Virus expressing a modified e2 protein and uses thereof, especially for specific targeting of alpha5beta1-positive cancer cells

Abstract

The present invention relates to a recombinant protein having at least 90% of identity with the protein of sequence SEQ ID NO: 1, said percentage of identity not including gaps due to possible insertions and / or deletions, characterized in that said protein comprises at least one substitution between amino acids 213 and 234, and / or at least one substitution between amino acids 70 and 78. It also relates to a nucleic acid encoding the recombinant protein, an expression vector into which the nucleic acid is inserted, a cell transfected with the expression vector, a virus expressing the recombinant protein, a virus for use as a medicament, and for use in the treatment of a cancer. The present invention relates in addition to a method of in vitro neutralisation of at least one cancer cell expressing integrin α5β1 at its plasma membrane, by administration of the virus, and to a method for producing a virus.

Description

[0001] VIRUS EXPRESSING A MODIFIED E2 PROTEIN AND USES THEREOF, ESPECIALLY FOR SPECIFIC TARGETING OF ALPHA5BETA1- POSITIVE CANCER CELLS

[0002] Technical field

[0003] The present invention refers to a recombinant protein having at least 90% of identity with the protein of sequence SEQ ID NO: 1, a nucleic acid encoding the recombinant protein, an expression vector into which the nucleic acid is inserted, a cell transfected with the expression vector, a viral vector, a virus expressing the recombinant protein, and a virus for use as a medicament.

[0004] Therefore, the present invention has utility in medical fields, especially in gene therapy.

[0005] In the description below, the references into brackets ([ ]) refer to the listing of references situated at the end of the text.

[0006] Background of the Invention

[0007] With almost 10 million deaths in 2020 (i.e. one in six), cancers are the second leading cause of death from disease worldwide. Most OECD (Organisation for Economic Co-operation and Development) countries have a prevalence of cancer above 1000 per 100,000 population. Although there are local variations, cancers are the first or second cause of death before the age of 70 in 112 out of 183 countries, and the third or fourth cause in a further 23 countries. The prevalence of cancer is tending to increase because of rising life expectancy and growing exposure to carcinogenic pollutants. However, the constant improvement in the quality and earliness of diagnoses is leading to better detection and follow-up of cases. This improvement in diagnosis means that treatments need to be improved to cope with this growth and the diversity of cases.

[0008] Even if they are grouped together under the same term, cancers are not uniform and define a group of diseases with complex origins withoutbeing able to determine a general cause for their occurrence. Thus, a wide range of characteristics must be taken into account to establish an effective treatment protocol for a given type of cancer. However, certain characteristics tend to be common to certain cancers and are frequently found in different cases. Many therapies have been developed with this in mind. However, basing treatments on shared characteristics has led to the development of treatments that are not very specific, and that can cause undesirable side effects. In many cases, a fraction of non-cancerous cells share characteristics with the transformed cells, making them unwanted targets for treatment. A great deal of attention is therefore being paid to the development of increasingly specific treatments to enable more personalised and, hopefully, more effective medicine.

[0009] Therefore, to avoid the side effects that can lead to treatment discontinuation, new approaches have been designed to enable more specific targeting. The deregulation of cancer cells sometimes results in the ex novo expression or strong overexpression of certain markers compared with healthy cells, providing a new criterion for discriminating between healthy and transformed cells. These approaches are based on the identification of cancer cells using their surface identity. Generally, molecules capable of specifically binding target markers are used to activate the treatment. Implementing these treatments requires in-depth and exhaustive characterisation of cancers, and opens the way to personalised protocols depending on the cancer diagnosed.

[0010] The therapeutic delivery of a gene is a promising approach. This is why viral vectors are an attractive option because of their ability to transduce human cells. Here, the strategy relies on the ability of the virus to induce cell death by efficiently delivering suicide genes. Viral vectors are systematically modified to render them incapable of replication while preserving their ability to deliver and enable expression in the host cell of genes contained in the genetic material they carry. These are known as viral vectors. Initially developed with the aim of correcting hereditary genetic disorders, their interest has gone beyond this use. Indeed, as cancers are generally causedby multi-gene disorders, the corrections that would be necessary are so complex that the solution of eliminating the cell by introducing a single toxic gene is often preferable.

