WO2007093449A2 - Method and means for high- throughput -screening of compounds that exhibit anti -arenavirus activity - Google Patents

Abstract

The invention relates to a recombinant cell-based system in which the expression of the Lassa Virus receptor in one cell and the expression of the GP-C protein from Lassa Virus in another cell is used to have cells fused and thereby to mimic cellular entry of Lassa Virus. The system of the invention further uses reporter genes in one cell population and a transactivator protein in the other cell population to quantify the occurring cell fusion. The system is applied for screening anti-fusion-acting compounds that are thus identified as having anti -Lassa Virus activity. Such compounds are applicable in the treatment of Lassa Virus- induced disease.

Description

TITLE

Method and means for high-throughput-screening of compounds that exhibit anti -arenavirus activity

FIELD OF INVENTION

The invention relates to the field of medicine. In particular the invention relates to the field of screening anti-viral compounds. More in particular it relates to compounds that act against arenaviruses, and methods for identifying same.

BACKGROUND OF THE INVENTION

Arenaviruses are enveloped animal viruses that can cause severe illnesses in infected humans. Following their geographical distribution they are subdivided into the new- world and the old-world group of the Arenaviridae family. Some arenaviruses can induce hemorrhagic fevers in humans,- those include Junin Virus, Machupo Virus, Guanarito Virus and Sabia Virus within the new-world subgroup and the Lassa Virus in the old-world group. The hemorrhagic fever arenaviruses are listed in the CDC category A as possible bioterrorism agents and additionally they are grouped within the highest biological safety level BSL-4. The natural arenavirus reservoirs are chronically infected rodents, where infection is generally not associated with illness. Virus transmission to humans can occur by exposure to infected rodents or infected humans. Virus spread happens either directly through virus- containing body fluids or indirectly via virus-loaded aerosols . The publicly most perceived arenavirus is Lassa Virus, since this pathogen has recently been exported by travelers from its natural habitat West Africa to Europe, e.g. to Germany, Switzerland, the Netherlands, and Great Britain. Albeit sporadic, these cases have achieved great public concern within the pertained populations and focused special interest on this emerging pathogen {Crowford N. S. 2002, Euro Surveill. 7(3) :50-2) .

For Lassa Virus, which is endemic in West Africa, the CDC (Centers for Disease Control and Prevention) estimates 100,000 to 300,000 human infections with approximately 5,000 deaths per year. Lassa Virus infections proceed mildly or asymptomatically in about 80% of cases, whereas the remaining 20% develop severe disease with symptoms ranging from fever, retrosternal pain, sore throat, back pain, cough, abdominal pain, vomiting, diarrhea, conjunctivitis, facial swelling, proteinuria to mucosal bleedings resulting in death in 15%-20% of hospitalized patients.

No vaccine against Lassa Virus is available to date. Lassa Virus infections can successfully be treated with Ribavirin, a synthetic nucleoside analogue. However, Ribavirin therapy is effective only if started early after onset of the first symptoms. Additionally, supportive care consisting of maintenance of appropriate fluid and electrolyte balance, oxygenation and blood pressure, as well as treatment of any other complicating infections should be given {CDC Lassa Virus FACTS sheet) .

Attempts to provide pharmaceuticals against Lassa Virus have so far led to the development of antibodies that showed anti-viral activity in diverse animal models (Bredenbeek et al. 2006, Virol. 345 (2).-299-304 ; Geisbert et al . 2005, PLoS Med. 2 (6) :el83; Fisher-Hoch 2004, Expert Rev. Vaccines 3 (2).-189-197) . However, a prophylactic or therapeutic medicament for human use is still not available.

The development of drugs against Lassa Virus usually requires special equipment since BSL-4 conditions have to be used to work with this highly pathogenic virus. To overcome the problems related to experimentation under such strict safety conditions, one could envision the use of recombinant DNA technology. Use of these techniques allows solitary expression of viral proteins outside the context of viral infections. The developing knowledge of the discrete steps in the viral lifecycle may provide specific strategies to investigate how to inhibit or completely block the virus somewhere in its lifecycle.

The Lassa Virus lifecycle comprises at least the following general steps: receptor-binding on the host cell surface, virus- internalization through endocytosis, replication, transcription, translation of viral proteins, and assembly of the viral genome and viral proteins within the cell, and finally virus-budding from the cell surface. Principally all of these steps are potential targets for anti-viral drugs. Virus entry is of major importance in respect to the present invention (see below) and is therefore described in some more detail.

The entry of Lassa Virus is initiated by binding to at least one specific receptor on the host cell membrane. The only Lassa Virus receptor known to date is alpha- dystroglycan (Cao et al . 1998, Science 282 (5396) :2079-81) , which interacts with the Glycoprotein- 1 polypeptide (GP-I) on the surface of the Lassa Virus particles. After the initial binding, the virus is internalized within specific endosomal vesicles by endocytosis (Borrow and Oldstone 1994, Virol. 198 (1) :l-9) . The acidic pH that prevails in the endosomal compartments is believed to induce refolding of the virus-membrane-anchored Glycoprotein-2 (GP-2). Specific GP-2 domains, so-called "fusion peptide segments", insert into to cytoplasma membrane and initialize the fusion between virus envelope and endosomal membrane. The protein known as GP-C is the precursor of the two proteins derived therefrom: GP-I and GP-2. As a result of the fusion event the Lassa Virus genome and the regulatory proteins associated with it are released into the cytoplasm of the host cell and the viral lifecycle continues. Some fusion peptide segments responsible for the fusion were identified within GP-2 {Gallaher et al . 2001, BMC Microbiol. 1:1; Glushakova et al . 1994, Biochim. Biophys. Acta 1110 (2) :202-208) .

In 1994, Nussbaum et al . presented an assay that allowed analysis of the fusogenic mechanisms used by enveloped viruses {Nussbaum et al . 1994, J. Virol. 68 (9).-5411-5422) . This assay is based on the use of heterologous DNA comprising a promoter element and a downstream- located reporter gene. Expression of the product encoded by the reporter gene is achieved by the binding of a transactivator protein to the promoter. The sleight of this assay lies in the use of two different cell populations: one cell harboring a nucleic acid encoding the transactivator protein, and the second cell harboring a nucleic acid comprising the promoter/reporter-gene construct. When transfecting a nucleic acid encoding the glycoprotein of an enveloped virus in one of the two cells and transfecting a nucleic acid encoding the respective virus receptor in the other cell, this method of reporter-gene-transactivation can be used to quantify glycoprotein- induced cell-cell fusion. Clearly, cell fusion only occurs after appearance of the expressed receptor on the cell surface of one cell and the appearance of the glycoprotein on the cell surface of the other cell. After cell fusion has taken place, the transactivator protein is able to interact with the promoter from the fusion partner and subsequently induce expression of the reporter gene product, which expression can be measured. Additionally, prevention of cell-cell fusion by any substance would indicate a potential anti -viral drug, as suggested by Nussbaum et al .

