Posttranslational Modifications of Rab Gtpases Help Their Insertion Into Membranes

COMMENTARY

Posttranslational modiﬁcations of Rab GTPases help their insertion into membranes

Olena Pylypenko and Bruno Goud1 Unité Mixte de Recherche 144, Centre National de la Recherche Scientiﬁque, Institut Curie, 75248 Paris Cedex 05, France

TPases of the Rab family, which comprises more than 60 mem- G bers in mammals, have emerged as key regulators of virtually all transport steps between intracellular compartments. By interacting with a di- verse range of effector proteins, such as lipid kinases and phosphatases, molecular motors, tethering factors, and scaffolding proteins, they regulate the formation of transport carriers from donor membranes, their movement along cytoskeletal tracks, and their targeting to acceptor membranes (1). Like other GTPases, Rab proteins function as molecular switches, cycling between an inactive GDP-bound form and an active GTP-bound form. Because Rabs display high affinities for both GDP and GTP and very low intrinsic GTPase Fig. 1. (A) The Rab GTPase cycle. (B) Rab modifications by Legionella proteins dramatically change Rab– activity, the GDP/GTP cycle must be reg- GDI complex affinity. A Rab protein prenylated with a fluorescently labeled farnesyl moiety (red star) ulated by guanine nucleotide exchange forms a high-affinity complex with GDI (B, 1); the intrinsic dissociation rate of the complex is very low (B, factors (GEFs) that promote GDP release 1 and 2), and equilibrium is shifted toward RabGDP–GDI complex (gray arrows). DrrA GEF activity con- and by GTPase-activating proteins (GAPs) verts Rab to the GTP-bound form (B, 2 and 3) snatching a small fraction of the RabGDP spontaneously that stimulates GTP hydrolysis. released from GDI. The GTP loaded Rab has low affinity to GDI, and the equilibrium shifts toward A salient feature of Rab proteins is a complex dissociation. The activated Rab is recognized as a substrate by DrrA ATase domain that ad- enylylates the GTPase (B, 3 and 4). AnkX also relies on the free Rab pool generated by intrinsic Rab–GDI cycle of association/dissociation to/from fi membranes that is superimposed to the complex dissociation. The phosphocholine transferase modi es the Rabs (B, 2 and 5), rendering them GDP/GTP cycle (Fig. 1A). Rabs therefore incompetent to rebind to GDI and shifts the equilibrium to the complex dissociation state, whereas the lpg0696 dephosphocholination activity reverses the process (i.e., 5, then 2,inB). mainly exist as two pools, one membrane and GTP-bound and one cytosolic and GDP-bound (estimated depending on geting of different Rabs to specific mem- Legionella effectors reorient the host cell the Rab to range between 10% and 40% at branes and their activation (GDP/GTP vesicle-trafficking pathway to grab the the steady state). The stable insertion of exchange), with both processes being in- host ER-derived vesicles necessary for the Rabs in the outer leaflet of intracellular timately linked. Given the high affinity formation and maintenance of a special- membranes is mediated by one or, in most of GDI for prenylated RabGDP and con- ized vacuolar compartment called Legion- cases, two geranylgeranyl lipid moieties sequently the low intrinsic rate of GDI ella-containing vacuole (LCV) (6, 7). that are attached to cysteine residue(s) dissociation, certainly the crucial event of DrrA/SidM specifically localizes, immedi- present at their C terminus. This irre- this step is the dissociation of RabGDP– ately after infection, to the LCV mem- versible posttranslational modification GDI complexes. Two main mechanisms brane, where it can efficiently recruit occurring on newly synthesized Rabs have been proposed to explain GDI dis- Rab1, a Rab regulating the early secretory involves a protein named REP (Rab placement. Based on the observations pathway (8). Schoebel et al. first showed escort protein), which presents the Rab that nucleotide exchange occurs shortly that the DrrA GEF domain is sufficient fi – proteins to Rab geranylgeranyl trans- after binding of puri ed Rab5 and Rab9 to displace GDI from Rab1GDP–GDI ferase (RabGGTase). Cytosolic Rabs are GDI complexes to endosomal membranes complex by catalyzing nucleotide exchange bound to the protein GDI (GDP Dissoci- using in vitro or semi-in vitro assays (3, 4), fi to GTP (9). Wu et al. then demonstrated, ation Inhibitor; two GDIs exist in mam- the rst mechanism suggests the existence by using a semisynthetic farnesyl-NBD mals, GDIα/GDI-1 and GDIβ). GDI binds of membrane-associated proteins termed fi fi Rab7 probe, that the af nity of GDI for to Rabs with very high af nity (Kd in the GDF (GDI Displacement Factor) that Rab7GTP is approximately three orders of nanomolar range) when they are preny- favor the release of GDI (Fig. 1A, pathway magnitude lower than the affinity for lated and in the GDP state, which explains 2). One protein with a GDF activity has fi Rab7GDP. Based on these results, the that GDI functions as a chaperone dur- been identi ed, Yip3/PRA1, and shown authors proposed that GEFs are necessary to achieve the dissociation of Rab9–GDI ing the Rab cycle, extracting them from and sufficient for Rab membrane asso- membranes after GAP-induced GTP hy- complexes and to facilitate Rab9 mem- ciation (10). However, the GEFs cannot drolysis (Fig. 1A, pathway 1) and solubi- brane insertion (5). However, it is difficult lizing them in the form of RabGDP–GDI to generalize this mechanism, as Yip3 re- complexes. In a study in PNAS, Oesterlin mains the only GDF identified so