a for Arno, and is mostly this be very as with which is to phospholipids. binding of the proteins to membranes greatly the This is also by the that in GDP and are as [3H]GDP and [35S]GTPγS from were more by a catalytic of Arno GDP was used as the nucleotide This suggests that the Arno more from than it does from which be to the affinity of for membranes. on the other [35S]GTPγS release was in the presence of GDP It not a as with and Arno was acting on myrARF1, of a by the GDP-bound of that Arno a affinity for ARFGDP, Arno more from than it does from the is on of Arno on the of [3H]GDP and [35S]GTPγS from in the presence of 1 was first with in the presence of for 1 at or with [35S]GTPγS in the presence of 1 and for at In the also and 1 Mg2+ was to 1 and 1 unlabeled GDP or was with or Arno at to of [3H]GDP and of of We previously that Arno is inactive on at 1 Mg2+ in the presence of and vesicles (16Chardin P. Paris S. Antonny B. Robineau S. Béraud-Dufour S. Jackson C.L. Chabre M. Nature. 1996; 384: 481-484Crossref PubMed Scopus (409) Google Scholar). of in PIP2 is not by phospholipids of the of a small activation of binding to on the addition of a of Arno but this as with that on or this be greatly by the of the (a) decreasing the ionic strength by of led to the of a of activated at 1 and (b) the of Mg2+ to 1 the activated We have that 1 Mg2+ exchange on (11Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1993; 268: 24531-24534Abstract Full Text PDF PubMed Google Scholar). Mg2+ that nucleotide exchange seem to also the activation by Arno, whereas decreasing the salt ionic interactions between the two proteins and phospholipids. the Mg2+ and salt we the of phospholipids on the of binding to by Arno and Arno Sec7 In the absence of phospholipids [35S]GTPγS binding to 1 was by Arno and more by Sec7 In the of exchange was a of the exchange factor and was identical for full-length Arno and Arno Sec7 domain in the absence of phospholipids, Arno and the Sec7 domain are but are very In the presence of the by Sec7 domain was only but Arno more active Arno was to domain of PIP2 in the vesicles not the by the Sec7 domain but increased the of Arno is soluble (12Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1995; 270: 1337-1341Abstract Full Text Full Text PDF PubMed Scopus (147) Google activation by Arno is greatly facilitated by the binding of Arno to which is in contrast with [Δ17]ARF1 1 The of PIP2 on the Sec7 domain was the PH domain is in this mutant In the Sec7 domain is the small in activity lipids or with lipids cannot be to an interaction between the Sec7 domain and phospholipids. It be ascribed to We previously that phospholipids the active of and the binding of [35S]GTPγS (11Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1993; 268: 24531-24534Abstract Full Text PDF PubMed Google Scholar). whether this is the for the we the by [3H]GDP the activation by the Sec7 domain in the presence of phospholipids than it was in absence exchange of the was in the presence of This that interaction of with phospholipids activation by the soluble truncated exchange whereas it does not the nucleotide at It can be that and interactions a role in the activation of by Arno. The present that Arno is to guanine nucleotide exchange on [Δ17]ARF1, a soluble truncated mutant of ARF1, in the absence of phospholipids. The exchange activity is unaffected by the addition of phospholipid vesicles of PC, PG, and PIP2, vesicles that have to Arno (16Chardin P. Paris S. Antonny B. Robineau S. Béraud-Dufour S. Jackson C.L. Chabre M. Nature. 1996; 384: 481-484Crossref PubMed Scopus (409) Google Scholar). a activity is with an of Arno a mutant of Arno lacking the PH domain, or of Arno Sec7 domain, a mutant to the central catalytic that in Arno the catalytic site is accessible to the soluble whether Arno is or in solution. the binding of the PH domain to vesicles does not the catalytic activity of Arno but promotes membrane of the exchange factor. The role of PIP2 binding for a of other PH In two the K. E. Gout I. R. R. P.C. J. 1996; PubMed Scopus Google Scholar) and the Cell. 1997; Full Text Full Text PDF PubMed Scopus Google the binding of proposed to directly regulate the catalytic activity, in the latter the of this recently by the of a protein that and P. Curr. Biol. 1997; 7: Full Text Full Text PDF PubMed Google Scholar). frequently the PH domain is thought to have a by of the in of the This demonstrated for G J.A. R.J. J. Biol. Chem. 1996; 271: Full Text Full Text PDF PubMed Scopus Google Scholar) and and of phospholipase P. R. S. Morris A.J. S. S. Biochemistry. 1995; PubMed Scopus Google Scholar, A. J. Biol. Chem. 1995; 270: Full Text Full Text PDF PubMed Scopus Google Scholar). for other small G nucleotide exchange factors of the family acting on G proteins R.A. A. J. Biol. Chem. 1996; 271: Full Text Full Text PDF PubMed Scopus Google the PH domain is to serve to the to ARF (5Brown H.A. Gutowski S. Moomaw C.R. Slaughter C. Sternweis P.C. Cell. 1993; 75: 1137-1144Abstract Full Text PDF PubMed Scopus (823) Google Scholar, 6Cockroft S. Thomas G.M.H. Fensome A. Geny B. Cunningham E. Gout I. Hiles I. Totty N.F. Truong O. Hsuan J.J. Science. 