The system from Nussbaum et al . dealt with cell-cell fusions due to the interaction of CD4 with the Human Immunodeficiency Virus (HIV-I) -derived env protein. HIV is an enveloped virus. Similar other systems based hereon have been described in the art. One transactivation assay applying an arenaviral surface glycoprotein has been described for the new-world arenavirus Junin Virus (York et al. 2004, J. Virol 78 (19) : 10783-92) . Junin Virus is the causative agent of Argentine Hemorrhagic Fever and, consistent with Lassa Virus, has two glycoprotein subunits on the virion surface. Junin Virus uptake is believed to happen in concordance with Lassa Virus uptake. The above described cell-cell fusion assay was used to quantify Junin Virus entry. The reporter gene encoding β-galactosidase (cloned into the recombinant vaccinia virus VCB21R-Z) was transactivated by the bacteriophage T7 polymerase (provided by infection with recombinant vaccinia virus vTF7-3) . So in this particular cell-cell fusion assay the Junin Virus glycoprotein-expressing cell line was infected with VTF7-3 and the target cell - expressing the Junin Virus receptor - was infected with VCB21R-Z. Both cell lines required infection with a recombinant vaccinia virus prior to cell- cell fusion, which might significantly influence the adequacy of the assay. Avoiding infection and using more sophisticated transactivator and promoter-reporter-gene elements would be more appropriate. This inconvenience was also predominantly present in the original assay as described by Nussbaum et al . Infact, certain other systems based on the transactivation of HIV-I long terminal repeat (LTR) promoter element by the HIV-I tat protein have been described that overcome this inadequacy (Schwartz et al . 1990, J. Virol. 64 (6).-2519-29; Kimpton and Emerman 1992, J. Virol. 66(4).-2232-39) .

Junin Virus and Lassa Virus do not use the same cellular receptor to enter cells (Spiropoulou et al . 2002, J Virol 76 (10).-5140-46) . Hence, the system described by York et al . is not fully applicable for studying Lassa Virus entry. Moreover, possible anti-fusion-, or anti-viral entry compounds that may be found in screens using the Junin-based system may not be functional whatsoever against Lassa Virus entry, as it relates to completely different interactions. Besides this, the actual mechanisms behind the Lassa Virus entry into cells are rather vague and relatively unknown. Most data gathered on arenavirus entry originate from studies with the Lassa Virus related, old-world arenavirus, lymphocytic-choriomeningitis-virus (LCMV) . LCMV is not a human pathogen and as such the prototypic old-world arenavirus for studies without BSL-4 constraints. As outlined above, some systems in the art exist that circumvent the need for high safety level experimentation. However, no systems exist that provide the means to provide anti-virals to Lassa Virus. Notably, the systems described above have not resulted whatsoever in compounds that also act against Lassa Virus infections. There is a clear need in the art for methods and means that enable the skilled person to screen for compounds that actually do act against Lassa Virus infections, and that may be used for prophylactic or therapeutic treatment of Lassa Virus induced disease.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows the principle of the recombinant cell-cell fusion assay of the present invention. Two different cell populations were transiently co-transfected with either Lassa Virus GP-C together with HIV-1-LTR-β-gal (referred to as population 1) or with HIV-1-tat (referred to as population 2, known to express Lassa Virus receptor alpha- dystroglycan) . 1 day post-transfection 1:1 cell mixtures were pH-shifted to induce membrane fusion and after an additional incubation for 1 day β-gal activity was measured.

Figure 2 shows the result of a recombinant cell -cell fusion assay for three different viral glycoproteins (Lassa Virus GP-C, LCMV GP-C and Influenza A-derived hemagglutinin: HA) and two negative controls (Lassa Virus nucleoprotein or no additional protein, respectively) . All five experiments were subdued to different pH-shifts to determine the optimal pH required for cell -cell fusion. Influenza HA served as positive control as HA is known to fuse at pH 5.3 in the late endosome (see WO 94/02125) , whereas Lassa Virus GP-C- induced and LCMV GP-C- induced cell-cell fusion required more acidic pH conditions (_pH^optimal 4.0).

Figure 3 shows the schematic representation of the adapter plasmid pAdApt used for the eukaryotic expression of the codon-optimised Lassa Virus GP-C. Important features of the vector are predominately the long (and therefore very strong) CMV promoter, and the downstream located Lassa Virus GP-C encoding sequence (strain Josiah) . Most significantly is the fact that the GP-C encoding gene was codon-optimised to ensure high expression in human cells.

Figure 4 shows the codon-optimised nucleic acid sequence of GP-C.

SUMMARY OF THE INVENTION

The present invention relates to an in vitro cell -based system for mimicking the entry of a Lassa Virus into a host cell by cell fusion, said system comprising two cell populations, wherein the first cell population expresses the glycoprotein (GP-C) or a functional homologue thereof from Lassa Virus, and the second population expresses the natural receptor for Lassa Virus, said system further comprising: a first nucleic acid comprising a nucleic acid sequence encoding a reporter gene product, and a responsive promoter element; and a second nucleic acid encoding a protein able to act on said responsive promoter element, wherein said first and said second nucleic acid are separately present in said first or said second cell population before cell fusion between the two cell populations has occurred. In a preferred embodiment, the nucleic acid sequence encoding the Lassa Virus GP-C protein is codon-optimised.

In another aspect, the invention relates to a method for fusing cells comprising culturing the cell -based system according to the invention in an acidic medium, and allowing the cell populations in said system to fuse. Furthermore, the invention relates to a method for identifying a compound that inhibits Lassa Virus cell entry. The invention also relates to a use of a system according to the invention for identifying a compound that inhibits fusion between the two cell populations. Importantly, the invention also relates to the compounds identified by a method according to the invention or identified by a use according to the invention.

DETAILED DESCRIPTION

The present invention relates to a system, which is applicable in methods for identifying compounds that act as anti-arenaviral drugs. Since work with wild-type Lassa Virus requires BSL-4 facilities it is useful to have an assay system to allow high throughput screening of possible antiviral compounds under BSL-2 conditions. The present invention now enables one to study the interaction between the Lassa Virus and its host cell without the need of such strict safety precautions. Although similar systems for HIV and Junin Virus were available in the art, the inventors of the present invention have now surprisingly found that such a recombinant system is also applicable for Lassa Virus, as some unexpected hurdles were overcome during the course of the work. The problems that were not envisioned and the manner in how these problems were solved will be explained in more detail below. A recombinant assay was set up that is based on the expression of only one Lassa Virus protein: the glycoprotein generally referred to as GP-C. As discussed above, this allows working under less strict, BSL-2 conditions.