[0011] Various therapeutic and suicide genes can be carried by viral vectors as part of cancer therapy. These include tumour suppressor genes, tumour-associated antigens, pro-inflammatory interleukins, immune checkpoint inhibitors, anti-angiogenic proteins, small interfering RNAs (siRNAs) and cytotoxic proteins. The vectors used are mainly of three types: Adeno-associated viruses (AAV), Adenoviruses and Retroviruses / Lentiviruses (LV). The difficulty with virus-based cancer therapies lies in the need for in vivo transduction.

[0012] In the treatment of cancer, LVs play an important role in research in the vectorisation of gene therapy against cancers. LVs are enveloped viruses of the RV family, with the main characteristic of being able to integrate their genome into that of the host cell. This property has made these viruses the preferred choice for the development of vectors when transgene expression needs to be particularly durable. The first applications of these vectors did not focus directly on anti-cancer therapies, but on the correction of monogenic disorders. When the disease is caused by a single defective gene, the ultimate aim of these therapies is to replace the defective gene with an exogenous gene enabling recovery of a wild-type phenotype, which can therefore be provided by a lentiviral vector (LV vector).

[0013] However, this approach may prove impossible in many cancers in which genetic disorders have multiplied during tumorigenesis. Given the limited encapsidation capacity of the vector, correcting genetic disorders by adding therapeutic genes would require numerous transductions to achieve complete correction. The simplest solution is therefore to eliminate the deregulated cell rather than correct it. To achieve this, the therapeutic transgene must enable the cancer cell to be eliminated - suicide gene therapy (SGT). Different strategies using different genes are proposed to achieve this, for examples enzymes, mediators of apoptosis, and tissue specific promoters.In this purpose, when transfecting producer cells, the choice of envelope glycoprotein (EGP) determines the tropism of the viral vector. In natural infections, the type of EGP carried by a virus is dictated by the selection of the most appropriate target cells for replication. The choice of target cell is based on the interaction between the EGPs (for enveloped viruses) and a protein (defined as a receptor) present on the surface of the target cell. This interaction must be specific and sufficiently strong to ensure that the virus and the cell are kept in close proximity, but must not interfere with the subsequent membrane fusion step. For viral vectors, the aim is for this interaction to take place between the EGPs and a surface protein specific to the cell targeted by the therapy (in the case of cancer, a tumour marker). The receptor of natural infections almost never corresponds to a specific tumour marker, which means that, in order to target cancer cells specifically, the natural EGPs have to be modified. Thanks in particular to pseudotyping, which makes it possible to substitute the EGPs of one virus with those of another, it is possible to use EGPs with more interesting characteristics for vectorisation. Interestingly, the pseudotyped LV vector also acquires the mechanism of entry of the EGPs it carries.

[0014] Another problem is that not all cancer cells have a sufficiently specific marker to enable targeted therapy by recognition of their surface identity.

[0015] Integrin a501 is a transmembrane protein belonging to the integrin family, which plays a fundamental role in cell-extracellular matrix (ECM) communication. Integrin a501 is a heterodimer composed of two subunits, a5 and 1. This integrin is a member of the RGD integrin family. It interacts with the tripeptide motif Arginine-Glycine-Aspartic Acid present on fibronectin, its natural ligand. Structural studies of a501 have shown that the specific residue (Asp154) can be used to distinguish a5 from other a-subunits because this amino acid helps direct the dimer's preference towards fibronectin, a preference not possessed by other RGD integrins. Another factor in this preference is the dimer's ability to bind a second motif present on fibronectin, the PHSRN motif (Feng and Mrksich, 2004 ([1])), which only a501 is capable of binding. Overexpression of integrin a501 hasbeen observed in several types of cancer, including breast, prostate, pancreatic and brain (glioblastoma). It is associated with an increased capacity for tumour cells to invade neighbouring tissues and form metastases. This is largely due to its role in signalling cell migration and invasion. a5 1 is a marker of severity; when a cancer cell expresses it, the prognosis is inevitably negative.

[0016] Therefore, integrin a5 1 could be an interesting target in some cancers. For example, the expression of integrin a5 1 is low in normal brain tissue, but high in certain cases of glioblastoma multiforme (GBM), opening the way to integrin-dependent targeting. Furthermore, Dr Dontenwill's team have shown that a5 1 proves to be a notable severity factor, its overexpression in GBM is linked to decreased survival of patients under standard treatments , invasion of surrounding tissues, distant migration, angiogenesis and resistance to therapies. Indeed, Dr Dontenwill's team demonstrated that when integrin a5 levels are low, p53 is fully functional. However, when a5 is overexpressed, p53-mediated responses, enabling sensitivity to conventional treatments, are reduced. It therefore seems appropriate to target a5 1 in order to provide a new therapeutic solution for these cases, which initially have the worst prognosis.