far. A et al. (2) address the question of Rab second mechanism postulates that nucle- Author contributions: O.P. and B.G. wrote the paper. membrane attachment mechanism—the otide exchange plays a central role in GDI The authors declare no conflict of interest. less well understood Rab cycle step. release from RabGDP (Fig. 1B, pathway See companion article 10.1073/pnas.1121161109. The Rab membrane attachment step 3). It is based on the study of DrrA/SidM, 1To whom correspondence should be addressed. E-mail: couples, in fact, two processes: the tar- an effector of Legionella pneumophila. [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1202494109 PNAS Early Edition | 1of2 Downloaded by guest on September 28, 2021 interact directly with RabGDP–GDI SidM (15). This modification blocks early which may favor its membrane insertion complexes, as available crystal structures secretory events involving vesicular trans- (Fig. 1B). Phosphocholination could how- of Rab–GEF and Rab–GDI complexes port on microtubules after exit from the ever affect the interaction with GEFs, as show that GEFs and GDI bind to over- ER and inhibits endocytic pathways (16). shown for phosphocholinated Rab35 and lapping sites and thus would have to Oesterlin et al. (2) use a fluorescence- its GEF connecdenn (15). However, in- compete for Rab binding (9, 11). The based assay to explore the ability of DrrA terestingly, Legionella also expresses a de- GEF-mediated Rab membrane attach- phosphorylcholinase, Lgp0696, that can ment mechanism then implies that the fi – Oesterlin et al. address reverse AnkX-mediated modi cation of dissociation rate of the RabGDP GDI Rab1 (17), and thus possibly permits is high enough to present enough the question of Rab binding of Rab1 to its GEF(s). RabGDP molecules to GEFs for generat- Posttranslational modifications of Rab ing a sufficient number of active Rabs GTPases represent a tentative mechanism that will interact with their effectors. membrane attachment Oesterlin et al. (2) demonstrate the ex- to explain, at least in part, how Rab istence of other mechanisms that can mechanism—the less GTPases can be targeted to membranes. lower the affinity of GDI for RabGDP and Evidence exists that phosphorylation can thus stabilize the dissociated RabGDP. well understood Rab affect the membrane/cytosol distribution – This study is again based on the bio- of a few Rabs (18 20). However, it re- chemical properties of Legionella effec- cycle step. mains to be established whether posttrans- tors. In addition to a GEF domain, DrrA/ lational modifications can be considered SidM contains a N-terminal domain that as a general regulatory mechanism. In- and AnkX to displace GDI from NBD- possesses adenosine monophosphory- terestingly, a human protein containing farnesyl-Rab:GDI complexes in vitro. lation (AMPylation or adenylylation) a Fic domain, termed HYPE (huntingtin They show that recombinant DrrA and activity toward a conserved tyrosine resi- yeast-interacting protein E) has been AnkX can efficiently adenylylate and due (Tyr77) in the switch II region of recently shown to possess the ability to phosphocholinate in vitro Rab1b and Rab1b, a region involved in binding to fi AMPylate Rho GTPases, as bacterial Fic different protein partners (12). The active Rab35, respectively. Both modi cations drastically lower their affinities for GDI by domain proteins do (21). GTP-bound form of Rab1 is a preferable Another issue is to explain how Rabs, substrate for DrrA/SidM-catalyzed ad- several orders of magnitude and could thus efficiently displace GDI from Rab1b/ which are all geranylgeranylated by the enylylation, suggesting that Rab1 activa- – same enzyme and interact with only one or tion is coupled to its modification. The Rab35 GDI complexes. However, GDI fi displacement by adenylylation is depen- two different GDIs, are targeted to spe- recently reported identi cation of SidD cific membranes. It has long been thought with deadenylylation activity toward Rab1- dent on DrrA-mediated Rab1 activation, fi fi which makes it unlikely that such a modi- that this speci city relies on the existence AMP (13) con rms that AMPylation is of a targeting signal within the C-terminal a regulated reversible modification like fication is meaningful in vivo to promote hypervariable domain of Rab GTPases many other posttranslational modifica- GDI dissociation from RabGDP–GDI tions. The other Legionella effector stud- complexes during membrane insertion (22). However, this view has been chal- ied by Oesterlin et al. (2) is AnkX/ (Fig. 1B). In contrast, the AnkX catalytic lenged by several studies showing that this Lpg0695. AnkX is a Fic domain-contain- activity does not depend on the nucleotide is not the case for several Rabs and that ing protein that was predicted to function status of Rab1b or Rab35. Even though membrane targeting involves multiple Rab as an AMPylating toxin (14), but it was phosphocholination does not accelerate regions (23, 24). In light of the results proven later that this protein is, in fact, the intrinsic rate of GDI dissociation from discussed here, it seems more and more a phosphocholine transferase (15). AnkX RabGDP–GDI complexes (as also ob- likely that Rab targeting relies on specific modifies Rab1 and Rab35 on a serine served for GEF-mediated nucleotide GEFs and/or specifically localized en- residue adjacent to the conserved tyrosine exchange), phosphocholination would zymes that catalyze posttranslational that is the adenylylation site for DrrA/ prevent RabGDP from rebinding to GDI, modifications.

