EP1214446A1

EP1214446A1 - Assay to detect substances useful for therapy

Info

Publication number: EP1214446A1
Application number: EP00966051A
Authority: EP
Inventors: Erik Nielsen; Savvas Christophoridis; Carol Murphy; Marino Zerial; Stefano De Renzis
Original assignee: Max Planck Gesellschaft zur Foerderung der Wissenschaften eV
Current assignee: Max Planck Gesellschaft zur Foerderung der Wissenschaften eV
Priority date: 1999-09-16
Filing date: 2000-09-18
Publication date: 2002-06-19
Also published as: WO2001020022A1; CA2384306A1; EP1088898A1

Abstract

The present invention relates to the use of effectors/regulators for Rab and Rho GTPases in in vitro and in vivo assays that recapitulate and measure the role of these effectors/regulators in membrane transport and membrane-cytoskeleton interactions in the endocytic pathway as novel targets to find therapeutic drugs to prevent or inhibit cancer cell growth and arrest cancer cell invasiveness as well as for stimulating and/or restoring endocytic transport and phagosome maturation in cells infected by intracellular parasites, which drugs are therefore useful for the therapy and optionally also the prophylaxis of cancer and infectious diseases. In addition, the present invention is also directed to kits useful as a means to detect drugs suitable as anti-cancer and anti-infectious diseases drugs.

Description

Assay to Detect Substances Useful for the Therapy

The present invention relates to the use of effectors/regulators for Rab GTPases in in vitro and in vivo assays that recapitulate and measure the role of these effectors/regulators in membrane transport and membrane-cytoskeleton interactions in the endocytic pathway as novel targets to find therapeutic drugs to prevent or inhibit cancer cell growth and arrest cancer cell invasiveness as well as for stimulating and/or restoring endocytic transport and phagosome maturation in cells infected by intracellular parasites, which drugs are therefore useful for the therapy and optionally also the prophylaxis of 1) cancer and other proliferative (skin repair diseases such as psoriasis), invasive or cell migration disorders (endometriosis, atherosclerosis, inflammation and allergic diseases), 2) infectious (bacterial and viral) diseases, 3) diabetes, 4 ) Alzheimer's disease. In addition, the present invention is also directed to kits useful as a means to detect drugs suitable as anti-cancer and anti-infectious diseases drugs.

A class of molecules shown to play an important role in the regulation of intracellular transport and organelle function is represented by Rab proteins, small GTPases of the Ras superfamily, which are required in virtually every transport step which has been investigated. Similar to other GTPases, these molecules use the conformational change induced by GTP hydrolysis to regulate downstream events necessary for vesicle formation, docking and fusion. In the GTP-bound, active form Rab proteins bind to effector proteins and in this way transmit their signal to the transport machinery. For example, in membrane docking and fusion, Rab proteins regulate the activity of SNAREs. SNAREs are integral membrane proteins that by pairing on opposite membranes engaged in docking lead to membrane fusion. The pairing of SNAREs requires the activity of Rab proteins and therefore of Rab effectors. For example, endosome membrane docking requires the presence of the Rab5 effector EEA1 which, upon bridging the two opposite membranes, allows SNAREs to pair in trans, thus leading to membrane fusion (Chistoforidis et al., 1999a). The role of Rab proteins is not restricted to membrane docking and fusion but recent data provide evidence that Rab proteins also regulate the association with and motility of vesicles along cytoskeletal filaments. Membrane-cytoskeleton interactions play an important role in determining the intracellular distribution and motility of organelles and transport vesicles. The cytoskeleton is made of different types of filaments, actin filaments and microtubules being the best characterised. Actin filaments and microtubules undergo continuous remodelling to serve the needs of the cell in cell architecture, motility, and organelle movement. Microtubules play a pivotal role in cell division but also constitute tracks along which cargo is transported across long distances such as between the periphery and the centre of the cell, from the basal to the apical pole in epithelial cells, or along the axon in neurons. Actin filaments are well known to function in muscle contraction. They are also important for cell motility, cell shape and vesicular transport. Different proteins are necessary to attach the organelles and transport vesicles to, and move them along, these filaments. Rab5 has recently been demonstrated to regulate the attachment of endosomes to, and motility along, microtubules. This suggests that Rab5, and in general members of the Rab proteins family, regulate various aspects of intracellular transport, including the ability of organelles to move along cytoskeletal tracks.

The identification of a large number of effectors and regulators for the small GTPase Rab5 provides a molecular explanation for the multiplicity of functions of Rab5 and allow to predict similar mechanisms for other Rab GTPases. Rab5 regulates a molecular network of several effector proteins, each contributing a specific function in membrane organisation, vesicle formation, vesicle and organelle movement, membrane docking and fusion. Most importantly, by functioning in a cooperative fashion Rab5 effectors modify the membrane environment and thus contribute to the biogenesis of the early endosome membrane. This mode of action is exemplified by the following mechanism elucidated by the present inventors. Upon activation by the effector/exchange factor Rabaptin-5/Rabex-5 complex, Rab5 locally recruits and activates phosphoinositide PI3-Kinases, leading to the generation of PI(3)P and consequently allowing the membrane recruitment of other Rab5 effectors (e.g. EEAl; see below) that bind to both Rab5:GTP and PI(3)P (Christoforidis et al., 1999b). Furthermore, Rab5 effectors are engaged in the formation of oligomeric complexes on the early endosome membrane (McBride et al., 1999). By the same criteria, other Rab proteins present in the early endosomes would be expected to recruit multiple effectors within a separate membrane environment. Consistent with this, studies using (green fluorescent protein) GFP -tagged Rab5, Rab4, and Rab 11 have demonstrated that these GTPases are present in separate sub-compartments of the early endosome membrane. Endosomes are therefore organised as a mosaic of different Rab-domains created through the recruitment of specific effector proteins, which co-operatively act to generate a topologically restricted and functionally specialized environment on the endosome membrane. Laying at the core of the organelle biogenesis and membrane transport mechanisms, Rab GTPases and their effectors are placed in a strategic position to regulate and mediate the trafficking of cellular membrane and soluble components along the biosynthetic and endocytic pathways. Their function includes the regulation of transport of cellular constituents (proteins and lipids) that are implicated in pathological alterations. Consequently, the present inventors' understanding of the mechanisms underlying membrane trafficking and membrane-cytoskeleton interactions, has applications in biotechnological research. Applications are in particular possible for those diseases where modulation of the membrane trafficking properties of endosomes provides an opportunity for therapeutic intervention. The main research areas for which this knowledge has proven to have important implications are: 1) cancer and other proliferative (skin repair diseases such as psoriasis), invasive or cell migration disorders (endometriosis, atherosclerosis, inflammation and allergic diseases), 2) infectious (bacterial and viral) diseases, 3) diabetes, 4 ) Alzheimer's disease. In the following section, the Rab GTPases and Rab effectors machinery will be presented in the context of the development of a therapeutic strategy for the treatment of these diseases.

1. Cancer

The molecular analysis of the transport machinery represented by Rab effectors operating in the endocytic pathway is relevant to two related aspects of the molecular mechanisms underlying tumor growth: intracellular signalling and cell spreading.

Intracellular signalling

Cell growth is regulated by growth factors that by binding specific receptors on the cell surface trigger a signalling cascade that culminates with the regulation of gene expression. Growth factors and their receptors are important mitogens both in normal and in transformed cells. Mutations in their genes or overexpression can cause tumors or stimulate tumor cell growrth. For example, many tumors overexpress growth factors and surface tyrosine kinase receptors thus enhancing tumor cell growth. Growth factors and growth factor receptors have therefore been considered as excellent targets for new therapeutic drugs in the treatment of cancer and other disorders of excessive cellular proliferation. For example, the Herc2 receptor implicated in metastatic breast cancer provided the target for the development of Herceptin, a monoclonal antibody therapeutic from Genentech. The strategies pursued so far include the search for compounds that i) block ligand-receptor interactions, ii) the kinase activity of the receptors or the subsequent interactions with downstream factors of the signalling cascade. Another approach consists of the search for drugs that modulate receptor trafficking through the endosomal system and result in receptor downregulation at the plasma membrane. Growth factor receptors and their bound ligands are internalised into endosomes, and endocytic trafficking of these molecules plays a critical role not only in attenuating the signalling response but also in establishing and controlling specific signalling pathways. The ability of growth factors to trigger signaling depends on the surface appearance of their receptors. This is determined by the kinetics and extent of receptor 1) endocytosis, 2) recycling to the cell surface and 3) transport to, and degradation in, late endocytic compartments (Sorkin et al., 1991). Molecules that can modulate these events and decrease the fraction of growth factor receptors on the plasma membrane are therefore valuable tools for an anti-cancer therapy. For example, enhanced degradation of epidermal growth factor receptor correlates with the lack of epidermal growth factor- induced proliferation and mitogen-activated protein kinase stimulation (Caraglia et al., 1999). Given their established role in the regulation of endosome structure and function, Rab GTPases and their network of Rab effectors represent an ideal target for screening of molecules that can reduce the surface content of growth factor receptors and, consequently, inhibit the growth properties of tumor cells. Rab5 for example regulates the internalisation of receptors from the plasma membrane into clathrin-coated vesicles (CCV) as well as the subsequent delivery of these carriers to early endosomes. Other Rab proteins such as Rab4 and Rab 11 regulate the sorting function of early and recycling endosomes and, consequently, the recycling to the surface of growth factor receptors. The small GTPase Rab7 regulates transport of molecules from early to late endosomes and lysosomes, the degradative endocytic organelles.

The impact of Rab GTPases and their effectors/regulators on mitogenic signaling also includes the trafficking of signaling molecules downstream of growth factor receptors. An increasing number of components of the signalling machinery have recently been localised to early endocytic organelles. This remarkably specific localisation is very likely an important element in the signalling process. Such localisation depends on the structural and functional properties of the early endosomes, which in turn depends on the activity of Rab GTPases and their corresponding effectors/regulators. For example, the localisation of FYYE finger proteins depends on the synthesis of PI(3)P on the early endosome membrane (Christoforidis et al., 1999b). In the absence of this phosphoinositide these proteins are no longer capable of residing on early endosomes and are therefore released into the cytosol. Several novel signaling molecules possess a FYVE finger and, consequently, their ability to reside on early endosomes and to participate in the signaling cascade depends on the synthesis of PI(3)P. This concept is corroborated by recent data indicating that hVPS34, the PI3-K that produces PI(3)P is required for mitogenic signaling (Siddhanta et al., 1998). The synthesis of PI(3)P is subjected to the regulation by Rab5, since PI-3 kinases (including hVPS34) are Rab5 effectors (Christoforidis et al., 1999b). Indeed, recent studies of the present inventors have demonstrated that the production of PI(3)P is stimulated by Rab5 and inhibited upon removal of Rab5 from endosomes. These observations therefore establish a direct link between the activity of Rab5 and its effectors such as PI-3 kinases and the membrane recruitment of signalling molecules to early endosomes and their ability to function in the signalling process.