The invention is directed to a system in which the process by which arenaviruses enter cells is mimicked in a recombinant cell-cell fusion assay: Two different cell- populations are fused under conditions that resemble those needed for arenaviral entry into cells. One cell acts as the virus, the other acts as the host cell.

One of the two cell populations expresses the arenaviral glycoprotein (Lassa Virus-derived GP-C) , thereby playing the role of the virus. The other cell population expresses the Lassa Virus-specific arenavirus receptor alpha-dystroglycan, thus playing the role of the host cell that becomes infected. Additionally, one of the two cell populations contains a reporter gene element (HIV-1-LTR-β- gal) that is solely expressed upon binding of the transactivator protein (HIV-1-tat) to the tat-specific promoter elements upstream of the β -gal -encoding sequence. The transactivator is expressed in the other cell population . Clearly the cell in which the transactivator or, on the other hand, the reporter gene construct is present is a matter of choice. Upon fusion of the cells, after attachment of the viral glycoprotein expressed on the cell surface of the first cell population, to the receptor expressed on the cell surface of the other cell population, the transactivator expressed in one cell is able to come in contact with its specific responsive promoter elements and subsequently is able to induce β-gal expression, which product can be readily measured through means known to the person skilled in the art.

As discussed above, it is important that the process of cell entry is mimicked as carefully as possible. Clearly, the environmental circumstances in which the endosomes are formed and maintained are essential for the process to occur. It was already mentioned that the pH of the endosome has an influence on the processing of the glycoprotein. In the present invention, the endosomal environment during virus entry is mimicked by the application of cell culture media or cell culture buffers adjusted such that they resemble endosomal pH conditions. So, the invention also relates to the specific pH conditions used in the system, which enables one to proceed through the entire process of initial binding to final expression of the reporter gene product. The method outlined herein enables one not to only grow and maintain cells in culture, it enables one also to have them fused and mimic the viral entry of arenviruses into target cells. Preferably, the method is used to mimic the entry of Lassa Virus. However, the specific conditions disclosed herein may also be applicable for other arenaviruses, which is further a matter of routine experimentation, based on what has now been disclosed for the first time herein. The system of the present invention enables one to screen for compounds that inhibit viral entry induced by the binding of the particular glycoprotein to its receptor . One of ordinary skill in the art would readily understand that the present invention may be employed with different cell types, with different expression plasmids/constructs and with different reporter genes. Moreover, it is clear that different kinds of compounds may be screened, ranging from small synthetic compounds present in large libraries, to large compounds such as antibodies, or parts thereof .

Clearly, the compounds can be screened on a small scale, but also in a high-throughput setting in which millions of different compounds are screened. The read-out is the measurement of the activity of the protein encoded by the reporter gene. However, one could also envision a reverse system in which the reporter gene is constantly on, but only after fusion is down-regulated by a repressor expressed in the other cell. A negative read-out would then be expression of the reporter gene (since no fusion would have occurred) .

In a preferred embodiment, the reporter gene is up- regulated upon fusion, but off when fusion is inhibited by whatever compound, which compound has a negative impact on the fusion of the cells. Clearly the compound may act only on the glycoprotein, or only on the receptor, or both. The compound may inhibit the binding of the glycoprotein to the receptor. The compound may also prevent subsequent fusion, despite of a positive glycoprotein-receptor interaction. Hence, the compound may act on all levels within the process of cell fusion, thereby influencing the viral entry of the Lassa Virus into the cell in a negative fashion.

The compound may be a small compound from a small compound library. It may be proteinaceous, such as a peptide, a protein or a polypeptide. It may also be a chemical compound. It may be synthetic or derived from a natural source, such as plants, soil or animals, including humans. It may also be an antibody, such as a monoclonal antibody, which may be animal-derived, manipulated to become humanized or fully human. It may be of any immunoglobulin class, such as IgA, IgG or IgM. The compound may also be internalized into the cell, and for instance act through inhibiting the expression, trafficking or posttranslational modification of the glycoprotein (or the receptor) . Hence, the compound may act extra-cellularly or intra-cellularly, or both. In another embodiment, the compound may be the soluble form of the Lassa Virus specific receptor: alpha- dystroglycan, such that the soluble form inhibits further interaction with the cellularly expressed receptor (see US 6, 083 , 911) . It may also be an interacting part of the receptor in soluble form or bound to a carrier. This interacting part of the receptor may be made synthetically or split and purified from the wild type version.

The examples (see below) describe the use of certain cell lines in the cell fusion system according to the present invention. It is understood that other cell lines may also be employed. Other constructs and reporter genes than those disclosed in the examples can also be used.

Examples of suitable reporter genes that can be used are β-galactosidase, β -glucuronidase, Renilla or firefly luciferase, alkaline phosphatase, β-lactamase, horseradish peroxidase, chloramphenicol acetyltransferase , and fluorescent proteins (GFP, YFP, RFP, BFP, etc.). Preferably, β-galactosidase is used.

Examples of other cell lines that may be used in the system according to the invention are rodent cells such as Chinese Hamster Ovary (CHO) cells, Baby Hamster Kindney (BHK) cells, murine embryo cells such as NIH3-3T3, or other mammalian cells like simian cells such as Vero cells and CV- 1 cells (both from the Simian Green Monkey) . In a preferred embodiment human cells are used. Examples of suitable human cells are HeLa cells, Jurkat cells (T cell leukemia cells), U-937 cells (lymphoma cells) , Human Embryonic Kidney (HEK- 293) cells, Human Embryonic Retinoblast Cells (such as PER.Cβ¹⁸ cells), Human Embryonic Lung Cells, or any other mammalian cell either primary or continuous, as long as the cell supports the expression of the transactivator protein, the receptor, the glycoprotein from Lassa Virus or supports sustaining the presence of the reporter construct. All of these features can be measured by well-known systems and methodology used by the person skilled in the art. Expression from the reporter gene (for whatever reporter used) can be measured by methods that are described in detail in the art. Moreover, selection procedures to ensure the presence of the different nucleic acids encoding the separate components are also widely applied in the art.

In a preferred embodiment of the method, cells expressing the natural host cell receptor are used as receptor presenting cells (Broder et al . 1993, Virol.