1. Hutagalung AH, Novick PJ (2011) Role of Rab GTPases 9. Schoebel S, Oesterlin LK, Blankenfeldt W, Goody RS, versible phosphorylcholination. Proc Natl Acad Sci USA in membrane traffic and cell physiology. Physiol Rev Itzen A (2009) RabGDI displacement by DrrA from 108:21212–21217. 91:119–149. Legionella is a consequence of its guanine nucleotide 18. Bailly E, et al. (1991) Phosphorylation of two small GTP- 2. Oesterlin LK, Goody RS, Itzen A (2012) Posttransla- exchange activity. Mol Cell 36:1060–1072. binding proteins of the Rab family by p34cdc2. Nature tional modifications of Rab proteins cause effective 10. Wu Y-W, et al. (2010) Membrane targeting mechanism 350:715–718. displacement of GDP dissociation inhibitor. Proc Natl of Rab GTPases elucidated by semisynthetic protein 19. van der Sluijs P, et al. (1992) Reversible phosphoryla- – Acad Sci USA, 10.1073/pnas.1121161109. probes. Nat Chem Biol 6:534 540. tion—dephosphorylation determines the localization 3. Ullrich O, Horiuchi H, Bucci C, Zerial M (1994) Mem- 11. Rak A, et al. (2003) Structure of Rab GDP-dissociation of rab4 during the cell cycle. EMBO J 11:4379–4389. brane association of Rab5 mediated by GDP-dissocia- inhibitor in complex with prenylated YPT1 GTPase. Sci- 20. Fitzgerald ML, Reed GL (1999) Rab6 is phosphory- ence 302:646–650. tion inhibitor and accompanied by GDP/GTP exchange. lated in thrombin-activated platelets by a protein 12. Müller MP, et al. (2010) The Legionella effector protein Nature 368:157–160. kinase C-dependent mechanism: Effects on GTP/GDP DrrA AMPylates the membrane traffic regulator 4. Soldati T, Shapiro AD, Svejstrup AB, Pfeffer SR (1994) binding and cellular distribution. Biochem J 342: Rab1b. Science 329:946–949. Membrane targeting of the small GTPase Rab9 is ac- 353–360. 13. Neunuebel MR, et al. (2011) De-AMPylation of the companied by nucleotide exchange. Nature 369:76–78. 21. Worby CA, et al. (2009) The fic domain: Regulation of small GTPase Rab1 by the pathogen Legionella pneu- 5. Sivars U, Aivazian D, Pfeffer SR (2003) Yip3 catalyses cell signaling by adenylylation. Mol Cell 34:93–103. mophila. Science 333:453–456. the dissociation of endosomal Rab-GDI complexes. Na- 22. Chavrier P, et al. (1991) Hypervariable C-terminal do- 14. Roy CR, Mukherjee S (2009) Bacterial FIC proteins AMP – main of rab proteins acts a targeting signal. Nature ture 425:856 859. up infection. Sci Signal 2:pe14. – 6. Swanson MS, Isberg RR (1995) Association of Legion- 15. Mukherjee S, et al. (2011) Modulation of Rab GTPase 353:769 712. ella pneumophila with the macrophage endoplasmic function by a protein phosphocholine transferase. Na- 23. Beranger F, Paterson H, Powers S, de Gunzburg J, – reticulum. Infect Immun 63:3609 3620. ture 477:103–106. Hancock JF (1994) The effector domain of Rab6, plus 7. Kagan JC, Roy CR (2002) Legionella phagosomes inter- 16. Pan X, Lührmann A, Satoh A, Laskowski-Arce MA, a highly hydrophobic C terminus, is required for cept vesicular traffic from endoplasmic reticulum exit Roy CR (2008) Ankyrin repeat proteins comprise a di- Golgi apparatus localization. MolCellBiol14: sites. Nat Cell Biol 4:945–954. verse family of bacterial type IV effectors. Science 320: 744–758. 8. Murata T, et al. (2006) The Legionella pneumophila 1651–1654. 24. Ali BR, Wasmeier C, Lamoreux L, Strom M, Seabra MC effector protein DrrA is a Rab1 guanine nucleotide-ex- 17. Tan Y, Arnold RJ, Luo Z-Q (2011) Legionella pneumo- (2004) Multiple regions contribute to membrane tar- change factor. Nat Cell Biol 8:971–977. phila regulates the small GTPase Rab1 activity by re- geting of Rab GTPases. J Cell Sci 117:6401–6412.

2of2 | www.pnas.org/cgi/doi/10.1073/pnas.1202494109 Pylypenko and Goud Downloaded by guest on September 28, 2021