Cell spreading

The growth of human tumors depends not only on the proliferative advantage characteristic of cancer cells but also on the ability of these to invade tissues and to colonise different organs. The ability of tumors to form metastatic tumours depends on the invasive properties of transformed cells. In addition, cancer cells require a higher supply of nutrients and, therefore, require de novo formation of blood vessels. Clearly, the spreading of endothelial cells during angiogenesis that occurs concomitantly with tumor growth also requires an increase in cell motility. Cell locomotion is therefore a central mechanism in tumor cell invasiveness and metastasis. On the one hand, cell spreading and invasiveness depend on the signalling function of various growth factor and growth factor receptors. In many different cell types, including tumor cells, EGFR signaling produces a pleiotropic response that includes mitogenesis or apoptosis, enhanced cell motility, protein secretion, and differentiation or dedifferentiation. The formation of new blood vessels is instead tightly regulated by specific growth factors that target receptor tyrosine kinases (RTKs) such as vascular endothelial growth factor (VEGF) (McMahon, 2000). A novel approach to treat human cancers emerging from these studies consists in the development of therapeutic strategies aimed at blocking tumor cell invasiveness and deregulated angiogenesis through the inhibition of various growth factor receptor signaling pathways (both in tumor cells and in the neighbouring tissue such as in endothelial cells). The potential use of Rab GTPases and their effectors/regulators as targets for the screening of molecules that can reduce the signaling function of growth factor receptors applies therefore not only for the mitogenic signaling but also for the spreading of tumor cells. On the other hand, the use of Rab GTPases and their effectors/regulators can also be extended to the cellular machinery that executes the signaling program resulting in increased cell locomotion. Cell locomotion occurs by a complex mechanism involving the coordinated activity of both the cytoskeleton - actin and microtubules - and membrane trafficking - especially endocytic and recycling structures. The structural/functional properties of endosomes are important elements of the cell motility apparatus. For example, the endocytic cycle between endosomes and the plasma membrane sustains the deposition of adhesion molecules to the leading edge of motile cells. Consequently, the function of molecules that play an important role in endocytic transport can be rate-limiting important factors for the survival and spreading of cancer cells. This proposal is supported by recent studies that demonstrate that the small GTPase Rab5 is required for cell motility. The use of Rab GTPases and their effectors/regulators is therefore here applied for the screening of molecules that can reduce the ability of the cell to respond to mitogens through increased cell motility and, consequently, inhibit the spreading of tumor cells.

The described application of Rab effector intervention as a means of inhibiting or regulating cell motility defines the potential of the intervention for the treatment of other diseases that depend on cell locomotion such as endometriosis, atherosclerosis and monocyte migration into plaques, asthma and allergic diseases, and any other disease dependent on cell migration or locomotion.

2. Infectious diseases

Another area according to the present invention where membrane trafficking molecules, i.e., Rab GTPases and their effectors/regulators, can serve as therapeutic targets is that of infectious diseases. Cells internalise soluble ligands through endocytosis and large particles through phagocytosis. The latter mechanism is exploited by parasites that in this way gain access to the endomembrane system of the host cell. Increasing numbers of studies indicate that phagocytosis is regulated by the same molecular principles that govern endocytosis and recycling. For example, amphiphysin II and dynamin 2, which participate in receptor-mediated endocytosis, are also required for phagocytosis at the stage of membrane extension around the bound particles. Since the phagocytic process results in the internalisation of a large portion of the plasma membrane, mechanisms exist to ensure rapid renewal of plasma membrane. Given the role of Rab proteins in the regulation of membrane transport, it is obvious that they play an important role in the phagocytic cycle as well. Indeed, recent studies have shown that Rabl 1 operates in an endocytic compartment which is essential for phagocytosis. The function of endocytic Rab GTPases and their corresponding effectors/regulators are therefore expected to play an important regulatory role in the host cell invasion by parasites mediated by phagocytosis (Rab5) and in the compensatory membrane recycling (Rab4, Rabl 1) sustaining this process.

Once internalised, engulfed parasites or particles reside intracellularly in phagosomes. Under normal conditions, phagosomes mature by docking and fusing progressively with early endosomes, late endosomes and lysosomes, thus causing the destruction of their content due to the low pH and increasing exposure to hydrolytic enzymes. However, it is well established that certain microbes can sabotage the cellular defence mechanisms and survive intracellularly. This is the case for example for Mycobacterium tuberculosis that once internalised by phagocytosis resides within the phagosome without coming in contact with harmful organelles such as lysosomes. A possible strategy to facilitate the killing of the pathogen is to re-activate the membrane trafficking route, thus resuming the maturation of the phagosome and, therefore, the exposure of the internalised bacterium to low pH and harmful hydrolytic enzymes. The strategy should focus on different Rab proteins each regulating a distinct stage of phagosome maturation (i.e. Rab5, Rab4, Rabl 1, Rab7).

Another area where Rab GTPases and their effectors can be used for therpeutic intervention is that of viral infection, especially HIV. AIDS continues to be cause of many deaths both in developed and under-developed countries. It has been recently established that, besides CD4, members of the family of chemokine receptors are required as co-receptors, which in conjunction with CD4, allow the virus to enter cells. Downregulation of chemokine receptors by endocytosis and inhibition of receptor recycling protects cells from HIV. These observations imply that receptor internalization and inhibition of receptor recycling represent novel therapeutic strategies to prevent HIV infection and therefore combat AIDS. Since Rab GTPases and their effectors/regulators regulate the trafficking of receptors to the cell surface, these molecules represent new targets for therapeutic agents aimed at decreasing the surface expression of HIV receptors and therefore inhibit HIV entry into cells and prevent HIV infection.

3. Diabetes

Another area where the endocytic trafficking apparatus has potential commercial applications is the trafficking of glucose transporters and the defect in this process, which is a cause of diabetes. In fat and muscle cells, stimulation by insulin causes the relocation of glucose transporters (e.g. GLUT4) from intracellular stores to the cell surface. These intracellular stores are thought to be specialised membrane domains of the recycling endomembrane system. Insulin sensitivity depends on the surface levels of such glucose transporters (GLUTs). Under physiological conditions, the increase in muscle insulin sensitivity of glucose transport observed after exercise is due to translocation of more GLUT-4 to the cell surface. Under pathological conditions, GLUTs trafficking is perturbed. Certain forms of diabetes are in fact caused by deficiency in the surface appearance of the GLUTs. This implies that the elucidation of the transport machinery that regulates the trafficking of membrane proteins, including GLUTs, along the endocytic and recycling pathway offers therefore interesting opportunities for intervention in diabetes. In particular, the discovery of Rab effectors that play a role in the transport to endosomes and from endosomes to other destinations, e.g., the cell surface or to degradative compartments, identifies potential targets for the screening of drugs that, by altering the trafficking properties of GLUTs, can restore or stimulate their delivery to the plasma membrane in response to insulin.

4. Alzheimer's Disease At the core of the molecular mechanism underlying the generation of Alzheimer's disease lies the aberrant processing of the Amyloid Precursor Protein (APP) by β-secretase to form Aβ(l-42). It is at present unclear where precisely the aberrant cleavage occurs, but endosomes have been implicated as the compartment or one of the compartments where processing of APP occurs. This implies that 1) APP is transported to endosomes and follows the endocytic/recycling pathway and 2) β-secretase activity is localised to endosomes. Since Rab GTPases and their effectors/regulators regulate the trafficking of molecules (lipids and proteins) in the endocytic pathway, these molecules represent new targets for therapeutic agents aimed at decreasing the presence of APP in endocytic compartments and/or the presence of β-secretase activity in endocytic compartments, therefore inhibiting the generation of beta- amyloid and retarding and or preventing the onset/progression of Alzheimer's disease.

Based on the above experimental results, the object of the present invention was to develop an assay useful for the detection and identification of molecules (pharmaceutical drugs). In order to achieve this object, the present inventors developed the strategy that molecules (proteins) interacting with GTPases such as GTPases of the Rab family and, thus, functioning at the level of endosome trafficking and coordinating the interaction of these organelles with the cytoskeleton, can provide novel targets for molecules (pharmaceutical drugs) aimed at inhibiting cell locomotion and invasiveness (and thereby cancer) as well as infectious (both viral such as ATDS and bacterial and also eukaryotic), allergic, and inflammatory diseases, diabetes, Alzheimer's disease, endometriosis, and atherosclerosis. A further object of the present invention was to provide a kit comprising molecules interacting with GTPases (effector or regulator proteins/molecules; for exact definitions see further below), which kit is useful as a means for the detection and identification of pharmaceutical drugs exhibiting properties making them preferred drugs for treating cancer, inflammatory, allergic, and infectious diseases (eukaryotic, prokaryotic and viral) including ATDS but also atherosclerosis, endometriosis, diabetes and Alzheimer's disease. Hereinafter, these diseases are termed "the above mentioned diseases".

The following definitions are given in order to more fully specify the terms as used in the present application. "Invasiveness" refers to the migration and spreading of cancer cells or other cells such as endothelial cells (rather than the invasion of bacteria). "GTPase-interacting protein" or "regulator/regulatory protein/regulatory molecule" refers to any protein that interacts with the GTPase under any nucleotide conformation and for any purpose. GTPase-interacting proteins/regulators regulate the nucleotide-bound state of the GTPase, i.e., the GTPase activating protein (GAP) interacts with the GTP -bound GTPase and down regulates the activity of the GTPase by stimulating the hydrolysis of GTP. A GDP/GTP exchange factor promotes nucleotide exchange, thereby activating the GTPase. "Effector" or "effector/effector protein/effector molecule" is defined to mean a protein or a protein complex that interacts with the GTPase only if the GTPase is in the GTP -bound, also called active, form, that by binding stabilises this conformation and that mediates the "effect" of that GTPase. Effectors transmit the function of the GTPase, Rab is upstream and recruits the effectors which do the work. Regulators instead switch the Rab on and off. Rab-on (=GTP) binds the effectors and is functional. Thus, both effectors and regulators may therefore also be termed GTPase effectors and GTPase regulators, respectively. Additionally, it is apparent for the skilled reader that the term "regulator" is broader than the term "effector" and includes the latter.

The above mentioned strategy implies the first step of employing known effectors/regulators or identifying novel effectors/regulators of Rab or of other (small) GTPases, specifically regulating endosome trafficking and endosome-cytoskeleton interactions, and the additional step of designing an in vitro and/or in vivo assay to screen for inhibitors or activators of these effector/regulator molecules. In particular, the strategy includes the identification of substances useful as pharmaceutical drugs due to their interference with the process of cell locomotion, i.e., it includes the identification of inhibitors of membrane-cytoskeleton interactions.