193 (1).-483-91) . Such cells do not need to be transfected with anything that encodes the receptor because of its natural presence. In a preferred embodiment, if cells need to be transfected, cells are used that can be transfected and that sustain high expression of the encoded proteins. Cells that already express the receptor without being transfected are nevertheless preferred, which implies that human cell lines are favored for embodiments of the method where human pathogenic arenaviral glycoproteins are examined. In particular HEK-293 cells are preferably used. However, other mammalian cell lines can be used as receptor presenting cell line as well, provided they express the respective viral receptor, which can be selected for by the person skilled in the art using different methods to check for expression (northern blots, western blots, real time PCR, etc) .

When choosing a cell -line for the recombinant cell -cell fusion assay as provided by the present invention, one should test for an intrinsic promoter-reporter-gene activation (without transactivation from the transactivator of choice being present) . Unspecific reporter-gene expression would rule out that particular specific combination of reporter-gene and cell line. Background expression of the reporter should preferably be as low as possible to ensure a proper read-out. Most preferred is a system in which there is a complete shut -off of the reporter gene expression when the transactivator protein is absent .

Examples of transactivators that may be used in the system according to the invention are transactivator proteins of retroviruses and their respective responsive promoter elements, or other DNA-binding-proteins with activating or repressing activity on genes located downstream of the DNA-bmding-protem binding site. Highly preferred is the tat transactivator protein from HIV. Transactivator proteins may also be fusion proteins of a DNA-bindmg domain and an activator domain (from the same or different protein or even species) . An example of such a system that is known to the person skilled in the art, is the Gal4-VP16 fusion in which the Gal4 part binds DNA and the VPlβ provides strong expression of a downstream located transgene . Notably, some system known in the art that have been used to study cell-cell fusion and thereby mimicking the viral entry into a cell, apply vaccinia virus infections to deliver nucleic acids to the cells used. In a preferred embodiment of the present invention, infection is not applied. Actually, in a preferred embodiment, the nucleic acid (when delivery is required) is introduced into the cell through transfection of the DNA. More preferably, plasmids are transfected.

Interestingly, cell-cell fusion induced by Lassa Virus glycoprotein in the method of the present invention occurred only when transfectmg a manipulated version of the Lassa Virus GP-C encoding gene. The wild-type version of the glycoprotein provided no detectable differences of reporter gene activity when compared to unspecific reporter gene activity (empty vector, used as a negative control) . It may be that the background noise was too high in the not- optimised Lassa Virus GP-C. This problem, that was encountered unexpectedly, was solved by providing a version of the gene that was manipulated such that the expression of the protein was significantly enhanced. This resulted in a positive effect in the amount of protein rising that was readily detectable over the background values. The manipulation of the gene is generally referred to as ^λcodon- optimization' , which is a technique known in the art to provide a version of a nucleic acid that yields higher levels of the gene product depending on the host in which the product is to be expressed. It is known in the art that RNA products from such genes may be more stable in one cell in comparison to other cells. Also translational effects may play a role. As human cells are preferably used in the methods according to the invention, codon-optimization for human host cells is preferably applied. Clearly, if one would apply hamster cells (as an example) one would preferably use a codon-optimised version for optimal expression in hamster cells.

Codon-optimization of Lassa Virus Glycoprotein GP-C (strain Josiah) for optimal expression in human cells was performed based on findings from Haas and co-workers (Haas et al . 1996, Current Biol. 6:315-24) , who found that expression of human genes coincided with codon-sequences (triplets) that are preferentially used to encode specific amino acids. For instance, the two codons that both encode tyrosine - TAC and TAT - prevail with frequencies of 74% (TAC) and 26% (TAT) in highly expressed human genes. Therefore codon-optimised sequences are synthesized in a way where rare codons [e.g. TAT) are replaced by more abundant ones [e.g. TAC) . Clearly, the same considerations would have to be made for codon-optimised gene expression in cell lines other than human ones, i.e. canine codon-optimization for expression in, e.g. MDCK cells, murine codon-optimization for expression in, e.g. NIH3-3T3 cells, etc. [Grantham et al. 1980, Nucl. Acid Res. 8:1893-1912).

The present invention disclosed herein, relates to an in vitro cell-based system for mimicking the entry of a Lassa Virus into a host cell by cell fusion, said system comprising two cell populations, wherein the first cell population expresses the glycoprotein (GP-C) or a functional homologue thereof from Lassa Virus, and the second population expresses a receptor for Lassa Virus, said system further comprising: a first nucleic acid comprising a nucleic acid sequence encoding a reporter gene product, and a responsive promoter element; and a second nucleic acid encoding a protein able to act on said responsive promoter element, wherein said first and said second nucleic acid are separately present in said first or said second cell population before cell fusion between the two cell populations has occurred. A ^afunctional homologue' of GP-C is defined as a partly or entire GP-C protein capable of binding to its specific receptor and inducing cell-cell fusion. The functionality of the protein is based on its capability of binding to the receptor and enabling a subsequent fusion of the cells. Hence, the GP-C protein may be mutated, shortened, altered in any way by technologies known to the person skilled in the art, as long as the resulting functional homologue is able to interact with the GP-C receptor and allow subsequent fusion. Clearly, a homologue is not functional if the homologue cannot bind and/or cannot induce fusion. A receptor for Lassa Virus is a receptor that interacts with the Lassa Virus GP proteins expressed on the viral surface, and that subsequently allows or stimulates viral entry. One of the known Lassa Virus receptors is alpha-dystroglycan . This receptor is preferably used. However, if new receptors are identified that also enable attachments and subsequent viral entry of the Lassa Virus particle, such receptors are also seen as natural receptors and are within the present invention. Preferably, the cell population not expressing the GP-C protein from Lassa Virus (hence, the other cell population) expresses the receptor by itself, without the need of introducing it into the cell.

Before cell fusion occurs in the system of the present invention, the first nucleic acid comprising a nucleic acid sequence encoding a reporter gene product, and further comprising a responsive promoter element is present in one cell population and is separately present when cell fusion has not yet occurred, from said second nucleic acid encoding a protein able to act on said responsive promoter element . It is to be understood that the system comprises at first two separate cell populations. However, after expression of the receptor (be it natural or induced) in the one cell population and the GP-C protein from Lassa Virus in the other cell population, and after bringing the two cell populations together, the system comprises fused cells, unless such fusion is inhibited by an inhibitory compound. Importantly, before fusion has occurred, said first and said second nucleic acid are 'separately present' in the system. In a preferred embodiment the reporter gene is selected from the group consisting of: β-galactosidase, β- glucuronidase, Renilla luciferase, firefly luciferase, alkaline phosphatase, β-lactamase, horseradish peroxidase, chloramphenicol acetyltransferase, green fluorescent protein, red fluorescent protein, yellow fluorescent protein, and blue fluorescent protein, whereas it is more preferred to use β-galactosidase. In another preferred embodiment, the responsive promoter element is the HIV-I- long-terminal-repeat. Also preferred is an embodiment, wherein said protein able to act on said responsive promoter element is a transactivator protein. A 'responsive promoter' element is a promoter that responds in ways of transcription enhancement or reduction upon binding of a DNA-binding protein and so, this binding is characterized in that the binding either activates or represses the expression from the downstream located gene. The gene may also be located upstream depending on the type of responsive element is used. Preferably however, the responsive element is up- regulated in the sense that the expression of the protein encoded by the reporter gene is increased. Also preferred is an embodiment, wherein said transactivator protein is the HIV-1-Tat protein.

