DeutschEnglish

Marino Zerial

Ongoing Projects

The following subprojects will give more detailed information on
a) The identification of Rab5 effectors and the first direct demonstration of a role in vesicle tethering
b) How Rab proteins and their effectors molecularly interact with SNAREs
c) The cooperativity between Rab effectors and how via protein-lipid and protein-protein interactions they cluster in a separate membrane domain on early endosomes
d) How in addition to membrane tethering and fusion Rab5 regulates endosome motility along microtubules


a) Identification and characterization of Rab5 effectors: Rab5 and EEA1 in membrane tethering

A breakthrough into the functional role of Rab5 came from the establishment of an affinity chromatography method based on recombinant GST-Rab5:GTP to search for novel Rab5-interacting proteins. All steps of the procedure (preparation of the affinity matrix, binding and elution conditions, etc.) have been optimized and specificity of the purification was guaranteed by the nucleotide-dependent elution step. This method allowed us to purify over 30 Rab5-binding proteins from bovine brain cytosol on the basis of their functional interaction with Rab5:GTP vs. Rab5:GDP. Among these proteins we identified the previously identified Rab5 effectors Rabaptin-5 and Rabaptin-5b as well as their co-factor Rabex-5. Strikingly, we could show that the Rab5:GTP column eluate was able to replace cytosol in an in vitro early endosome fusion reaction, indicating that the Rab5 effectors account for the requirement for cytosol in the fusion reaction. More strikingly, a single 170 kDa protein when present in excess could support early endosome fusion. This protein was identified as EEA1, a Rab5 effector that we have previously characterized in collaboration with H. Stenmark and colleagues. EEA1 is a predominantly alpha-helical protein peripherally associated with early endosomes. The binding to early endosomes is mediated by the C-terminal FYVE finger that binds to phospatidylinositol 3-phosphate PI(3)P, a lipid specifically produced on endosomes. We have found that EEA1 can cluster endosomes in an in vitro assay, thus providing the first direct demonstration that a Rab5 effector can act as a tethering molecule. These results together with other published studies in vitro and in vivo contributed to establish the general principle that Rab GTPases and their effectors act in the first step of membrane transport, i.e. tethering membranes, prior to SNARE pairing. Furthermore, we found that EEA1 can only be recruited on early endosomes and not on clathrin coated vesicles, suggesting that Rab effectors are the primary determinants of vesicle targeting to the acceptor compartment.

Simonsen et al. (1998)
EEA1 links phosphatidylinositol 3-kinase function to Rab5 regulation of endosome fusion.
Nature 394, 494-498

Chistoforidis et al. (1999)
The Rab5 effector EEA1 is a core component of endosome docking.
Nature 397, 621-625

Rubino et al. (2000)
Selective membrane recruitment of EEA1 suggests a role in directional transport of clathrin-coated vesicles to early endosomes.
J. Biol. Chem. 275, 3745-3748


b) Molecular link between Rab5 effectors and SNAREs

Our group provided the first biochemical demonstration of a direct, functional interaction between a Rab effector and a SNARE. Rab proteins regulate membrane tethering and they have long been thought to be upstream modulators of the integral membrane SNARE proteins implicated in bilayer fusion. SNARE proteins are essential in all vesicle transport events to date. We have therefore searched for molecular links between the Rab machinery controlling early endosome tethering and SNARE mediated fusion. For this we have explored possible interactions between the Rab5 effectors and the SNARE machinery on the early endosome membrane. We found that EEA1, which typically exists as a homodimer in cytosol, upon recruitment to the endosome membrane assembles into macromolecular complexes containing Rabaptin-5, Rabex-5, alpha-SNAP and NSF. We further demonstrated that EEA1 directly binds the t-SNARE syntaxin 13 and this interaction is essential for early endosome fusion.