The above object in mind, the present inventors established an assay system based on the use of such effectors/regulators, as defined hereinbefore, as targets for pharmaceutical drugs useful in the therapy of cancer and infectious diseases and any other disease which affects or requires the function of the endocytic/recycling trafficking machinery. According to a preferred embodiment of this aspect of the present invention, an assay is provided using both an effector/regulator and a Rab protein (or any other small GTPase). Accordingly, one aspect of the present invention is the use of the above mentioned GTPase effector/regulator molecules or proteins as targets in an in vitro or in vivo assay system to detect substances useful as a pharmaceutical/therapeutic agent for the prophylaxis and the treatment of the above mentioned diseases (including cancer, allergic, inflammatory, and infectious diseases, diabetes, endometriosis, atherosclerosis and Alzheimer's disease). Another object of the present invention is a kit useful for carrying out the assay of the invention. The kit necessarily contains at least one type of effector/regulator molecule. At least one GTPase effector/regulator molecule is either native and biochemically purified from a vertebrate, or recombinant and biochemically purified from bacterial cultures, from yeast cultures, or from other cultured eukaryotic cells, in either case labelled by a covalent modification or radioactivity suitable for use in the assay.

According to a preferred embodiment of this aspect of the present invention the kit additionally contains at least one type of GTPase and/or endosomal membrane fractions obtained by subcellular fractionation and labelled by internalisation of transferrin, either fluorescently labelled or labelled by any other modification that allows its detection, and/or cytosolic extracts, and/or an ATP-regenerating system and/or a number of chemicals to be tested for their suitability as an drug against any of the above mentioned diseases (cancer, endometriosis, diabetes, atherosclerosis, allergic, infectious, inflammatory diseases). Other specific and preferred embodiments of both the kit and the use according to the present invention may be readily taken from the claims attached hereto.

Drugs exhibiting any of the above activities include those molecules (including peptides) mimicking the peptide structure of the respective target molecule. Typical representatives of the above classes are molecules/substances that carry one or more of the following functional groups: a halide atom bound to an alkyl, alkenyl, alkinyl, or aryl residue, an alcohol group (primary, secondary, tertiary), an ether group, a carbonyl function (aldehyde or ketone), a carboxylic acid group, a carboxylic anhydride group, a carbamoyl group, a haloformyl group, a cyano group, an ester group including a lactone group, a benzyl, phenyl, tolyl, tosyl, sulfonyl group, an amino group (primary, secondary, tertiary), an isocyanate, a cyanate, a thioisocyanate, a thiocyanate, a carbamate, an azide, a diazo group, and a quinone group. Other representatives of suitable drugs are organometallic compounds, sterol moiety(ies)-containing molecules, β- hydroxy carboxylic acids, inorganic acids, and complexes such as metallocenes; nucleic acids, cytokines, hormones, antibodies, or oligopeptides comprising up to 20, preferably 8, 10, or 12, amino acid residues.

Preferably, the antibodies are polyclonal or monoclonal antibodies, or fragments thereof, humanised or even human antibodies (obtained for example by a phage display method), inhibitory or stimulatory, raised against and targeted towards any of the aforementioned GTPase effectors. A preferred nucleic acid is genomic DNA, cDNA, or mRNA, or a fragment thereof, an oligonucleotide, an oligoribonucleotide, all being based on or derived from any of the GTPase effector gene sequences having any of the sequences as depicted in SEQ ID NO: 1, 3, 5, 7, 9, 11, 13, and 15, or gene therapy vectors derived from the aforementioned GTPase effector gene sequences.

The advantage (and the gist) of the present invention consists in (1) the development of a comprehensive screening method based on molecular networks, (2) the sensitivity of the assays, and (3) the selectivity of the targets. The activity of the factors under control of the central switch can be responsible for coordinating various interconnected functions such as (a) endocytosis, (b) the distribution, i.e., sorting of proteins and lipids within endosomes and from endosomes to other intracellular locations (e.g. interactions of phagosomes with endosomes and lysosomes), (c) the motility of endosomes along the cytoskeleton, and (d) cell migration. This multiplicity of factors responsible for a wide number of cellular functions is defined as a molecular network. The present invention resides in enabling the skilled person to study the molecular network of many different GTPases, in particular those GTPases localised to endosomes, e.g., Rab5, Rab4, Rabl l, Rab7, Rabl7, Rabl8, Rab22, as a whole in order to establish various (GTPase) effector/regulator proteins that are suitable as targets for pharmaceutical drugs. In this approach, the GTPase itself will not serve as a target molecule because its activity is so crucial for cell homeostasis that it would likely invariably result in toxic effects. Rather, the effector/regulator molecules specifically functioning in protein and lipid sorting, endosome motility and endosome- cytoskeleton interactions will be the target. For example, compounds that affect cell motility could be screened more specifically without affecting other important cellular functions (i.e. endocytosis). In the alternative, and according to a preferred embodiment of the present invention, the target may well be one of the effectors/regulators as defined above, but additionally a GTPase such as Rab5, Rab4, Rab7, Rabl l, or Rab 17, will be present in the assay system. The present inventors have functionally characterised several Rab GTPases (e.g., Rab5, Rab4, Rabl l, Rab 7, Rabl7, Rabl8, Rab20) and effectors/regulators of these GTPases, localised to the early endosomes and the recycling endosomes, which GTPases and effectors/regulators, respectively, are thus suitable for the assay according to the invention. These Rab proteins sequentially control transport in the endocytic and recycling pathway. Rab5 controls transport from the plasma membrane to early endosomes whereas Rab4 and Rabl l function in protein sorting and recycling from early and recycling endosomes back to the plasma membrane. The expression of Rab 17 is restricted to polarised cells such as epithelial cells and neurons. This latter GTPase regulates trafficking through the apical recycling endosome, thus contributing to the generation and maintenance of cell polarity. Thus, effectors/regulators of all these GTPases and the respective GTPases (in combination with their effectors/regulators) may be employed in accordance with the present invention in an in vitro or in vivo assay system as a target in order to screen for substances that are useful as a pharmaceutical agent for the prophylaxis and treatment of the above mentioned diseases, i.e., cancer (benign tumours, malignant tumours, carcinomas, sarcomas, melanomas, leukemias, gliomas, or neuroblastomas), diabetes, Alzheimer's disease, endometriosis, atherosclerosis, allergic, inflammatory, and infectious diseases such as ATDS tuberculosis, cholera, malaria, pseudotuberculosis, gastroenteritis, enteric fever, typhus, and any other infectious diseases caused by an infectious agent that infects cells by the endocytic route and resides intracellularly in phagosomes escaping the cellular killing mechanisms (in this regard, it should be recalled that the GTPases per se and alone can not serve as a target for the drugs to be detected; rather, the target for the drug is an effector/regulator molecule). Other examples of infectious diseases are those diseases caused by the following pathogens (bacteria or organisms): Mycobacterium, Staphylococcus, Toxoplasma, Trypanosoma, Listeria, Salmonella, Legionella, Leishmania, Coxiella, Shigella, Yersinia, Neisseria, Vibrio, Bartonella, and any other intracellular pathogen that resides in intracellular phagosomes and escapes cellular killing mechanisms.

As the proteins described herein, i.e., the Rab proteins and their effectors/regulators, are, unless exceptions, ubiquitously expressed, pharmaceutical agents for the prophylaxis and treatment of tumours of any cellular origin can be screened for and detected according to the present invention. Specifically, the following tumours may be mentioned: lung carcinoma, osteosarcoma, lymphoma, leukemia, soft tissue sarcoma, breast carcinoma, bile cancer, cervix carcinoma, cancer of the (small) intestine, of the kidneys, of the cavity of the mouth, penis carcinoma, ovary cancer, stomach cancer, cancer of the tongue, brain cancer, bladder carcinoma, prostate carcinoma, liver carcinoma, and carcinoma of the pancreas and every tumor that invades other tissues and organs.

Drugs with anti-cancer and/or an activity against any of the other above mentioned diseases include those molecules (including peptides) mimicking the peptide structure of the respective target molecule. Typical representatives of the above classes are molecules/substances that carry one or more of the following functional groups: a halide atom bound to an alkyl, alkenyl, alkinyl or aryl residue, an alcohol group (primary, secondary, tertiary), an ether group, a carbonyl function (aldehyde or ketone), a carboxylic acid group, a carboxylic anhydride group, a carbamoyl group, a haloformyl group, a cyano group, an ester group including a lactone group, a benzyl, phenyl, tolyl, tosyl, sulfonyl group, an amino group (primary, secondary, tertiary), an isocyanate, a cyanate, a thioisocyanate, a thiocyanate, a carbamate, an azide, a diazo group, and a quinone group. Other representatives of suitable drugs are organometallic compounds, sterol moiety(ies)-containing molecules, β-hydroxy carboxylic acids, inorganic acids, and complexes such as metallocenes; nucleic acids, cytokines, hormones, antibodies, or oligopeptides comprising up to 20, preferably 8, 10, or 12, amino acid residues. Effectors transmit the function of the GTPase, Rab is upstream and recruits the effectors which do the work. Regulators instead switch the Rab on and off. Rab-on (=GTP) binds the effectors and is functional.

To elucidate the molecular mechanism whereby these GTPases exert their function in membrane trafficking and cytoskeleton organisation, the present inventors have developed a method to biochemically identify Rab5 protein effector molecules (Chistoforidis et al., 1999a). This method is based on previously described affinity chromatography (Amano et al., 1996) but has been optimised to result in the specific elution and large-scale purification of Rab effectors and GTPase interacting proteins in general, in amounts sufficient for both their identification by microsequencing techniques and their functional characterisation (Chistoforidis et al., 1999a).

In addition to that method, the present inventors have developed a novel method. This method is described in Example 3 below. With this method, the inventors have re-purified several of the proteins that bind Rab5. Using this modified affinity chromatography procedure, more than 20 effector and/or regulator proteins interacting directly or indirectly with Rab 5 (Figure la in (Chistoforidis et al., 1999a)) have been identified and demonstrated to exhibit functional activity as shown by their ability to substitute cytosol in an in vitro endosome fusion reaction (Chistoforidis et al., 1999a). Some of these molecules such as Rabex-5 regulate the activity of Rab5 (i.e., its nucleotide state) (Horiuchi et al., 1997). Others such as Rabaptin-5 and EEAl participate in the endosome docking and fusion process (Stenmark et al., 1995). Still others regulate the ability of endosomes to move along microtubules, and all of these proteins are suitable candidates as a target for the detection of chemical drugs (compounds) that may turn out to be highly efficient agents to combat the above mentioned diseases.