In yet another preferred embodiment, the cell population that expresses said Lassa Virus receptor is made of human cells. More preferably, both cell populations are human. In the event that the cell population does not express enough of the natural occurring receptor on its surface, it is preferred that the second cell population is transfected with a nucleic acid encoding the Lassa Virus receptor alpha-dystroglycan, whereas it is even more preferred that the nucleic acid encoding the Lassa Virus receptor is then stably integrated in the genome of said cells upon transfection with the encoding nucleic acid.

In another preferred aspect, said first cell population is transfected with a nucleic acid sequence encoding the Lassa Virus GP-C protein, whereas it is even more preferred that the nucleic acid encoding the Lassa Virus GP-C protein is stably integrated in the genome of said cells upon transfection .

Since expression levels are important for the present invention to obtain cell fusion and thus, to apply the system in anti -viral compound screening, it is required that the expression levels of the receptor as well as of the GP-C protein (cleaved into GP-I and GP-2) is high enough to indeed obtain cell fusion. An important aspect of the present invention is the expression level and the stability of the resulting RNA and protein in the cell populations used. Therefore, it is preferred that the nucleic acid sequence encoding the Lassa Virus GP-C protein is codon- optimised, as this results in (stable) levels that are high enough to rise above background. In another preferred aspect, the promoter regulating the expression of the GP-C protein is strong enough to yield high expression levels. Preferably, the GP-C gene is under the control of the CMV promoter. As the stability of the RNA and the protein in a certain cellular environment depends on the cell in which it is expressed, and since human cells are preferably used in the system according to the present invention, it is preferred that the nucleic acid sequence encoding the Lassa Virus GP-C protein is codon-optimised for improved expression in human cells. As used herein, codon- optimization relates to nucleic acid sequences that have been modified in a way, that the triplet codons have been altered such that increased stability and/or increased expression of the encoded protein is obtained without influencing the encoded amino acid of such altered codons . Hence, the encoded protein is preferably still wild type in the sense of amino acid content, but is altered on the nucleic acid level such that triplets are used that deliver more stable RNA.

The invention also relates to a method for fusing cells comprising culturing the cell-based system according to the invention in an acidic medium, and allowing the cell populations in said system to fuse. An acidic medium is a medium having a pH lower than pH 7. In a preferred embodiment of the invention the medium wherein said cell populations fuse is lower or approximately equal to 4.5. Additionally preferred is an embodiment, wherein the medium wherein said cell populations fuse is approximately equal to 4.0. Of course, there is a lower limit to the pH of the acidic medium that can be applied. However, such lower limit depends on the cells used. It can be envisioned that some cells can still survive at pH levels as low as 3.0, whereas other cells cannot. Also the time period in which a certain cell can withstand the low pH differs from cell to cell. However, in the course of the present invention, it was found that the optimal pH for the claimed system was around 4.0 as this enabled one to perform the cell fusion in a period of time that was long enough to have the cells survive and yet to obtain sufficient cell fusion that can be studied. The person skilled in the art will be able to determine the optimal pH level depending on the cell lines used and the period of time required to perform the fusion and/or to subsequently perform anti-fusion activity from a certain compound is a screen. It seems however, that pH 4.0 is more or less the optimal pH for providing the most appropriate environment for performing the methods according to the invention. Interestingly, this value is significantly- lower than the known pH of the endosomes, which is around 5.3.

In another preferred embodiment, the acidic medium is applied for a time period shorter than 30 minutes, as this leaves enough cells surviving the assay. Also preferred is an embodiment, wherein cell fusion is quantified by reporter gene activity in a time range of 15 minutes to 48 hours after incubation with said acidic medium, and wherein reporter gene activity is determined in unlysed cells or in lysed cells. When unlysed cells are used, one does not have to include another step of lysing the cells, which would free the reporter gene product. However, if the reporter gene product cannot be determined when still present within the cell, it is preferred to use an extra step in which the cells are lysed before measuring the reporter activity. Hence, the choice of lysing or not lysing the cells depends on the reporter gene used, and on the ease of handling the cells in a high-throughput assay.

Furthermore, the invention relates to a method for identifying a compound that inhibits Lassa Virus cell entry, comprising the above-described method with the further step of incubating said cell populations with a compound, wherein said method is performed in a high-throughput setting and/or wherein said compound is part of a compound library. A high- throughput setting refers to an experimental set-up, wherein more than one compound is tested simultaneously. A compound library is a collection of more than one compound, preferably thousands or even millions of independent compounds. In a preferred embodiment the invention relates to a method, wherein said compound is selected from the group consisting of: a proteinaceous molecule, a small nature-derived compound, a small synthetic compound, and an antibody or a binding fragment thereof .

The invention further relates to a use of a system according to the invention for identifying a compound that inhibits fusion between the two cell populations. And in yet another aspect of the invention, the invention relates to the compounds itself that are identified by a method or a use according to the invention.

The invention is further illustrated by the following non- limiting examples.

Example 1. Plasmids and cell lines used

Cells:

Vero cells (ATCC No. CCL- 81; kidney cells from the African green monkey) and 293 cells (ATCC No. CRL- 1573; human embryo kidney cells) were grown in Dulbecco's modified Eagle medium (DMEM) with 1 mM pyruvate and supplemented with 10% heat- inactivated fetal calf serum (FCS) , 100 units/ml penicillin, 0.1 mg/ml streptomycin, and 2 mM L-glutamine. 293 cells where chosen due to their intrinsic expression of the Lassa Virus receptor, reflected by findings showing that Lassa

Virus replication to high titers in those cells was possible {Lukashevich et al . 1983, Acta Virol. 27 (3) : 282-5) .