McBride et al. (1999)
Oligomeric complexes link Rab5 effectors with NSF and drive fusion via direct interactions with EEA1 and syntaxin13.
Cell 98, 377-386

A second interaction between a Rab5 effector and SNAREs was provided by the functional characterization of Rabenosyn-5. Like EEA1, Rabenosyn-5 contains a FYVE-finger domain and is recruited in a phosphatidylinositol-3-kinase dependent fashion to early endosomes. Rabenosyn-5 is complexed to the previously identified Sec1-like protein, hVPS45. Sec1 family members are thought to “present” syntaxins to other SNARE complex members. In the case of endosomal SNAREs, this function would be contributed by Rabenosyn-5 through its interaction with hVPS45. Consistent with a role of modulating SNARE prtiming and pairing, we found that hVPS45 can directly bind multiple syntaxin family members.

Nielsen et al. (2000)
Rabenosyn-5, a novel Rab5 effector, is complexed with hVPS45 and recruited to endosomes through a FYVE finger domain.
J. Cell Biol. 151 (3): 601-612


c) Rab5 effectors act cooperatively: Organization of a membrane domain
The membrane association of both EEA1 and Rabenosyn-5 is shown to require Rab5 as well as Phosphatidylinositol 3-kinase (PI 3-K) activity. PI3Ks are a family of lipid modifying enzymes that play a role in signal transduction, cytoskeleton organization and membrane transport. Three different classes of PI 3-Ks have been described, that differ in their substrate specificities and regulation. EEA1 and Rabenosyn-5 bind to PI(3)P through their FYVE motif and their recruitment to the membrane is contingent upon the presence of this particular phosphoinositide species concomitant with Rab5:GTP. We have found that the cell uses the simplest mechanism to ensure the presence of both binding sites, namely by directly coupling Rab5 to PI3Ks.

We have identified two distinct PI 3-Ks, hVPS34/p150 and p85-alpha/p110-beta among the Rab5 effectors and we have recently demonstrated that PI(3)P is regulated by Rab5 (Hye-Won Shin et al. in preparation). An important implication of our work is that Rab5 mediates the recruitment of EEA1 not only by binding directly to this protein but also catalytically, through an interaction with hVPS34/p150. Whereas production of PI(3)P primarily depends on the activity of hVPS34/p150, also PI 3-K-beta contributes to this activity. This is surprising in view of the fact that Type I PI 3-K produce PI(3,4,5)P3 instead of PI(3)P. Once again, the explanation came from the analysis of the Rab5 affinity column eluate. We established that also PI5-phosphatase and PI4-phosphatase are Rab5 effectors and that, in addition to direct phosporylation of PI, turnover of PI(3,4,5)P3 accounts for the synthesis of PI(3)P on early endosomes. We believe this system is designed to convert excess of PI(3,4,5)P3 and PI(3,4)P2 transiently produced at the plasma membrane e.g. during signaling, into the PI species, i.e. PI(3)P necessary to maintain the organization and function of early endosomes.

Christoforidis et al. (1999)
Phosphatidylinositol-3-OH kinases are Rab5 effectors.
Nature Cell Biol. 4, 249-252

The identification of PI 3-Ks as Rab5 effectors illustrates once more the cooperativity among these molecules. Collectively, these data led us to propose a model to explain the organization of early endosomes based on the clustering of Rab5 and its effectors in a distinct membrane sub-compartment. Briefly, the membrane domain is based on 1) the localized production of PI(3)P, 2) on the cooperativity among rab5 effectors and 3) on their oligomerization properties that limit their diffusion. NSF regulates the oligomerization of the Rab effectors and the transition from membrane docking to fusion, by a yet unknown mechanism.

The unique features of these membrane domains, i.e. size, biochemical composition, localization and functional properties, suggest that Rab5 and its effectors may play a central role not only in the regulation of membrane trafficking to early endosomes but also in the biogenesis of this organelle.
The Rab-domain mosaic model of early endosomes is supported by the finding that three Rab proteins associated with early and recycling endosomes (Rab4, Rab5, Rab11) do not randomly colocalize but are segregated in distinct membrane territories, or Rab-domains of the organelle membrane. Cargo is transported from one domain to another and contiguous Rab-domains are functionally connected by so called “divalent Rab effectors”. Rabaptin-5 was the first molecule described capable of binding and functionally connecting two contiguous Rab proteins. This is a kind of “binary switch” system that allows cargo to be sequentially transported along the endocytic/recycling pathway. We believe that the Rab-domain organization of early endosomes is likely to be a general principle underlying organelle biogenesis in eukaryotic cells. Interestingly, an increasing number of molecules have been identified that appear to functionally couple one Rab GTPase to another, both in the secretory and endocytic pathways.