In addition, expression of an activated mutant of Rab5 (Rab5Q79L) induces an increased recruitment of actin on the expanded endosomes suggesting that Rab5 either directly or indirectly regulates the attachment/nucleation of actin filaments with the endosome membrane. Based on the above data, in particular in view of this large number of Rab 5 interacting molecules (more than 20), the present inventors concluded that the activity of Rab5 is not restricted to endocytic membrane docking and fusion. Most importantly, they also concluded that modulating the activity of Rab5, and therefore of its effectors/regulators, would result in changes in organelle motility in vivo and in cell motility. This is perfectly consistent with the effect observed by the present inventors that expression of activated Rab5 leads to an increase in motility of fibroblasts and endothelial cells, implicating this GTPase in the regulation of cell migration. The effector/regulator protein(s) responsible for this activity (to regulate membrane trafficking and membrane-cytoskeleton interactions including the ability of endosomes to move along microtubules) are therefore, in accordance with the present invention, specific targets for drugs aimed at inhibiting the ability of the Rab5 machinery to sustain cell migration.

Similar experiments have been and are being conducted on two other GTPases functioning in endosome trafficking, Rab4 and Rabl l. Several candidate proteins for Rab4 effectors/regulators have been purified and one of these proteins has been sequenced and identified as the human homologue of the yeast Adenyl Cyclase-associated protein CAP (Field et al., 1990). This protein is a component of the Ras-activated cyclase complex in yeast and therefore is required for the proliferative signal in yeast. In addition, this protein has been implicated in the interaction with the actin cytoskeleton, since the carboxy terminus of CAP binds actin. Another protein which has been identified is Rabenosyn-5, a Rab5 effector. The realisation that this molecule binds both Rab5 and Rab4 but does not interact with Rabl 1, suggests that it may control trafficking between the Rab5 and Rab4 sub-compartments within early endosomes, and therefore regulate the sorting of proteins trafficking through this organelle. This hypothesis is consistent with the new experimental data recently obtained (see below). Given the role of Rab proteins in the regulation of membrane transport and membrane- cytoskeleton interactions, the effectors/regulators for these small GTPases are targets for pharmaceutical drugs aimed at (1) inhibiting cell invasiveness and (2) enhancing cell-defence mechanisms against intracellular pathogens, such drugs therefore being valuable weapons in the battle against tumor (and cancer) and all other of the above mentioned diseases. For example, it may be possible to inhibit the stimulatory activity of Rab5 to reduce cell motility and consequently to reduce metastasis and the progression of cancer.

A strategy similar to that established and described above for Rab5 has been applied for other Rab proteins regulating endocytic trafficking. For example, the small GTPases Rab4, Rabl 1, and Rab 17 reside onto and regulate trafficking through the early and recycling endosomes. Given that in motile cells recycling endosomes preferentially deliver membranes to the leading edge of the cell and given that the deposition of adhesion molecules to the leading edge of motile cells depends on the endocytic cycle between endosomes and the plasma membrane, the established functional role of Rab proteins functioning on endosomes, including Rab4 and Rabl l characterised by the present inventors, implies that they may participate in the regulation of cell motility. Therefore, the effectors/regulators of these and other endosomal GTPases will also represent novel potential targets according to the instant invention.

On the other hand, further Rab effectors/regulators to be identified and characterised in the future will also act in various aspects of endosome function, including endosome and cell motility, and endosome-phagosome interactions. An approach to identify further effectors/regulators has been described (Chistoforidis et al, 1999a), and the procedure described by the authors is similarly suitable to detect, identify, and characterise further, as yet unknown, effector/regulator molecules (proteins) of small GTPases of the Rab family.

The multiplicity of Rab5 (or any other Rab GTPases) effector/regulator molecules serves to execute distinct functions in endosome dynamics and cell dynamics all coordinated by the GTPase switch of Rab5 (or any other of the GTPases). It is therefore a "package" also referred above as a network of molecules that is necessary to coordinate different functions. These molecules can operationally be divided into the following classes on the basis of either their established identity and/or function and on the basis of their possible function for which different claims in drug discovery can be proposed. The effectors/regulators of all classes may be used for the assay as defined herein, i.e., for the detection of suitable drugs for the treatment of any of the above mentioned diseases.

Class 1 : Representative effectors/regulators: Rabaptin-5, Rabaptin-5b, Rabex-5, RN Tre, EEAl, PI 3-K, Inositol Polyphosphate 4-phosphatase and Inositol Polyphosphate 5-phosphatase. Function: endocytic trafficking Assay: endosome fusion, endosome motility

This class of Rab5 effectors/regulators includes molecules that 1) have either been already described as functionally linked with Rab5 (Rabaptin-5, Rabaptin-5b, Rabex-5, EEAl, PI 3-K), or 2) have previously been identified and whose DNA and deduced protein sequence have been disclosed (RN Tre, Inositol Polyphosphate 4-phosphatase and Inositol Polyphosphate 5- phosphatase) but that have been now functionally linked to Rab5 by the present inventors. Rab5 effectors/regulators of this class 1 belong to the "package" also referred above as a network of molecules that is necessary to coordinate different functions of Rab5.

Since Rab5 regulates endocytosis, effectors/regulators of this molecule also control endocytosis as well as all functions controlled by Rab5; i.e., endocytic trafficking, intracellular distribution of endosomes, motility of endosomes along microtubules, and cellular motility. Consequently, class 1 Rab5 effectors should be considered potential targets for the therapeutic strategies above described in conjunction with the other classes of Rab5 effectors/regulators.

Class 1 includes

1) Regulators that play a role in the nucleotide cycle of Rab5. Rabex-5 has been identified as a nucleotide exchange factor for Rab5 (Horiuchi et al., 1997), and therefore converts the GTPase from the inactive, GDP -bound, to the active, GTP-bound form. Consequent to the activity of Rabex-5 is the interaction between Rab5 and the Rab5 effectors that transmit Rab5 function. The present inventors have discovered that RN Tre (Matoskova et al., 1996), a protein related to the Tre oncogene, is a Rab5 GAP, which discovery provides an interesting link between the signal transduction and membrane trafficking machinery. Most importantly, like Rabex-5 that regulates activation of Rab5, RN Tre regulates its inactivation and therefore represents an ideal target for drugs aimed at interfering (positively and negatively) with Rab5 function. 2) Effectors that play a role in endosome docking and fusion: Rabaptin-5 (Stenmark et al., 1995), Rabaptin-5b (Gournier et al., 1998), and EEAl (Chistoforidis et al., 1999a; Simonsen et al., 1998), or in the regulation of this process via the generation of phosphoinositide-3-phosphate (PI 3-K) which is necessary for the membrane recruitment of EEAl (Christoforidis et al., 1999b). Two distinct phosphatidyl-inositol 3-kinases (PI 3-K), hVPS34 and p85a/pl l0b, have been identified among the Rab5 effectors. The two kinases are differentially distributed along the endocytic pathway, and the activity of hVPS34 is specifically required for EEAl membrane recruitment and endosome fusion. It has also been demonstrated that VPS34 regulates the motility of endosomes along microtubules, suggesting that PI 3-K activity is necessary to coordinate the function of various Rab5 effectors/regulators, including those controlling membrane-cytoskeleton interactions (Nielsen et al., 1999). PI 3-K are therefore considered to represent a particularly suitable target of drugs that block the motility activity. Similarly, targets other than PI 3-K may be identified among the Rab5, Rab4, and Rabl l effectors/regulators that couple the activity in membrane docking and fusion with the organelle dynamics. One such target coupling the activity in membrane docking and fusion with the organelle dynamics is Rabaptin-5 due to its capacity to couple Rab4 to Rab5 function.

The present inventors have determined that the list of Rab5 effectors includes two more proteins that play a role in phosphoinositide metabolism: Inositol Polyphosphate 4-phosphatase and Inositol Polyphosphate 5-phosphatase. First tests conducted by the present inventors have proven that both these phosphatases bind to a Rab5 affinity column and directly interact with Rab5 in a yeast two-hybrid system assay. The function of Inositol Polyphosphate 4-phosphatase and Inositol Polyphosphate 5-phosphatase is to catalyse the removal of the phosphate from the 4- and 5-position of the inositol ring, respectively. In the context of the Rab5 network of effectors/regulators, these proteins are functionally linked to the PI3-Ks, especially p85a/pl l0b, since Inositol Polyphosphate 4-phosphatase and Inositol Polyphosphate 5-phosphatase would use as substrate the product of this kinase. The discovery that 1) Rab5 regulates a sophisticated cycle of phosphorylation/dephosphorylation of phosphoinositides and 2) that phosphoinositides regulate the recruitment of factors playing a role in the endosome transport machinery as well as signaling molecules, implies that Rab5 and its effectors/regulators can be used as drug targets for diseases where alterations in the function of endosomes can result in therapeutic intervention.

The class 1 effector/regulator molecules may be used according to the present invention as drug targets in cancer on the basis of regulating cell motility and invasiveness. The inventors of the present invention have found out that Rab5 regulates the motility of endothelial cells in vitro

(this result was the outcome of experiments described in the Examples below). Since the Rabaptin-5/Rabex-5 complexes regulate the nucleotide cycle of Rab5, and since the kinases PI 3- K regulate the recruitment of effectors/regulators, these molecules have been identified as targets for drugs aimed at inhibiting the activity of Rab5 in cell motility. The class 1 effector/regulator molecules may also be used as drug targets in cancer on the basis of the signalling molecules regulating cell growth. Growth factors and growth factor receptors are transported through the endosomes before being degraded in the late endosomes and lysosomes. Molecules that regulate the trafficking of receptors along the endocytic pathway can be used as targets for drugs aimed at interfering with transport (i.e., clearance from the plasma membrane, inhibition of recycling, transport to endosomes where signalling molecules are located, increase transport to degradative compartments where receptor-ligand complexes are degraded) and, consequently, inhibiting mitogenic signalling. Finally, the class 1 effector/regulator molecules may also be used as drug targets in infectious diseases (see the above list exemplifying infectious diseases particularly suitable according to the present invention). The small GTPase Rab5 has been observed to accumulate on Mycobacterium tuberculosis phagosomes. Interestingly, the Rab5 effector Rabaptin-5 is not present on the phagosome membrane. Thus, it must be concluded that there are differences between endosomes and phagosomes in the utilisation of the Rab5 effectors/regulators. In the case of Listeria monocytogenes phagosomes, it has been observed that an upregulation in expression of the small GTPase Rab5 occurs in the course of infection of macrophages by a Listeria monocytogenes non-lytic mutant. This increased Rab5 recruitment results in an increased fusogenicity of the phagosomes with endosomes leading to intracellular killing of the bacterium. Accordingly, in order to detect and identify substances suitable for the theurapeutic treatment of pathogen infections, the Rab5 effector/regulator responsible for this effect may be used as a target in the assay described in detail further below. In this case, the upregulation of the host cell defence mechanism could be accomplished by a suitable drug. The activator of Rab5, Rabex-5, that functions as a nucleotide exchange factor and converts the protein into the GTP -bound active form (Horiuchi et al., 1997) is a particularly suitable drug target in this regard. By enhancing its activity (by means of a drug activating Rabex-5), the inventors have been able to up-regulate the Rab5 machinery, stimulate endosome fusion and facilitate phagosome-endosome interactions. Alternatively, other Rab5 effectors/regulators (still to be identified) playing a role in endosome trafficking could serve as targets.