Plasmids : pAdApt-LV-GP-C, a eukaryotic expression plasmid encoding a codon-optimised form of the Lassa Virus GP-C protein (strain Josiah, NCBI Ace. No. AY628203) was used as the expression plasmid as it contains a very strong CMV promoter. The codon-optimised gene was purchased from GeneArt (Regensburg, Germany) .

The pAdApt plasmid, which was used as an expression vector for the codon-optimised Lassa Virus GP-C, was described before in detail (see WO 00/70071) . The codon-optimised Lassa Virus GP-C (strain Josiah) was generated with the GeneOptimizer^® sequence optimisation system of GeneArt, and the complete optimised nucleotide sequence is shown in figure 4 (SEQ ID N0:l) . SEQ ID NO : 2 provides the encoded amino acid sequence of the coding sequence of the GP-C gene. The codon usage was adapted to the codon bias of human genes, as explained above. In addition, regions of very high (i.e. >80%) or very low (i.e. <30%) GC content were avoided where possible. Moreover, the following cis-acting sequence motifs were avoided: internal TATA boxes, chi-sites and internal ribosomal entry sites, AT-rich or GC-rich sequence stretches, repeat sequences and RNA secondary structures, cryptic splice donor and acceptor sites, branch points, poly (A) sites, and Hindi I I and Xbal restriction sites.

Instead, HindiII and Xbal sites were chosen as restriction sites for the cloning of the entire GP-C into the pAdApt vector, and therefore synthesized upstream of the start codon (Hindlll) or downstream of the stop codons (Xbal) . For efficient translation termination two stop codons were added before the Xbal site. Additionally, a Kozak-consensus sequence was introduced between the HindiII site and the ATG start codon in order to improve translational initiation. The newly generated codon-optimised Lassa Virus GP-C sequence contains 1476 nucleotides, as does the wild-type

Lassa Virus GP-C, specified above. Prior to the transfection experiments described hereafter, pAdApt-LV-GP-C plasmid DNA was transformed by methods well known in the art into E. coli (strain XL-I blue) in order to multiply the plasmid. Multiplication was performed under ampicillin selection and the vector was then purified with the QIAgen-High-Speed- Plasmid-Maxi kit (QIAGEN, Hilden, Germany) , according to the manufacturer's description. Before transfecting the vectors the sequences were checked using the following primers: forward primer 5'-CAT TGG AAG CTT GCC ACC ATG GGC-3' (SEQ ID NO: 3) and reverse primer 5' -ATC TCG TCT AGA TTA TCA TCT CTT- 3' (SEQ ID N0:4) . Sequencing was done using the ABI Prism 3700 DNA Analyser (Amersham Biosciences) .

The plasmid encoding the HIV-1-tat protein (pL3tat) and the reporter construct harbouring the HIV-I long terminal repeat (LTR) linked to the downstream located β- galactosidase (β-gal) encoding gene (pJK2) , were both a kind gift of V. Bosch (DKFZ Heidelberg, Germany) and described before {Schwartz et al . 1990, J. Virol. 64 (6) :2519-29; Kimpton and Emerman 1992, J. Virol. 66 (4) :2232-39) . The expression plasmid encoding the LCMV glycoprotein GP-C and used in the experiments as a positive control (LCMV-GP- C; strain WE) , was used in combination with plasmid pCAGGS- LCMV-GP-C as described before. The plasmids were kindly provided by O. Lenz {Beyer et al . 2003, J Virol. 77:2866- 72) . The expression plasmid of the Influenza Virus hemagglutinin HA protein, subtype H7 , was kindly provided by R. Wagner and also described before {Wagner et al . 2005, J. Virol. 79:6449-58) .

Example 2. Cell fusion induced by expression of the Lassa Virus glycoprotein This example describes how cells are transfected and subsequently used in a cell fusion assay followed by measuring the reporter gene product activity. Figure 1 shows the principle of the recombinant cell -cell fusion assay used to quantitatively determine the fusion process used by enveloped viruses to enter cells. The adequacy of the assay was tested using Influenza Virus HA (subtype H7) expressing Vero cells and fusing them with 293 cells - measured by HIV- 1-tat transactivation of HIV-1-LTR-β-gal - by applying different pH conditions for 15 mm at 37⁰C.

Transfections

Cells were precultured at 37°C and 5% CO₂ in several 75 cm² flasks (Greiner Bio-One, Germany) until they reached confluency, which corresponded to approximately 1.5xlO⁶ cells. The day before transfection the cells were split into 6-cm dishes to achieve densities of approximately 4xlO⁵ cells (i.e. confluency of ca. 90%) on the day of transfection. All transfections for the assay were done using Lipofectamine® 2000 ( Invitrogen) . A total of 8 μg of DNA was transfected in each 6-cm dish using 20 μl of Lipofectamine 2000 according to the manufacturer's instructions. If two plasmids were transfected (e.g. Lassa Virus GP-C and HIV- 1 -LTR- β-gal) 4 μg of each plasmid were transfected using 20 μl of Lipofectamine 2000.

Cell fusion

6 h post-transfection in 6-cm dishes, Vero cells that were transfected with nucleic acid encoding the viral glycoproteins HA (H7) , LCMV GP-C, and Lassa Virus GP-C and that were co-transfected with the reporter construct pJK2 , trypsmated, resuspended in DMEM/FCS, and co-cultured overnight in 12-well dishes with 293 cells that were transfected with the plasmid encoding the transactivator (pL3tat) and that were equally trypsinated and resuspended. The cells were mixed at a ratio of 1:1. The next morning cell mixtures were exposed for 15 min to RPMI 1640 medium (GIBCO) supplemented with 2% FCS and adjusted to different pH with citric acid, and then incubated at 37⁰C for 15 min. The cell mixtures were then washed twice with PBS and incubated with DMEM/FCS for a further 24 h to allow the transactivation on the reporter construct pJK2 by the transactivator protein HIV-1-tat. As controls pAdApt without an insert or pCAGGS-LV-NP (encoding the Lassa Virus nucleoprotein) each together with pJK2 co-transfected Vero cells were mixed with pL3tat transfected 293 cells and treated as above.

Measurement of β-gal activity

Cell-cell fusion was quantified by lysing pH-shifted cells

24 h post-pH-shift in 100 μl β-gal reporter gene buffer (Roche) as described in the product descriptions. Lysates were cleared of cellular debris by short centrifugation (2 min, 10,000 rpm in a table top centrifuge) . Then, 100 μl 1 mM chlorophenol red β-D-galactopyranoside (CPRG) as β-gal substrate was added to the cleared lysate supernatants and the colorimetric β-gal -induced absorption change of CPRG was measured as optical density (OD) at 570 nra in an ELISA-plate Dynatech MR7000 reader (Dynatech, Germany) .