1994; 263: 523-526Crossref PubMed Scopus (586) Google and activity is on PIP2 of the S. K. S. M.A. Morris A.J. J. Biol. Chem. 1997; Full Text Full Text PDF PubMed Scopus Google it is possible that the of Arno by PIP2 serves to ARF in the of It be that membrane of Arno greatly enhances the activity of the exchange factor myrARF1, the catalytic site is not directly This is to from an increased of the two proteins on the membrane surface that The that Arno efficiently catalyzes GDP from [Δ17]ARF1 that the amino-terminal helix of ARF1 is not the site of interaction with the exchange which the of Kahn R.A. Nature. 1994; 372: PubMed Scopus Google Scholar) on the of The formation of a stable complex between Arno Sec7 domain and nucleotide-free [Δ17]ARF1 of for and of the between the two It is that in contrast to [Δ17]ARF1, the full-length is not readily activated by Arno in solution. This suggests that the amino-terminal helix the exchange as it the we previously that at Mg2+ GDP does not from whereas it is from myrARF in the presence of phospholipid vesicles (12Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1995; 270: 1337-1341Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar). the amino-terminal helix in is in a hydrophobic Kahn R.A. Nature. 1994; 372: PubMed Scopus Google Scholar, H. C. R. Fensome A. Thomas G.M.H. S. B. Biol. 1995; PubMed Scopus Google we proposed that insertion of the myristate into the bilayer the helix from the protein and the of the nucleotide binding site M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1996; 271: Full Text Full Text PDF PubMed Scopus Google Scholar). the myristate is the amino-terminal helix to the protein and the nucleotide binding site in a In this the only to GDP is to the affinity of ARF for GDP by decreasing the Mg2+ (11Franco M. Chardin P. Chabre M. Paris S. J. Biol. Chem. 1993; 268: 24531-24534Abstract Full Text PDF PubMed Google Scholar). The present that to GDP release from unmyrARF1, it is necessary to phospholipids, to the ionic strength of the and to the of we show that the of phospholipids is not to Arno binding to the a of the exchange is still Arno is by the Sec7 studies the of ARF with vesicles of have that with phospholipids hydrophobic interaction of the myristate with the bilayer and electrostatic interactions of cationic residues with anionic lipids (14Antonny B. Béraud-Dufour S. Chardin P. Chabre M. Biochemistry. 1997; 36: 4675-4685Crossref PubMed Scopus (281) Google Scholar). The ionic to the on the surface of the protein by residues and Kahn R.A. Nature. 1994; 372: PubMed Scopus Google Scholar, H. C. R. Fensome A. Thomas G.M.H. S. B. Biol. 1995; PubMed Scopus Google were not between and phospholipids at a ionic strength (14Antonny B. Béraud-Dufour S. Chardin P. Chabre M. Biochemistry. 1997; 36: 4675-4685Crossref PubMed Scopus (281) Google but they be at a salt concentration. the for a ionic strength here the to electrostatic interactions to for the absence of the decreasing the Mg2+ only the affinity for GDP or also the of the amino-terminal helix is not but that efficient activation of full-length ARF1 by Arno requires a membrane surface and an interaction of the bilayer with the amino-terminal portion of is the sequence of during the exchange the helix with the bilayer or the binding of Arno to In the first the binding site of Arno be to the of the In the of the factor and nucleotide exchange factor it proposed that the the nucleotide binding site by the of the C. M. S. R. Nature. 1996; PubMed Scopus Google Scholar). the to Arno, it is possible that the nucleotide binding site of ARF to be the interaction with Arno. the nucleotide binding site of ARFGDP two possible (a) a to the of Kahn R.A. Nature. 1994; 372: PubMed Scopus Google Scholar, H. C. R. Fensome A. Thomas G.M.H. S. B. Biol. 1995; PubMed Scopus Google and (b) a induced by a release of the amino-terminal helix by the Arno only the latter Arno normally to ARFGDP with the amino-terminal helix the protein but not be to the nucleotide binding site on the of phospholipids to the amino-terminal We cannot between the two the the present to the importance of the amino-terminal helix of ARF1 and of the protein-phospholipid interactions in the activation of ARF1 by Arno. We that efficient of GDP release by Arno can only on a membrane surface with the two proteins interacting with Arno interacting with PIP2 by PH domain, and ARF interacting with anionic phospholipids by amino-terminal Arno and ARFGDP can in the presence of the amino-terminal helix nucleotide exchange and the membrane of the whereas the PIP2 for Arno binding serve to activated ARF to We for