Class 1 as well as the other classes of Rab5 effectors described below may include molecules playing a role in endosome motility. The present inventors have found that Rab5 regulates the motility of early endosomes in vivo and both the attachment of early endosomes to microtubules and the motility of early endosomes along microtubules in vitro (Nielsen et al., 1999). New data is that a Rab5 eluate is sufficient to reconstitute motility in an endosome-microtubule in vitro motility assay and in the absence of cytosol. Therefore a Rab5 effector(s) is/are sufficient to regulate or switch on a microtubule motor, or a Rab5 effector(s) is/are a microtubule motor. For this function we make three statements to support a claim:

1. Rab5 is required to bind certain or possibly all early endosomes to microtubules.

2. Rab5 effectors and possibly their binding partners (i.e. a Rab eluate) are required for endosome motility on microtubules.

3. RabGDI blocks all endosome motility in vivo (microinjection of RabGDI). These data demonstrate an essential role for Rab5 effectors in binding endosomes to microtubules and their subsequent motility along microtubules. Rab5 effectors can therefore be used as target molecules for drugs aiming to reduce endosome motility to block endosome cycling and, consequently, recycling of surface molecules necessary for cell motility.

Class 2

Representative effectors/regulators: Rabankyrin-5 and Rabenosyn-5 (pi 10 FYVE) Function: endocytic trafficking, possible interaction with the actin cytoskeleton . Assay: endosome fusion, endosome motility

This class includes at least two proteins which share the same PI(3)P-binding domain, the FYVE-finger. One protein is the human homologue of mouse Ankhzn (Ito et al., 1999), to be renamed Rabankyrin-5, given its ability to bind Rab5. The present inventors also found a Rabankyrin-5 splice variant (SV). Rabankyrin-5 contains ankyrin repeats. It is localised to early endosomes, binds Rab5, plays a role in endosome fusion, and is required for early endosome fusion. As regards the binding to Rab-5, it is worth noting that the present inventors determined the Rab5 binding region to reside in the ankyrin repeats in the C-terminus of the protein (between residues 542 and 1075).

The sequence of the gene coding for human Rabankyrin-5 is depicted in the sequence listing attached to this application (SEQ ID NO: 1) whereas the sequence of the protein itself is given in the sequence listing as SEQ ID NO: 2. The nucleotide and amino acid sequences of the SV gene and protein are indicated in the sequence listing as SEQ ID NO: 3 and SEQ ID NO: 4, respectively. To the best knowledge of the present inventors, there is no prior art that would suggest a role for the gene/gene product of ankhzn - or of its human analogues - as a Rab5 effector with a role in endosome fusion and membrane trafficking. The second molecule is Rabenosyn-5, previously named pl lOFYVE. It is localised to early endosomes as well, binds Rab5 as well, and is also known to be complexed with and bind to the mammalian homologue of yeast Vps45p, a protein involved in transport to the vacuole. This binding demonstrates that Rabenosyn-5 has a role in SNARE-mediated membrane fusion. Rabenosyn-5 (pi 10FYVE) probably also plays a role in the regulation of transport and participates in the movement of endosomes along the actin and microtube cytoskeleton.

The nucleotide and amino acid sequences of pi 10FYVE are given in SEQ ID NO: 5 and SEQ ID NO: 6, respectively. Further properties of the effector Rabenosyn-5 (pi 10 FYVE) are that it is essential in endosome fusion, fusion of clathrin coated vesicles derived from the plasma membrane with early endosomes. Over-expression of Rabenosyn-5 inhibits the processing of lysosomal enzymes.

The present inventors have found that Rabenosyn-5 is a Rab4 effector. It has been known that three small GTPases, i.e., Rab5, Rab4, and Rabl l, exist in separate domains on the same endosome. The present inventors have been able to demonstrate that over-expression of Rabenosyn-5 brings the Rab5 and Rab4 domains together, without changing the distribution of Rabl l, and changes the sorting properties of the endosome with respect to the cycling of certain receptors (e.g. transferrin receptor). This proves that Rabenosyn-5 has a function in receptor sorting and provides validation that manipulating Rabenosyn-5 has the potential to control receptor amount, distribution, and the kinetics of receptor cycling.

Class 3

Representative effectors/regulators: multi-protein complex including pi 00, p95, p60, p45, p25 Function: endocytic trafficking, possibly in polarised cells such as epithelial cells and neurons Assay: endosome fusion, endosome motility

This complex consisting of the above 5 proteins (pi 00, p95, p60, p45, p25) and exhibiting the molecular weights of 100 kDa, 95 kDa, 60 kDa, 45 kDa, and 25 kDa, respectively, has been isolated by the present inventors. At least one of these molecules, or the entire complex, interacts either directly with Rab 5 or indirectly through another Rab 5 -binding protein. The present inventors have also determined that the complex is detected in a clathrin coated vesicles-enriched fraction. This subcellular fraction includes the vesicles that transport protein receptors from the plasma membrane to the early endosomes.

The structure (sequence) and function of the 5 protein components have been investigated to some extent and are more or less known now and form part of the present invention.

In detail, the C-terminal half of the 100 kDa component shares sequence homology (52%) with the conserved catalytic domain of GTPase activating proteins and spindle assembly checkpoint proteins from yeast to mammals, while the N-terminal half shares homology (57%) with the conserved domain of serine/threonine protein kinases. Fusion between early endosomes is strongly inhibited when cytosol is immuno-depleted using anti-45 kDa component antibodies.

Immunofluorescence data show that the 45 kDa component is localised in intracellular structures which are not EEAl positive. Upon over-expression of Rab5Q79L though, there is partial co- localisation of these proteins on early endosomes.

The nucleotide and amino acid sequences of these proteins have been determined and are depicted in the attached sequence listing (p60: SEQ TD NO: 7 and SEQ ID NO: 8; p45: SEQ ID NO: 9 and SEQ ID NO: 10; p25: SEQ ID NO: 11 and SEQ ID NO: 12, plOO: SEQ ID NO: 13 and SEQ ID NO: 14; p95: SEQ ID NO: 15 and SEQ ID NO: 16, respectively). The p95 sequences are only partial sequences, however.

Examples Example 1 :

The following example describes the materials and methods used to measure the effect of a GTPase or mutant GTPase on the regulation of cell motility.

Expression constructs Myc-rhoDV26G (Murphy et al., 1996) was cloned into the Hindltl-Xbal sites of expression vector pHSVPUC (Geller et al., 1993) to generate construct: pHSV-myc-rhoDV26G. Myc- rhoDV26G cDNA was cloned in frame into the EcoRI-BaniHI sites of expression vector pEGFP- C2 (Clontech) to generate a green fluorescent protein (GFP) fusion: GFP-myc-RhodV26G. Transient transfection ofrhoDV26G and human transferrin receptor

BHK-21 cells were trypsinized 24h prior to transfection and were seeded onto 11mm glass coverslips. Cells were infected with T7 RNA polymerase recombinant vaccinia virus (Bucci et al., 1992) and transfected with either T7-myc-rhoDV26G and T7-human transferrin receptor (T7-hTR) or T7-hTR alone (Murphy et al., 1996) using DOTAP (Boehringer Mannheim, Mannheim, Germany). The cell number and also the amount of DNA and lipid were constant. The final concentration of DNA per coverslip was lμg.

Following transfection, the cells were incubated at 37°C, 5% CO2, for 2h, cyclohexamide was then added for lh. Cells were washed 8 times for lmin in preheated medium and rhodamine transferrin (50 μg/ml) was uptaken in a preheated humidified chamber at 37°C, 5% CO2, for 20min. Coverslips were washed in preheated medium and then mounted into chambers for video microscopy (Bradke and Dotti, 1997; Bradke and Dotti, 1996) as described below. TLVM (time lapse video microscopy) was carried out immediately. Hydroxyurea was present at all times to prevent late viral gene expression.

Transient transfection of T7-rhoDV26G and T7-gfp

To monitor lysosomal motility in the presence and absence of rhoDV26G we made use of a green fluorescent protein-expressing vector cloned into pGEMl harbouring the T7 promoter. Co- transfection of this plasmid with T7-myc-rhoDV26G allowed us to identify the transfected cells for video microscopy. Handling of the cells was as outlined above with the exception that they were incubated in the presence of 50 nM Lysotracker (Molecular Probes) for 15min at 37°C, 5% CO to label the lysosomes, and TLVM was carried out as outlined below.

BBCE cell motility

To address the effect of rhoDV26G on cell motility, BBCE cells were trypsinized and plated at a density of 1,500 cells per cm , 24h later nuclear microinjection was carried out with a RhoD V26G expression construct at 50μg/ml concentration, FITC dextran was coinjected to identify the injected cells. Approximately, 12h following injection the cells were stimulated with bFGF and subjected to TLVM.

Time-Lapse Video Microscopy The cells for TLVM were grown on coverslips and mounted into aluminum chambers as described in (Bradke and Dotti, 1997). Briefly, aluminum was cut into 7.5cm x 2.5cm x 2mm rectangles, the dimensions are similar to a standard glass slide and fit onto the microscope stage. In the centre a circular hole was cut of 8mm diameter, and around it 4mm area was milled from both sides. A glass coverslip was inserted onto one side of the chamber, sealed with a lubricant, medium was then added to fill the chamber, cells grown on coverslips were then inverted onto the other side of the metal slide and sealed with lubricant.

Image and statistical analysis of the data A dedicated automatic program was developed to detect and track endosomes/lysosomes as they move. It runs on a SPARC station Ultral (SUN, Mountain View, CA) to which a Series 151/40 digital image processor (Imaging Technology, Bedford, MA) is connected. The detection of fluorescent spots corresponding to endosomes/lysosomes is performed automatically by a multiresolution algorithm based upon selectively filtering an undecimated wavelet decomposition of the image through the use of wavelet coefficient thresholding and correlation. At the end of this step, all endosomes/lysosomes in the sequence are characterised and their coordinates are determined and stored. A tracking algorithm is then used to establish valid trajectories. The algorithm uses a first-order Kalman filtering approach whereby at each frame and on the basis of the previous ones, predictions of the endosome/lysosome positions are established and compared with the computed ones. The best matches are selected as trajectory points and tracks are finally analysed to compute the data that was used for generating values in Tables 1 and 2. For each data set, a sequence of 30 images was analysed. Speed values reported here are mean values of the speed of several endosomes/lysosomes and errors are standard deviations of these group estimated.