[0017] The development of targeted therapies against a5 1+ glioblastomas relies on specific recognition of the dimer. To achieve this, an arsenal of small chemical compounds has been developed to specifically inhibit a5 1-mediated effects. However, these molecules generally do not allow direct elimination of cells carrying the dimer. Their role is to limit or abolish the actions permitted by integrins, such as migration, proliferation and the resistance associated with their overexpression. The therapeutic efficacy of the treatment will therefore depend on the complementary strategies used.

[0018] Alternatives being considered for the direct elimination of tumour cells are to arm an oncolytic virus or a SGT viral vector with an a5 1 ligand in order to direct the tropism towards these cells. The adenovirus-derived oncolytic virus Delta-24-RGD (or DNX-2401) was armed with this tri-peptide to have a targeted effect on high-grade gliomas. The virus carries aninsertion of an RGD4-C peptide, a cyclised 4-RGD repeat, at the apex of the virus fiber protein to enable it to associate directly with RGD integrins. In vitro, oncolytic virus functionalized with RGD-4C increased the number of infected a5 1+ cells by a factor of 6, and in vivo intratumoral injection of Delta-24-RGD into xenograft mice was associated with prolonged survival compared with mice injected with non-functionalized virus. Indeed, 60% of the mice treated with Delta-24-RGD survived for more than 4 months compared with only 15% of those injected with non-functionalised viruses. Following these encouraging results, a phase 1 clinical trial of this oncovirus, published in 2018, showed significant tumour reduction in 72% of patients (18 out of 25) with glioblastoma that had recurred at least once, with 20% of these patients surviving beyond 3 years.

[0019] However, the problem with adenoviruses is that this type of virus is capable of targeting any type of healthy cell expressing RGD integrins. As a result, it must be administered by intracranial injection into the tumor to avoid side effects. Furthermore, all functionalizations based on the insertion of RGD peptides or derivatives will have the major drawback of targeting all RGD integrins, leaving a large field of off-target action. Indeed, even though RGD integrins are found to be over-expressed in many cancers, their presence in healthy tissue remains significant, especially if this entire class of integrins can be the target of treatment.

[0020] Thus, a need exists of alternative tools able to specifically target a target cell, for in vivo or in vitro vectorisation.

[0021] Description of the invention

[0022] Thanks to extensive research work, the inventors have identified positions in the protein responsible for binding to the receptor for insertion of motifs of interest, making it possible to target cells of interest, while not disrupting the functionality of the proteins in charge of membrane fusion allowing the transduction of viral material to the target cell.

[0023] Thanks to these results, they have obtained a specific vector for targeting, and advantageously eliminating, a5 1+ cells.More precisely, the inventors have managed to develop EGPs, which are specific of a5 1+ cells, and to arm SGT vectors or oncolytic viruses.

[0024] Moreover, in a completely inventive way, the inventors have come up with the idea of using two ligands acting in a complementary way, rather than multiplying identical ligands as was done in the prior art.

[0025] Surprisingly, the inventors have demonstrated that it was possible to use two different ligands, inserted at two distinct points at the apex of the viral E2 protein, to change its tropism.

[0026] Furthermore, they managed to identify specific positions in EGPs that allow to specifically transduce a5 1 cells, advantageously from SGT vectors.

[0027] Interestingly, these specific positions in EGPs make them useful to build a platform of versatile tools that can be functionalized for targeting various cells of interest.

[0028] Accordingly, in a first aspect, the present invention provides a recombinant protein having at least 90% of identity with the protein of sequence SEQ ID NO: 1 , said percentage of identity not including gaps due to possible insertions and / or deletions, characterized in that said protein comprises at least one substitution between amino acids 213 and 234, and / or at least one substitution between amino acids 70 and 78 relative to sequence SEQ ID NO: 1.