The recombinant cell-cell fusion assay with different viral glycoproteins and different pH-shift conditions delivered the following results: Figure 2 shows that H7-HA induced cell-cell fusion measured colorimetrically could be detected after a shift below pH 6, peaking at pH 5.5 - conditions considered to be those for Influenza Virus HA mediated fusion (Colman and Lawrence 2003, Nat. Rev. MoI. Cell Biol. 4 (4) : 309-19) . Vero cells expressing codon-optimised Lassa Virus GP-C also fused with 293 cells after a 15 min pH-shift. Remarkably, only a pH- shift ≤ pH 4.5 could induce cell-cell fusion, with maximal fusion at pH 4.0. When LCMV GP-C was transfected instead of Lassa Virus GP-C cell-cell fusion after acidic treatment within the same range as the one for Lassa Virus GP-C could be observed.

The conclusion from these experiments is that the assay described here offers an adequate system to study Lassa

Virus entry. This was shown by the finding that only after exposure to acidic pH Lassa Virus GP-C induced cell-cell fusion. This is consistent with the model that Lassa Virus entry in infected cells requires the fusion of the viral envelope with inner-cellular membranes, a process presumably taking place in the acidic milieu of the endosomal/lysosomal compartments .

Surprisingly, the pH to induce GP-C-induced cell-cell fusion was below pH 5.3, the lowest pH prevailing in endosomal compartments. Lassa Virus GP-C only showed cell- cell fusion at pH significantly lower than pH 5.3, namely 4.5 or 4.0, respectively. This effect was clearly not anticipated and fully unexpected. In view of what is known in the art in relation to cell entry by Lassa Virus, this is of high significance since this finding may imply that Lassa Virus enters the cells not via endosomes but rather via lysosomes, where pH can reach values of 4.4 (Overly et al . 1995, PNAS 92(8).-3156-60) . Additionally, this finding might imply that co-factors on top of the receptor alpha- dystroglycan are present to enable the entry of Lassa Virus into the cell. If so, these factors may be included in systems similar as to those described herein. Whatever the case, the inventors of the present invention have now found that the optimal cell fusion and thus, the optimal conditions for studying antiviral compounds are at such low pH levels. In sum, in a system according to the invention, preferably the pH is set to be approximately equal to or lower than 4.5. More preferably, the pH is approximately equal to 4.0.

Another important feature of the assay is the HIV-I- LTR- β -gal expression vector. It is important that it is not automatically or intrinsically activated in cells that are transfected with this construct. For instance, it was found when transfecting a 293 subclone cell line of unknown origin with HIV-I -LTR- β -gal, that 24 h post-transfection β-gal- activity could be detected in these cells, possibly because of a stably integrated HIV-1-tat gene in this cellular subclone (data not shown) . Such subclones were described in the art {Negrini et al . 1991, Biotechniques 10 (3) : 344-53) and care should be taken not to use those cells in the assay. However, one clearly understands that such cells could still be used as HIV-1-tat expressing cells. This seems even favorable in view of an extensive HIV-1-tat expression level, which could possibly enhance the read-out of the reporter-gene-activity.

A further important feature of the invention is the use of human cells as receptor expressing cells. For instance, chicken fibroblasts that were transfected with human alpha- dystroglycan and HIV-1-tat in the assay as described above, showed that such cells did not fuse with GP-C- and HIV-I- LTR-β-gal-transfected Vero cells after pH-shift (data not shown) .

Furthermore, the results of the recombinant cell-cell fusion assay were greatly improved by using the codon- optimised Lassa Virus GP-C. Transfecting wild-type Lassa Virus GP-C (strain Josiah, cloned into the pCAGGS vector) m the assay did also lead to cell-cell fusion after a pH-shift £ 4.5 (data not shown) . However, cell-cell fusion was less efficient when compared to codon-optimised GP-C. When the empty pCAGGS vector was transfected instead of the wild-type Lassa Virus GP-C in pCAGGS the difference between GP-C- induced fusion and background fusion was smaller, m some experiments background was even higher than GP-C-expressmg 'positive control' (data not shown) .

In addition, the time to apply acidic medium is an important feature of the assay. Since the optimal pH range {<_ 4.5) to induce cell-cell fusion is very low, damage can be done to the cells, if applied for extended time periods. Cell-cell fusion m the assay described here could be found after pH-shifts ranging from 1 mm to 30 mm (data not shown) . When incubating the cells longer with the acidic pH- shift medium a high number of cells detached from the plates, which indicated cell death of these cells, so that pH-shifts longer than 30 mm are not beneficial to the results of the assay. Hence, preferably the low pH is applied for periods of time that are shorter than 30 mm.

Moreover, it is highly preferred to use cells m the assay that are adherent. Suspension cells can also be used, but adherent cells are beneficial in that they do support the medium changes better and without detachment, especially in cases where receptor-binding of GP-C is not strong due to reduced alpha-dystroglycan expression levels.

Example 3. Screening for anti-viral compounds

The above-described system is to be used in an experimental setting to screen for compounds that may inhibit the cell-cell fusion process induced by Lassa Virus GP-C. A screening example is provided herein.

Such an experimental setting comprises seeding of cell population A and cell population B in two different 96-well plates, such that said cells are 90% confluent in each well of said 96-well plates on the day of transfection. Said cells A are transfected with Lassa Virus GP-C and HIV- 1-LTR-β-gal using Lipofectamine 2000. 0.1 μg of both constructs together with 0.5 μl Lipofectamine 2000 are transfected in each well. Cells A in three wells are not transfected (control). Said cells A (transfected and untransfected) will be mixed with cells B (equally seeded in 96-well plates) , which either intrinsically express alpha- dystroglycan or express it after transfection with an alpha- dystroglycan expression vector, said cell B additionally expresses HIV-1-tat, again either stably or after transfection. Again, transfection is performed using

Lipofectamine 2000, transfecting 0.1 μg Plasmid together with 0.5 μl Lipofectamine 2000 per well.

Then cells A and cells B are mixed after all above described proteins are expressed {e.g. 6 h post-transfection) into 2 96-well plates, thereby being incubated in cell culture medium (e.g. DMEM with 10% FCS) . A potentially anti-viral compound is added to said medium of said cell mixtures. This compound can be a proteinaceous molecule diluted adequately (e.g. at nano- or micromolar concentrations) in the cell mixture medium (e.g. DMEM with 10% FCS). A proteinaceous molecule may be a peptide, a polypeptide or a protein. Peptides are strings of amino acids linked together by a peptide bond. Although not precisely defined, peptides typically comprise between 2 and 50 amino acids. Polypeptides are longer peptides that may contain up to thousands of peptide bond-linked amino acids. The words polypeptide and protein are often interchangeably used to describe single, long polypeptide chains. In addition, proteins may consist of multiple polypeptide chains that collectively form the basis of a complex three-dimensional structure. A peptide, a polypeptide and/or protein may comprise modifications such as those generated by a cellular protein modification machinery.