Example 2:

Drugs that have been found by employing the assay of the present invention and that enhance or inhibit Rab5 effector/regulator activity can be distinguished and classified on the basis of their mode of action (activities 1 to 5, see below). Accordingly, 5 classes of drugs will be introduced (and searched for) that exhibit a specific mode of action on the target molecules, i.e., on the effectors or regulators:

1. Drugs that inhibit Rab5 activation. Such drugs would in principle act on the whole Rab5 network and are therefore less specific than drugs of classes 3 and 4 because they act on a regulator (protein or molecule) that affects the nucleotide state of Rab5 and, consequently, the ability of Rab 5 to signal to the totality of Rab 5 effectors rather than on a specific effector (protein or molecule).

2. Drugs that enhance Rab5 inactivation. Such drugs would in principle act on the whole Rab5 network and are less specific than drugs of classes 3 and 4 for the reason given for class 1 above.

3. Drugs that disrupt specific Rab5-effector interactions, e.g., between Rab5 and pi 10 FYVE.

4. Drugs that inhibit a specific Rab5 effector activity, e.g., endosome fusion, endosome- microtubule interaction, endosome motility along microtubules.

5. Drugs that inhibit cell motility.

Class 3 and 4 drugs are the most specific and are therefore preferred according to the present invention. However, drugs of the other classes, in particular of classes 1 and 2, can nevertheless be considered as similarly suitable agents to combat at least one of the above mentioned diseases.

Assay for Activity 1

The assay measures Rabex-5 dependent nucleotide exchange of Rab5. The drug could perturb Rab5-Rabex-5 interaction or inhibit the catalytic activity of Rabex-5. The result expected is an inhibition of GDP/GTP exchange of Rab5. This assay is based on the assay published in (Horiuchi et al., 1997) and may well be adequately adapted for high throughput screening. It will be readily feasible, therefore, to screen thousands of compounds for their potential anti-cancer or anti-infectious disease activity/potential.

GDP/GTP exchange activity. Unless otherwise specified, the standard [ SJGTPγS-bindmg assay was performed by the filter method (Sasaki et al, 1990) by incubating 200nM of recombinant Rab5 with lμM of [³⁵S]GTPγS (20,000cpm/pmol) at 37°C for lOmin with 5mM of MgCl2 in the presence of 20nM recombinant Rabex-5.

Substances that are able to enhance or inhibit nucleotide exchange above or below the standard activity measured in the presence of Rabex-5, respectively, may be searched for.

Assay for Activity 2

The assay measures the GTPase Activating Protein (GAP) RN-Tre-dependent stimulation of

GTP hydrolysis by Rab5. The drug could stabilise the Rab5-GAP interaction or stimulate the catalytic activity of GAP. The result expected is an increase in the inactive-form of Rab5, i.e., an increased amount of GDP-bound Rab5. This assay is based on the following assay:

GAP assays with purified recombinant RN-Tre fusion proteins were performed by the filter method (Sasaki et al., 1990) using 2μM of substrate [γ-³²P]-GTP-loaded Rab5 in the presence of recombinant RN Tre GAP domain (lOOnM), at 30°C for 2min. GAP activity was expressed as the percentage of non-hydro lysed [γ- P]-GTP which remained bound to the filters, relative to the radioactivity at time 0. Substances that are able to increase the GTP hydrolysis activity compared with the standard activity measured in the presence of RN Tre should be searched for. The assay may adequately be adapted for high throughput screening.

Assay for Activity 3

Drugs may be screened for their ability to inhibit specific Rab5-Rab5 effector interactions. The assay consists in the binding of an effector molecule (either in vitro translated or as purified recombinant protein; native or modified, e.g., by biotinylation) to GST-Rab5:GTPgS (GTPgammaS) immobilised on a matrix. This assay is based on the procedure developed by Christoforidis et al. (1999a and b), but may be readily adapted for high throughput screening.

GST-Rab5 affinity chromatography is performed as described previously (Chistoforidis et al., 1999a). Beads with GST-Rab5 were incubated with the nucleotide exchange buffer (NE buffer) containing 20mM Hepes, lOOmM NaCl, lOmM EDTA, 5mM MgCl2, ImM DTT, ImM GTPγS, pH7.5, for 90min at room temperature under rotation. Afterwards, NE buffer was removed and the nucleotide was stabilised with NS buffer containing 20mM Hepes, lOOmM NaCl, 5mM MgCl2, ImM DTT, ImM GTPγS, pH7.5, for 20min at room temperature under rotation. In vitro transcription-translation of S-methionine labeled proteins is performed according to Manufacturer's instructions (Promega). Recombinant proteins or transcribed-translated proteins (50μl of standard reactions) are incubated for 2h at 4°C with 20μl glutathione sepharose beads coupled with GST-Rab5:GTPgS, and 150μl buffer containing 20mM Hepes, lOOmM NaCl, 5mM MgCl₂, ImM DTT, and ImM GTPgS, respectively, such that the final volume is 220μl. The incubation is performed both in the presence and in the absence of the candidate drug. When the drug is dissolved in an organic solvent the buffer in the control reaction will be supplemented with the solvent in the same concentration and amount added in the reaction in the presence of the drug. Usually, however, the buffer/solvent dissolving the drug is the same buffer used for the assay, i.e., the buffer containing 20mM Hepes, lOOmM NaCl, 5mM MgCl₂, ImM DTT, and ImM GTPgS. Beads are subsequently washed 4x with buffer containing 20mM Hepes, lOOmM NaCl, 5mM MgCl₂, ImM DTT, lOμM GTPgS, lx with the same buffer but now containing 250mM NaCl, and lx with 20mM Hepes, 250mM NaCl, ImM DTT. Elution of bound proteins is performed as described before (Chistoforidis et al., 1999a; Christoforidis and Zerial, 1999c) and eluted proteins are loaded on SDS-PAGE gel followed by immunoblotting for recombinant proteins or autoradiography for in vitro translated proteins.

The assay measures the activity of drugs that are capable of disrupting the association of a specific or any Rab5 effector with Rab5 in a concentration-dependent manner under the same experimental conditions that allow the interaction. For example, a drug x will disrupt the interaction of effector X with Rab5:GTPgS in a concentration-dependent manner. This drug will then be used as inhibitor of effector X in the biological process X.

A variation of the assay is to use magnetic beads in an assay of the ELISA-type (similar to the in vitro endosome fusion assay described below). GST-Rab5 was biotinylated using chemical crosslinking (Pierce) and nucleotide exchange was performed. Beads with GST-Rab5 were incubated with NE buffer containing 20mM Hepes, lOOmM NaCl, lOmM EDTA, 5mM MgCl2, ImM DTT, ImM GTPgS, pH7.5, for 90min at room temperature under rotation. Afterwards, the NE buffer was removed and the nucleotide was stabilised with NS buffer containing 20mM Hepes, lOOmM NaCl, 5mM MgCl2, ImM DTT, ImM GTPgS, pH7.5, for 20min at room temperature under rotation. Beads (lμl) were subsequently incubated with lμg recombinant purified Rab5 effector coupled to a ruthenium trisbipyridine chelate (IGEN) in 20μl NS buffer in the presence or absence of chemical drug candidates. Beads were then washed three times with 500μl NS buffer and binding of the effector to Rab5 is measured with an Origen Analyzer (IGEN) according to the manufacturer procedure. The background signal (binding to Rab5:GDP, i.e., without nucleotide exchange) is deducted. Drugs that affect the interaction between Rab5 and the Rab5 effector decrease the signal in comparison with the signal obtained when the reaction is performed in the absence of the drug.

Assays for Activity 4 The present inventors have developed three main assays to measure Rab5 and Rab5 effector activity.

(i) In vitro fusion assay (Horiuchi et al., 1997)

This assay is intended to screen for drugs that inhibit endocytosis, clathrin coated vesicle (CCV) fusion with early endosomes, early endosome homotypic fusion and endosome trafficking in tumor cells or enhance endocytic membrane fusion in bacterial parasites infected cells.

Two distinct enriched populations of early endosomes labeled with either biotinylated transferrin or a sheep α-human transferrin antibody are prepared from sHeLa (suspension HeLa) cells by sucrose sedimentation (Gorvel et al., 1991). The basal fusion reaction consists of the two enriched endosome populations incubated for 45minutes at 37°C in the presence of 3mg/ml of HeLa cytosol, unlabelled transferrin and an ATP regeneration system (17.3mM creatine phosphate, 87μg/ml creatine kinase, and 2.2mM ATP) and in the presence or absence of the candidate drug. The fusion is quantified by incubation of the fusion mix with wash buffer (50mM Tris pH7.5, lOOmM NaCl, lg/1 BSA and 2% (w/v) Triton X-100) and streptavidin-coated magnetic beads (Dynal). After two washes in wash buffer the samples are incubated with a rabbit α-sheep secondary antibody coupled to a ruthenium trisbipyridine chelate (IGEN) and measured with an Origen Analyzer (IGEN). The background signal (5-10% of the fusion signal) is deducted and the data expressed as the percentage of the basal fusion reaction. The CCV-early endosome fusion assay is performed in the same way with biotinylated transferrin-labeled CCV and α-transferrin antibody-labeled early endosomes.

Substances that are able to inhibit CCV-endosome or early endosome fusion below the standard fusion activity in the presence of cytosol and an ATP -regenerating system may be searched for. To control that the inhibition is targeted to a Rab5 effector (by inhibiting the Rab5-Rab5 effector interaction or the Rab5 effector activity), the inhibitory effect of the drug should be rescued by increasing concentrations of that particular Rab5 effector. For this a titration curve of the drug in the presence of increasing concentrations of the Rab5 effector will be necessary. The following two assays are intended to screen for drugs that inhibit endosome movement and cell motility in tumor cells or enhance endocytic membrane fusion in infected cells. The assays score the overall activity, but as soon as the effector molecules specifically involved in these processes are identified, it will provide the additional possibility of using Assay for activity 3.

(ii) Microtubule spin-down assay

75 to lOOμg purified early endosomes are incubated at room temperature for 20min with lOOμg HeLa cytosol protein in a reaction brought to a final volume of 50μl by addition of BRB-80 (80mM K-Pipes, ImM MgCl₂, ImM EGTA, pH6.8). lOμl of taxol-stabilized microtubules (lOOμg tubulin equivalent) are added, and this mix is incubated for lOmin at room temperature, in the presence or absence of the candidate drug, and then overlayed on 600μl of a 35% (w/v) sucrose cushion. After sedimentation at 100,000 x g for 20min at 22°C in a TLA 100.4 rotor, the upper layer was removed, the cushion was washed with lOOμl BRB-80 and then also removed. The remaining pellet was resuspended in 30μl 2x SDS-PAGE buffer plus 30μl dH O, and analysed by SDS-PAGE and immunoblotting.