[0029] SEQ ID NO: 1 corresponds to the variant m168 of the E2 protein of the Sindbis virus, which is an alphavirus in the Togaviridae family, the genome of which being a single-stranded RNA, having positive polarity 11.7 kb, with an icosahedral capsid with T=4 symmetry. Sindbis viral entry is carried out by the dimer constituted by E1 / E2 proteins. E2 binds the receptor and, once bound, E1 carries out membrane fusion. The variant m 168 of the E2 protein is described in Morizono et al., 2005 ([2]). The mutations present in E2 of the m168 variant are, compared to the wild-type E2 protein: the insertion of the ZZ domain derived from the IgG binding domain of Staphylococcus aureus protein A between amino acids 71 and 74 of the E2 protein, and the insertion of the four point mutations K159A, E160A, E216Aand T218A. Sindbis virus E2 protein mutant m 168 is incapable of promoting cell infection due to its loss of ability to interact with the natural receptor of the Sindbis virus. In SEQ ID NO: 1, the ZZ motif, comprised between amino acids 78-188 of SEQ ID NO: 1 , is then deleted to allow the insertion of new motifs.

[0030] SEQ ID NO: 1 is as follows:

[0031] SVIDDFTLTSPYLGTCSYCHHTEPCFSPVKIEQVWDEADDNTIRIQTSAQF GYDQSGAASANKYRYMAAAAVTDNKFNKEQQNAFYEILHLPNLNEEQRN AFIQSLKDDPSQSANLLAEAKKLNDAQAPKVDNKFNKEQQNAFYEILHLP NLNEEQRNAFIQSLKDDPSQSANLLAEAKKLNDAQAPKVDANSSSVPGD PVTTVKEGTMDDIKISTSGPCRRLSYKGYFLLAKCPPGDSVTVSIVSSNSA TSCTLARKIKPKFVGREKYDLPPVHGKKIPCTVYDRLAATTAGYITMHRPG PHAYTSYLEESSGKVYAKPPSGKNITYECKCGDYKTGTVSTRTEITGCTAI KQCVAYKSDQTKWVFNSPDLIRHDDHTVQGKLHLPFKLIPSTCMVPVAH APNVIHGFKHISLQLDTDHLTLLTTRRLGANPEPTTEWIVGKTVRNFTVDR DGLEYIWGNHEPVRVYAQESAPGDPHGWPHEIVQHYYHRHPVYTILAVA SATVAMMIGVTVAVLCACKARRECLTPYALAPNAVIPTSLALLCCVRSAN A

[0032] Surprisingly, modifying the E2 protein does not alter its functionality, i.e. targeting cells of interest, and does not disrupt its association with E1 (which enables membrane fusion at a later stage of infection), which still allows them to work in concert.

[0033] The terms “identical” or percent “identity” in the context of two or more nucleotide or polypeptide sequences refer to two or more sequences or subsequences that are the same or have a specified percentage of amino acid residues or nucleotides that are the same, when compared and aligned for maximum correspondence, as measured using one of the sequence comparison algorithms known in the art, for example disclosed herein, or by visual inspection.

[0034] The term “at least 90% of identity”, in the context of two amino acid sequences, refers to two or more sequences or subsequences that have in some embodiments at least 90%, in some embodiments about 90%, in someembodiments about 91%, or in some embodiments about 92%, or in some embodiments about 93%, or in some embodiments about 94%, or in some embodiments about 95%, or in some embodiments about 96%, or in some embodiments about 97%, or in some embodiments about 98%, or in some embodiments about 99%, when compared and aligned for maximum correspondence, as measured using one of the sequence comparison algorithms known in the art, for example disclosed herein, or by visual inspection.

[0035] “Said percentage of identity not including gaps due to possible insertions and / or deletions” means herein that one or more insertions and / or one or more deletions may possibly occur outside the 213-234 and the 70-78 regions, with no effect with the functionalisation of these regions.

[0036] Therefore, when comparing two amino acid sequences, algorithms have to account for gaps to handle insertions or deletions that can occur. These gaps represent regions where one sequence has additional or missing residues compared to the other.

[0037] For example, alignment of sequences for comparison can be conducted, e.g., by the local homology algorithm of Smith & Waterman (1981 ) ([3]), or by the homology alignment algorithm disclosed in Needleman & Wunsch (1970) ([4]) or by the search for similarity method disclosed in Pearson & Lipman (1988) ([5]), or Altschul et al. (1990) ([6]); or by Clustal, e.g., ClustalW, Clustal Omega, or by MUSCLE or by computerized implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the GCG® WISCONSIN PACKAGE®, available from Accelrys, Inc., San Diego, California, United States of America), Snapgene such as Version 2,8,3 ; MacOS, or by visual inspection.