Vitale et al. (1998)
Distinct Rab Binding Domains mediate the interaction of Rabaptin-5 with GTP-bound rab4 and rab5.
EMBO J. 17, 1941-1950

Sönnichsen et al. (2000)
Distinct membrane domains in the endosomal recycling pathway visualised by multi-colour imaging of Rab4, 5 and 11.
J Cell Biol. 149(4): 901-14

Zerial and McBride (2001)
Rab proteins as membrane organizers.
Nature Rev. 2: 107-117

Miaczynska and Zerial (2002)
The mosaic organization of the endocytic pathway.
Exp Cell Res, 272, 8-14

De Renzis et al. (2002)
Divalent Rab effectors regulate the sub-compartmental organisation and sorting function of early endosomes.
Nat Cell Biol 4, 124—133


d) Rab5 regulates early endosome motility on microtubules

Dynamic associations with actin and microtubule cytoskeletal networks mediate both generation and movement of vesicular carriers as well as maintenance of the characteristic spatial distribution and morphology of endocytic organelles within the cell. We investigated whether Rab5 might regulate interactions of early endosomes with microtubules in mammalian cells. Using video microscopy we observed that the expression of Rab5 correlated with increased endosome movement towards the center of the cell, suggesting a regulation of the association of endosomes with, and/or motility along, microtubules. In collaboration with Tony Hyman we found that Rab5 regulates both the interaction of early endosomes with, and the motility of early endosome along microtubules, in vitro. We found also that endosome motility is dependent upon PI-3Kinase activity.

We propose that the interaction of Rab5:GTP with hVPS34 results in a localized production of PI-3P that, combined with the recruitment of Rab5 effectors contributes to the formation of aformation of a PI(3)P-enriched “Rab5-domain” on the endosomal membrane could localize a microtubule motor. We have recently identified a candidate motor protein that may fulfill the criteria for Rab5 and PI(3)P-dependent endosome movement along microtubules.

We have previously found that early endosomes move bi-directionally between the cell periphery and the interior through a mechanism regulated by the small GTPase RhoD. We identified a novel splice variant of human Diaphanous, hDia2C, which specifically binds to RhoD and is recruited on early endosomes. Expression of RhoD and hDia2C induces a striking alignment of early endosomes along actin filaments and reduces their motility. This activity depends on the membrane recruitment and activation of c-Src kinase, thus uncovering a new role in endosome function. Thus, a novel signal transduction pathway where hDia2C and c-Src are sequentially activated by RhoD regulates the motility of early endosomes via interaction with the actin cytsoskeleton.

In conclusion both Rab5 and RhoD concur to regulate endosome localization and movement within the cell, via processes regulated by both the actin and tubulin cytoskeleton.

Nielsen et al. (1999)
Rab5 regulates motility of early endosomes on microtubules.
Nature Cell Biology 1, 376-38

Gasman et al. (2003)
RhoD regulates endosome dynamics through Diaphanous-related Formin and Src tyrosine kinase.
Nat Cell Biol. 5(3):195-204


Standing of the research project in national and international comparison

The functional analysis of Rab5 effectors has provided an important contribution towards the understanding of the function of Rab proteins in the regulation of intracellular transport. This work has established general principles such as 1) that Rab effectors are vesicle tethering molecules, 2) that directly interact with SNAREs and 3) that regulate both vesicle tethering/fusion and organelle motility. The importance of this work is that it has showed that the role of Rab GTPases goes beyond the mere regulation of SNARE function and act as membrane organizers, recruiting and coordinating the activity of various effector molecules within restricted membrane domains. The Rab-domain model of organization of early endosomes is a first attempt to formulate a mechanism to explain the biogenesis of some if not most intracellular organelles. Furthermore, the molecular dissection of the Rab machinery has uncovered novel connections between transport and signaling pathways that will help understanding how cells respond to signaling molecules and regulate cell proliferation and developmental processes.