Substances that are able to inhibit the association of early endosomes with microtubules below the standard fusion activity in the presence of cytosol and an ATP -regenerating system may be searched for. To control that the inhibition is targeted to a Rab5 effector (by inhibiting the Rab5- Rab5 effector interaction or the Rab5 effector activity), the inhibitory effect of the drug may be rescued by increasing concentrations of that particular Rab5 effector. For this a titration curve of the drug in the presence of increasing concentrations of the Rab5 effector is necessary.

(iii) In vitro endosome motility assay Early endosomes and cytosol are isolated as previously described (Horiuchi et al., 1997), except that HeLa spinner culture cells were allowed to internalise rhodamine-labeled transferrin for lOmin at 37 C in order to label early endosomes. Oregon-green labeled tubulin is polymerized in vitro, and the resulting microtubules isolated, stabilised with taxol, and perfused into a microscope slide/glass coverslip chamber as previously described (Marlowe et al., 1998). The chamber is then washed with BRB-80 plus lOμM taxol to stabilise microtubules and remove non-polymerized tubulin. A mixture of HeLa cytosol (2mg/ml), MgATP (200μM), fluorescently- labeled endosomes (3mg/ml), purified bovine haemoglobin (3mg/ml; Sigma), and an anti-fade system (Howard and Hyman, 1993), in the presence or absence of the candidate drug, is perfused into the chamber, and images of fluorescently-labeled microtubules (FITC filter set) or endosomes (Rhodamine filter set) are collected at 2sec-intervals using a time-lapse fluorescence videomicroscope. All image acquisition, processing, and analysis of movies is performed using the NTH Image vl.60 software package. Endosome movements are defined as linear, vectorial motions that occur on fluorescent microtubules over four or more consecutive images. These are discriminated from tethered Brownian motions, which generally display non-linear shaking, or flip randomly back and forth from image to image. Movements are counted from at least three and in most cases many more, individual movies, averaged and population significant differences calculated using the Excel spreadsheet program (Microsoft, Inc.). Substances that are able to inhibit the motility of early endosomes with microtubules below the standard fusion activity in the presence of cytosol and an ATP -regenerating system can be searched for. To control that the inhibition is targeted to a Rab5 effector (by inhibiting the Rab5-Rab5 effector interaction or the Rab5 effector activity), the inhibitory effect of the drug may be rescued by increasing concentrations of that particular Rab5 effector. For this a titration curve of the drug in the presence of increasing concentration of the Rab5 effector is necessary.

Assay for activity 5

This assay describes the measurement of cell motility in vitro. The assay may be standardised for high throughput screening.

To address the effect of drugs on cell motility, BBCE cells or other tumor cell lines are trypsinized and plated at a density of 1,500 cells per cm². 24h later the cells are stimulated with bFGF or other growth factors and subjected to Time-Lapse Video Microscopy (TLVM). The cells for TLVM are grown on coverslips and mounted into aluminum chambers as described in (Bradke and Dotti, 1997). Briefly, aluminum is cut into 7.5cm x 2.5cm x 2mm rectangles, the dimensions are similar to a standard glass slide and fit onto the microscope stage. In the centre a circular hole is cut of 8 mm diameter, and around it 4mm area is milled from both sides. A glass coverslip is inserted onto one side of the chamber, sealed with a lubricant, medium is then added to fill the chamber, cells grown on coverslips are then inverted onto the other side of the metal slide and sealed with lubricant. TLVM is carried out using a standard video microscopy set up. A dedicated automatic program was developed to detect and track fluorescently labeled cells as they move. Example 3:

A novel Procedure for the Purification of Rab 5 effectors.

1201 of bacterial DH5α cells were grown to express GST-Rab5 for the large scale isolation of Rab5 effectors. The cells were induced, harvested, and broken according to manufacturer instructions (Pharmacia). Lysis buffer (2500ml) consisted of a PBS solution containing lOOμM GDP, 5mM MgCl2, 5mM 2-mercaptoethanol, 5μg/ml DNAase, 5μg/ml RNase, and a cocktail of protease inhibitors. High speed supernatant of the cell lysate was incubated with 20ml glutathione sepharose beads (Pharmacia) at 4°C for 2h under rotation. Subsequently, beads were loaded on a column and washed with lysis buffer without DNAase, RNase, and protease inhibitors. This procedure resulted in lg of GST-Rab5 bound to the column. The molecule is recovered in the GDP-bound inactive form and nucleotide exchange is necessary to convert it into the GTP -bound form to allow effector binding.

The present inventors have designed an alternative procedure that overcomes the low efficiency normally affecting this step. The modification consists in the use of aluminum fluoride (A1F₃) which when bound to Ras:GDP and Mg ⁺ has been shown to mimic the transition state for GTP- hydrolysis of the GTPase complexed to GAP. The inventors have obtained evidence, however, that the method can also be applied for the complex between Rab5:GDP:Mg²⁺:AlF₃ and its effectors. In other words, it is possible to obtain a complex consisting of Rab5, GDP, Mg ⁺, A1F₃, and Rab5 effectors. This procedure has for example allowed the inventors to purify at large scale EEAl and Rabankyrin-5, suggesting that this method can be applied for the identification of novel effectors and regulators for small GTPases.

Example 4: Construction and expression of EGFP-Rab5 pEGFP-Rab5 was constructed by PCR amplification of the human Rab5a cDNA from pGEM- myc-Rab5 using the primers CCCAAGCTTATGGCTAGTCGAGGCGCAACA and AACTGCAGTTAGTTACTACAACACTGATT followed by cloning of the Hindπi-Pstl fragment from the PCR product into a pEGFP-C3 expression vector (Clontech, Inc.). A431 cells were grown to ~80%> confluency in 10cm petri dishes and transfected with 20-30μg of plasmid DNA using a calcium phosphate based protocol. Stably transfected clonal lines were isolated after incubation in selective (G418; 0.5μg/ml) growth medium for 7-10d and checked for GFP fluorescence on endosomal structures. Despite the presence of the selectable marker, variable levels of expression of EGFP-Rab5 in these cells were observed.

Time-lapse fluorescence videomicroscopy of EGFP-Rab5 in vivo Cells were grown on glass coverslips and were transferred to custom-built aluminum microscope slide chambers (EMBL workshop, Heidelberg) just prior to observation. Unless otherwise stated, cells expressing average levels of EGFP-Rab5 were selected and analyzed on a Zeiss Axioskop microscope using 100X/1.40 plan-Apochromat lens with a temperature-controlled objective sleeve attached (EMBL workshop, Heidelberg). Time-lapse imaging was performed, collecting images at 2sec-intervals using a computer-controlled shutter (Uniblitz, Inc.) with illumination by a 100W mercury arc lamp attenuated with two heat reflection filters and a KG-1 heat absorbance filter (Zeiss). GFP-fluorescence was visualised with Hi-Q FIT, or GFP filter sets (Chroma Technologies, Inc.), and images were acquired using a COHU 4913 CCTR video camera with on- chip integration controlled by the NIH-Image vl.60 software package.

Example 5 :

Microtubule spin-down assay

75 to lOOμg of purified early endosomes were incubated at room temperature for 20min with lOOμg HeLa cytosol protein in a reaction brought to a final volume of 50μl by addition of BRB- 80 (80mM K-Pipes, ImM MgCl₂, ImM EGTA, pH6.8). lOμl of taxol-stabilized microtubules (lOOμg tubulin equivalent) was added and this mix was incubated for lOmin at room temperature, and then overlayed on 600μl of a 35% (w/v) sucrose cushion. After sedimentation at 100,000 x g for 20min at 22°C in a TLA 100.4 rotor, the upper layer was removed, the cushion was washed with lOOμl BRB-80 and then also removed. The remaining pellet was resuspended in 30μl 2x SDS-PAGE buffer plus 30μl dH₂0, and analysed by SDS-PAGE and immunoblotting.

Example 6:

In vitro endosome motility assays

Early endosomes, and cytosol were isolated as previously described, except HeLa spinner culture cells were allowed to internalise rhodamine-labeled transferrin for lOmin at 37°C in order to label early endosomes. Oregon-green labeled tubulin was polymerized in vitro, and the resulting microtubules isolated, stabilised with taxol, and perfused into a microscope slide/glass coverslip chamber as previously described. Alternatively, purified centrosomes were allowed to stick to the glass surfaces of the chamber. Microtubule asters were then polymerized in situ by 30min incubation of the chamber at 37°C with fluorescently-labeled tubulin (4mg/ml) and ImM GTP in BRB-80. The chamber was then washed with BRB-80 plus lOμM taxol to stabilise microtubules and remove non-polymerized tubulin. A mixture of HeLa cytosol (2mg/ml), MgATP (200μM), fluorescently-labeled endosomes (3mg/ml), purified bovine haemoglobin (3mg/ml; Sigma), and an anti-fade system was perfused into the chamber, and images of fluorescently-labeled microtubules (FITC filter set) or endosomes (Rhodamine filter set) were collected at 2sec- intervals using the time-lapse fluorescence videomicroscope setup described above.

Analysis and quantification of videos

All image acquisition, processing, and analysis of movies was performed using the NTH Image vl.60 software package. Endosome movements were defined as linear, vectorial motions that occurred on fluorescent microtubules over four or more consecutive images. These were discriminated from tethered Brownian motions, which generally displayed non-linear shaking, or flipped randomly back and forth from image to image. Movements were counted from at least three and in most cases many more, individual movies, averaged and population significant differences calculated using the Excel spreadsheet program (Microsoft, Inc.).

References

Amano, M., Mukai, H., Ono, Y., Chihara, K., Matsui, T., Hamajima, Y., Okawa, K., Iwamatsu, A., and Kaibuchi, K. (1996). Identification of a putative target for Rho as the serine-threonine kinase protein kinase N. Science 271, 648-50.

Bradke, F., and Dotti, C. G. (1997). Neuronal polarity: vectorial cytoplasmic flow precedes axon formation. Neuron 19, 1175-86.

Bucci, C, Parton, R. G., Mather, I. H., Stunnenberg, H., Simons, K., Hoflack, B., and Zerial, M. (1992). The small GTPase rab5 functions as a regulatory factor in the early endocytic pathway. Cell 70, 715-728.

Caraglia, M., Di Gennaro, E., Barbarulo, D., Marra, M., Tagliaferri, P., Abbruzzese, A., and Budillon, A. (1999). Up-regulated EGF receptors undergo to rapid internalization and ubiquitin- dependent degradation in human cancer cells exposed to 8-C1- cAMP. FEBS Lett 447, 203-8. Chistoforidis, S., McBride, H. M., Burgoyne, R., D., and Zerial, M. (1999a). The Rab5 effector EEAl is a core component of endosome docking. Nature 397, 621-625.