Instead of a proteinaceous molecule other molecules like chemical compounds (see detailed description above) can be added accordingly. As a control three wells of cell-mixtures (cells A plus cells B) are incubated with cell culture medium that does not contain the potentially anti -viral compound.

As a further control, three wells of cell-mixtures (cells A plus cells B) are incubated with above described cell mixture medium that contains equal amounts of the solvent used to dissolve above described potentially antiviral compound, to account for effects of said solvent (e.g. dimethyl sulfoxide (DMSO) , that is used to dissolve hydrophobic proteinaceous compounds) .

After the interaction of Lassa Virus GP-C with its receptor alpha-dystroglycan (e.g. 24 h after mixing cells A and cells B) cell medium containing potential anti -viral compound is replaced by acidic medium (preferably RPMI 1640 medium containing 2% FCS and titrated to pH 4 with citric acid) also containing said anti -viral compound. As further controls three 96-well plate wells of cell- mixtures (cells A plus cells B) are either incubated with acidic medium that does not contain the potentially antiviral compound, or with neutral medium (preferably RPMI 1640 medium containing 2% FCS) . After incubation with acidic medium (or neutral medium for one set of controls) for, e.g. 15 min, but preferably not longer than 30 min, said pH- shift medium is replaced by fresh cell culture medium (e.g. DMEM containing 10% FCS), which again contains the anti -viral compound. As one additional control three wells of a 96-well plate of cell- mixtures (cells A plus cells B) are incubated with fresh cell culture medium that does not contain the potentially anti -viral compound.

In three wells containing said mixture of transfected cells A and transfected cells B all incubations are performed as described above, with the exception that no potentially anti-viral drug is added to said cells at any time of the experiment (positive control) . After cell-cell fusion (that occurs in the positive control) an adequate time to allow transactivation of the reporter- gene is allowed, e.g. 24 h post-pH-shift .

After reporter-gene expression, above described cell mixtures are tested for reporter gene activity, e.g. as β- gal reporter gene activity. In particular this is lysis of all samples in 7.5 μl β-gal-reporter-gene buffer, centrifugation to clear lysate from cellular debris, and subsequent incubation of cleared lysate with 7.5 μl 1 mM CPRG incubation of lysates . Finally, absorption at 570 nm is measured, where reporter-gene activity is expected to be found in the positive control, and no reporter-gene activity in samples that were in contact with the anti-Lassa Virus compound .

Claims

1. An in vitro cell-based system for mimicking the entry of a Lassa Virus into a host cell by cell fusion, said system comprising two cell populations, wherein the first cell population expresses the glycoprotein (GP-C) or a functional homologue thereof from Lassa Virus, and the second population expresses a receptor for Lassa Virus, said system further comprising: - a first nucleic acid comprising a nucleic acid sequence encoding a reporter gene product, and a responsive promoter element; and a second nucleic acid encoding a protein able to act on said responsive promoter element, wherein said first and said second nucleic acid are separately present in said first or said second cell population before cell fusion between the two cell populations has occurred.

2. A system according to claim 1, wherein said reporter gene is selected from the group consisting of: β- galactosidase, β -glucuronidase, Renilla luciferase, firefly luciferase, alkaline phosphatase, β-lactamase, horseradish peroxidase, chloramphenicol acetyltransferase, green fluorescent protein, red fluorescent protein, yellow fluorescent protein, and blue fluorescent protein.

3. A system according to claim 1 or 2 , wherein said responsive promoter element is the HIV-1-long-terminal- repeat .

4. A system according to any one of claims 1-3, wherein said protein able to act on said responsive promoter element is a transactivator protein.

5. A system according to any one of claims 1-4, wherein said transactivator protein is the HIV-1-Tat protein.

6. A system according to any one of claims 1-5, wherein the cell population that expresses the Lassa Virus receptor are human cells.

7. A system according to any one of claims 1-6, wherein both cell populations are human cells.

8. A system according to any one of claims 1-5, wherein said second cell population is transfected with a nucleic acid encoding the Lassa Virus receptor alpha-dystroglycan.

9. A system according to claim 8, wherein the nucleic acid encoding the Lassa Virus receptor is stably integrated in the genome of said cells.

10. A system according to any one of claims 1-9, wherein said first cell population is transfected with a nucleic acid sequence encoding the Lassa Virus GP-C protein.

11. A system according to claim 10, wherein the nucleic acid encoding the Lassa Virus GP-C protein is stably integrated in the genome of said cells upon transfection .

12. A system according to any one of claims 1-11, wherein the nucleic acid sequence encoding the Lassa Virus GP-C protein is codon-optimised, and preferably is under the control of a strong promoter, preferably the CMV promoter.

13. A system according to claim 12, wherein the nucleic acid sequence encoding the Lassa Virus GP-C protein is codon-optimised for improved expression in human cells.

14. A method for fusing cells comprising culturing the cell-based system according to any one of claims 1-13 in an acidic medium, and allowing the cell populations in said system to fuse.

15. A method according to claim 14, wherein the medium wherein said cell populations fuse is lower or approximately equal to 4.5.

16. A method according to claim 15, wherein the medium wherein said cell populations fuse is approximately equal to 4.0.

17. A method according to any one of claims 14-16, wherein said acidic medium is applied for a time period shorter than 30 minutes.

18. A method according to any one of claims 14-17, wherein cell fusion is quantified by reporter gene activity in a time range of 15 minutes to 48 hours after incubation with said acidic medium.

19. A method according to claim 18, wherein reporter gene activity is determined in unlysed cells.

20. A method according to claim 18, wherein reporter gene activity is determined in lysed cells.

21. A method for identifying a compound that inhibits Lassa Virus cell entry, comprising the method according to any one of claims 14-20 with the further step of incubating said cell populations with a compound.

22. A method according to claim 21, wherein said method is performed in a high-throughput setting and/or wherein said compound is part of a compound library.

23. A method according to claim 21 or 22, wherein said compound is selected from the group consisting of: a proteinaceous molecule, a small nature_.-derived compound, a small synthetic compound, and an antibody or a binding fragment thereof .

24. Use of a system according to any one of claims 1-13 for identifying a compound that inhibits fusion between the two cell populations.

25. A compound identified by a method according to any one of claims 21-23 or by a use according to claim 23.