Future plans
Stemming from our research on Rab effectors, we will pursue our analysis of the endosomal system with respect to membrane compartmentalization and organelle biogenesis as well as the role of the endocytic machinery in signal transduction.

A central aim of the group is to test the Rab-domain model of endosome biogenesis.

This will be accomplished by analyzing the dynamics of Rab-domains in vivo and by functionally characterizing novel effectors for Rab GTPases that play a role in endocytosis and recycling. The studies in vivo will address the question of whether Rab-domains are stable in time or can be formed de-novo and undergo maturation. This question is very important in the context of the “vesicular transport” versus “endosome maturation” models. It is also important with respect to organelle inheritance during mitosis (see below). Affinity chromatography approaches similar to the one successfully used for Rab5 have been employed to identify downstream interactors of Rab4, Rab11 and Rab7. We value with particular interest the identification of divalent effectors for these GTPases, in order to test the Rab-domain hypothesis. The data so far obtained are extremely encouraging in this respect. Furthermore, we have systematically expressed the Rab5 effectors as recombinant proteins aiming at reconstituting the Rab5 machinery in proteoliposomes.

The goal of this project is to assemble artificial early endosomes using purified components and reconstitute the minimal machinery that is capable of Rab-dependent membrane fusion and motility in vitro. This approach will not only enable a more detailed study on the functional relationship between Rab effectors and SNAREs, but will also address the mechanism of membrane segregation regulated by Rab5 and its downstream effectors.

To elucidate the mechanisms regulating endocytosis in polarised cells.

This aim includes the characterization of Rab effectors participating in polarised endocytic transport. For example, the Rab5 effector Rabankyrin-5 regulates apical but not basolateral endocytic uptake in polarized epithelial cells. We will explore the role of various Rab effectors, especially those so far uncharacterized, in polarised trafficking. However, beyond the analysis of Rab effectors, we aim at identifying and functionally characterizing novel molecules regulating endocytic transport using genetically amenable organisms, e.g. C. elegans, D. melanogaster (with S. Eaton and M. Gonzalez) and mice (Frank Buchholz). This will be accomplished with the technical support of the assay development and screening facility of the MPI-CBG that will enable us to perform genome-wide RNAi screenings.

An important aspect of the study of endosome biogenesis is to elucidate the mechanisms regulating microtubule-dependent endosome motility and partitioning between daughter cells during mitosis.

New assays have been designed to investigate this problem both in vitro and in vivo. The identification of a motor protein that specifically interacts with PI(3)P and is recruited on the early endosome membrane provides us with a molecular tool to begin the functional dissection of this process.

The role of Rab GTPases in the structural and functional organization of endosomes has profound implications for how endocytic transport is modulated by signalling molecules and, conversely, how the endocytic machinery modulates the responsiveness of cells to signalling.

For example, stimulation by EGF enhances the rate of endocytic membrane flow by increasing the fraction of active Rab5. We have recently discovered two novel Rab5 effectors, APPL1 and APPL2, that are excellent candidates for molecules linking endocytic trafficking to signalling. APPL1 and 2 reside on a novel intracellular vesicle or organelle, the hermesome. We have shown that the hermesome functions as a specialized endosome devoted to signalling from the plasma membrane to the nucleus. An important task is now to elucidate the role of this new compartment in the interface between signalling and endocytosis. Questions that will be addressed are: Which specific physiological role does this organelle exert in signal transduction? How is transport to the hermesome vs. endosome regulated? Do hermesomes and endosomes communicate and how? What is the role of Rab5 in this process? What is the link between APPL and endocytosis in cell proliferation?

Do hermesomes have a widespread role? Do other growth factors use the hermesome signaling pathway? To what extent this pathway regulates proliferation vs. differentiation in different cell types?

We expect that progress along these lines will reveal novel insights into the molecular principles of endosome functions in trafficking, signalling and in the establishment and maintenance of cell polarity.

Lanzetti, L., Rybin, V., Malabarba, M.G., Christoforidis, S., Scita, G., Zerial, M., and Di Fiore, P.P. (2000)
The Eps8 protein coordinates EGF receptor signalling through Rac and trafficking through Rab5.
Nature 408: 374-377