Christoforidis, S., Miaczynska, M., Ashman, K., Wilm, M., Zhao, L., Yip, S. C, Waterfield, M. D., Backer, J. M., and Zerial, M. (1999b). Phosphatidylinositol-3-OH kinases are Rab5 effectors. Nat Cell Biol 1, 249-52.

Christoforidis, S., and Zerial, M. (1999c). An affinity chromatography approach leading to the purification and identification of novel Rab effectors. .

Field, J., Vojtek, A., Ballester, R., Bolger, G., Colicelli, J., Ferguson, K., Gerst, J., Kataoka, T., Michaeli, T., Powers, S., and et al. (1990). Cloning and characterization of CAP, the S. cerevisiae gene encoding the 70 kd adenylyl cyclase-associated protein. Cell 61, 319-27.

Geller, A. I., During, M. J., Haycock, J. W., Freese, A., and Neve, R. (1993). Long-term increases in neurotransmitter release from neuronal cells expressing a constitutively active adenylate cyclase from a herpes simplex virus type 1 vector. Proc. Natl Acad. Sci. USA 90, 7603-7607.

Gorvel, J.-P., Chavrier, P., Zerial, M., and Gruenberg, J. (1991). Rab5 controls early endosome fusion in vitro. Cell 64, 915-925.

Gournier, H., Stenmark, H., Rybin, V., Lippe, R., and Zerial, M. (1998). Two distinct effectors of the small GTPase Rab5 cooperate in endocytic membrane fusion. Embo J 17, 1930-40.

Horiuchi, H., Lippe, R., McBride, H. M., Rubino, M., Woodman, P., Stenmark, H., Rybin, V., Wilm, M., Ashman, K., Mann, M., and Zerial, M. (1997). A novel Rab5 GDP/GTP exchange factor complexed to Rabaptin-5 links nucleotide exchange to effector recruitment and function. Cell 90, 1149-1159.

Howard, J., and Hyman, A. A. (1993). Preparation of marked microtubules for the assay of the polarity of micro tubule-based motors by fluorescence microscopy. Methods Cell Biol 39, 105-13. Ito, K, Ishi, N., Miyashita, A., Tominaga, K., Kuriyama, H., Maruyama, H., Shirai, M., Naito, M., Arakawa, M., and Kuwano, R. (1999). Molecular cloning of a novel 130-kDa cytoplasmic protein, Ankhzn, containing Ankyrin repeats hooked to a zinc finger motif. Biochem Biophys Res Commun 257, 206-13.

Marlowe, K. J., Farshori, P., Torgerson, R. R., Anderson, K. L., Miller, L. J., and McNiven, M. A. (1998). Changes in kinesin distribution and phosphorylation occur during regulated secretion in pancreatic acinar cells. Eur J Cell Biol 75, 140-52.

Matoskova, B., Wong, W. T., Seki, N., Nagase, T., Nomura, N., Robbins, K. C, and Di Fiore, P. P. (1996). RN-tre identifies a family of tre-related proteins displaying a novel potential protein binding domain. Oncogene 12, 2563-71.

McBride, H. M., Rybin, V., Murphy, C, Giner, A., Teasdale, R., and Zerial, M. (1999). Oligomeric complexes link Rab5 effectors with NSF and drive membrane fusion via interactions between EEAl and syntaxin 13. Cell 98, 377-86.

McMahon, G. (2000). VEGF receptor signaling in tumor angiogenesis. Oncologist 5, 3-10.

Murphy, C, Grummt, M., Saffrich, R., Gournier, H., Rybin, V., Rubino, M., Auvinen, P., Parton, R. G., and Zerial, M. (1996). Endosome dynamics regulated by a novel rho protein. Nature 384, 427-432.

Nielsen, E., Severin, F., Backer, J. M., Hyman, A. A., and Zerial, M. (1999). Rab5 regulates motility of early endosomes on microtubules. Nat Cell Biol 1, 376-382.

Sasaki, T., Kikuchi, A., Araki, S., Hata, Y., Isomura, M., Kuroda, S., and Takai, Y. (1990). Purification and characterization from bovine brain cytosol of a protein that inhibits the dissociation of GDP from and the subsequent binding of GTP to smg p25A, a ras-like GTP- binding protein. J. Biol. Chem. 265, 2333-2337.

Siddhanta, U., Mcllroy, J., Shah, A., Zhang, Y., and Backer, J. M. (1998). Distinct roles for the pi lOalpha and hVPS34 phosphatidylinositol 3'- kinases in vesicular trafficking, regulation of the actin cytoskeleton, and mitogenesis. J Cell Biol 143, 1647-59. Simonsen, A., Lippe, R., Christoforidis, S., Gaullier, J.-M., Brech, A., Callaghan, J., Toh, B.-H., Murphy, C, Zerial, M., and Stenmark, H. (1998). EEAl links phosphatidylinositol 3-kinase function to Rab5 regulation of endosome fusion. Nature 394, 494-498.

Sorkin, A., Krolenko, S., Kudrjavtceva, N., Lazebnik, J., Teslenko, L., Soderquist, A. M., and Nikolsky, N. (1991). Recycling of epidermal growth factor-receptor complexes in A431 cells: identification of dual pathways. J Cell Biol 772, 55-63.

Stenmark, H., Vitale, G., Ullrich , O., and Zerial, M. (1995). Rabaptin-5 is a direct effector of the small GTPase Rab5 in endocytic membrane fusion. Cell 83, 423-432.

Claims

C L A I M S

1. Use of an effector of a GTPase as a target in an in vitro or in vivo assay to detect substances useful as pharmaceutical agents for the prophylaxis and/or treatment of cancer and other proliferative, invasive or cell migration disorders such as endometriosis, atherosclerosis, inflammatory and allergic diseases, infectious diseases, diabetes, Alzheimer's disease and skin repair diseases such as psoriasis.

2. Use of claim 1, wherein the GTPase is of the Rab family.

3. Use of claim 2, wherein the GTPase is any of Rab4, Rab5, Rab7, Rabl 1, Rab 17, Rab 18, and Rab22.

4. Use of any of the preceding claims, wherein the infectious disease is AIDS, tuberculosis, pseudotuberculosis, cholera, malaria, gastroenteritis, enteric fever, typhus, those diseases caused by pathogens (bacteria or organisms) such as Mycobacterium, Staphylococcus, Toxoplasma, Trypanosoma, Listeria, Salmonella, Legionella, Leishmania, Coxiella, Shigella, Yersinia, Neisseria, Vibrio, Bartonella, or any other infectious disease caused by an infectious agent that infects cells by the endocytic route and resides intracellularly in phagosomes escaping the cellular killing mechanisms.

5. Use of any of claims 1 to 3, wherein the cancer is a benign tumor, a malignant tumor, a carcinoma, a sarcoma, a melanoma, a leukemia, a glioma, or a neuroblastoma, in particular a lung carcinoma, an osteosarcoma, a lymphoma, a soft tissue sarcoma, a breast carcinoma, a bile cancer, a cervix carcinoma, a cancer of the (small) intestine, of the kidneys, of the cavity of the mouth, a penis carcinoma, an ovary cancer, a stomach cancer, a cancer of the tongue, a brain cancer, a bladder carcinoma, a prostate carcinoma, a liver carcinoma, a carcinoma of the pancreas, and every tumor that invades other tissues and organs distinct from its site of origin.

Use of any of the preceding claims, wherein the assay is carried out in the presence of one or more GTPase effector/regulator molecule(s) which is/are either native and biochemically purified from a vertebrate, or recombinant and biochemically purified from bacterial cultures, from yeast cultures, or from other cultured eukaryotic cells, in either case labeled by a covalent modification or radioactivity suitable for use in the assay.

7. Use of claim 6, wherein the assay is carried out in the simultaneous presence of at least one type of GTPase and/or endosomal membrane fractions fluorescently labeled or labeled by any other modification that allows its detection, and/or cytosolic extracts, and/or an ATP -regenerating system and/or a number of chemicals to be tested for their suitability as an anti-cancer or anti-infectious diseases drug.

8. Use of any of the preceding claims, wherein the substance useful as pharmaceutical agent is a molecule/substance that carries one or more of the following functional groups: a halide atom bound to an alkyl, alkenyl, alkinyl or aryl residue, an alcohol group (primary, secondary, tertiary), an ether group, a carbonyl function (aldehyde or ketone), a carboxylic acid group, a carboxylic anhydride group, a carbamoyl group, a haloformyl group, a cyano group, an ester group including a lactone group, a benzyl, phenyl, tolyl, tosyl, sulfonyl group, an amino group (primary, secondary, tertiary), an isocyanate, a cyanate, a thioisocyanate, a thiocyanate, a carbamate, an azide, a diazo group, and a quinone group; or is an organometallic compound, a sterol moiety(ies)-containing molecule, a β-hydroxy carboxylic acid, an inorganic acid or complex such as a metallocene, a nucleic acid, a cytokine, a hormone, an antibody, or an oligopeptide comprising up to 20, preferably 8,

10, or 12, amino acid residues.

9. Use of claim 8, wherein the antibody is a polyclonal or monoclonal antibody, or a fragment thereof, humanised or human, inhibitory or stimulatory, raised against and targeted towards any of the aforementioned GTPase effectors.

10. Use of claim 8, wherein the nucleic acid is genomic DNA, cDNA, or mRNA, or a fragment there, an oligonucleotide, an oligoribonucleotide, all being based on or derived from any of the GTPase effector having any of the sequences as depicted in SEQ ID NO: 1, 3, 5, 7, 9, 11, 13, and 15, or gene therapy vectors derived from the aforementioned

GTPase effector gene sequences.

11. A kit useful for carrying out the assay of any of the preceding claims, the kit comprising in a suitable means at least one GTPase effector/regulator molecule which is either native and biochemically purified from a vertebrate, or recombinant and biochemically purified from bacterial cultures, from yeast cultures, or from other cultured eukaryotic cells, in either case labeled by a covalent modification or by radioactivity suitable for the use in the assay.

12. The kit of claim 11 , wherein the kit further comprises at least one type of GTPase and/or endosomal membrane fractions fluorescently labeled or labeled by any other modification that allows its detection, and or cytosolic extracts, and/or an ATP-regenerating system and/or a number of chemicals to be tested for their suitability as a drug effective against any of the diseases depicted in claims 1, 4, and 5.

13. The kit of claim 12, wherein the at least one GTPase is/are one or more of the GTPases Rab4, Rab5, Rab7, Rabl l, Rabl7, Rabl8, and Rab22.