[0038] “At least one substitution between amino acids 213 and 234, and / or at least one substitution between amino acids 70 and 78” refers herein to a number chosen among 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35 or 36 substitutions, that may occur in the site 213-234, and / or the site 70-78, for functionalizing these sites.In addition to the substitutions in the site 213-234 and / or the site 70-78, the recombinant protein of the invention may comprise at least one other substitution, as long as the percentage identity of the recombinant protein with SEQ ID NO: 1 is at least 90%. These other substitutions may replace, in the sequence SEQ ID NO: 1 , any amino acid, as long as there is no effect on the functionalization of the site 213-234 and / or the site 70-78. For example, substitution may be conservative amino acid substitutions that do not have a substantial effect on the functionalization.

[0039] The at least one substitution in the site 213-234 and / or the site 70-78 may introduce into the recombinant protein at least one envelope glycoprotein, part of envelope glycoprotein, DARPins, affibodies, nanobodies or any antibody derived peptides, which determines a viral tropism of interest.

[0040] Advantageously, the at least one substitution between amino acids 213 and 234 comprises a RGD tripeptide.

[0041] Advantageously, the at least one substitution between amino acids 70 and 78 comprises a PHSRN (SEQ ID NO: 2) peptide.

[0042] Possibly, the at least one substitution in the site 213-234 and / or the site 70-78 may comprise at least one linker. The at least one linker may be a flexible linker, i.e. any peptide capable of flexibly orienting ligands in space to allow correct interaction without constraining the structure of the base protein. For example, the at least one linker may have a length of 2, or of 3, or of 4, or of 5, or of 6, or of 7, or of 8, or of 9, or of 10, or of 11 , or of 12, or of 13, or of 14, or of 15, or of 16 amino acids. Preferably, the at least one linker may be mainly made up of Ala and / or Gly and / or Ser amino acid. For example, the RGD tripeptide may be surrounded by 2 flexible linkers. For example, the RGD tripeptide and the linkers may have the sequence SEQ ID NO: 3 (GGGSRGDGGG) or sequence SEQ ID NO: 4 (CGGSRGDGGA). Possibly, the PHSRN may be surrounded by 2 flexible linkers. For example, the PHSRN and the linkers may have the sequence SEQ ID NO: 5 (AGPHSRNGA) or SEQ ID NO: 6 (RAGPHSRNGAK).Preferably, the recombinant protein of the invention comprises a RGD tripeptide between amino acids 213 and 234 and a PHSRN peptide (SEQ ID NO: 2) between amino acids 70 and 78.

[0043] In an embodiment, the recombinant protein of the invention may have the sequence SEQ ID NO: 17 SVTDDFTLTSPYLGTCSYCHHTEPCFSPVKIEQVWDEADDNTIRIQTSAQ FGYDHSGAASANKYRYMTRAGPHSRNGAKLVDANSSSVPGDPVTTVKE GTMDDIKISTSGPCRRLSYKGYFLLAKCPPGDSVTVSIVSSNSATSCTLAR KIKPKFVGREKYDLPPVHGKKIPCTVYDRLAATTAGYITMHRPGPHAYTS YLEESSGKVYAKACGGGSRGDGGGASCVAYKSDQTKWVFNSPDLIRHD DHTAQGKLHLPFKLIPSTCMVPVAHAPNVIHGFKHISLQLDTDHLTLLTTR RLGANPEPTTEWIVGKTVRNFTVDRDGLEYIWGNHEPVRVYAQESAPGD PHGWPHEIVQHYYHRHPVYTILAVASATVAMMIGVTVAVLCACKARRECL TPYALAPNAVIPTSLALLCCVRSANA

[0044] The recombinant protein of the invention may be obtained by conventional techniques known from the skilled person. For example, designing a gene construct comprising a nucleic acid encoding a recombinant protein of the invention, possibly comprising adding tags such as His-tag, GST-tag, or FLAG-tag for easier purification and detection, adding cleavage sites, cloning the gene construct into an expression vector such as pET, pGEX, pcDNA, pCMV or pYES, transforming the vector into a host cell such as E. coli, Pichia pastoris, Sf9, Sf21, HEK293, HeLa, COS, NIH-3T3 or CHO, inducing protein expression under appropriate conditions for example with inducers like IPTG or adjust temperature to enhance solubility, or by controling expression with the promoter (e.g., CMV) or additives, harvesting and lysing the cells, and purify the protein, for example using SDS-PAGE, Western Blot and / or Mass Spectrometry.

[0045] Another object of the invention is a nucleic acid encoding a recombinant protein of the invention as defined above.

[0046] For example, if the recombinant protein of the invention includes RGD and PHSRN (SEQ ID NO: 2), the nucleic acid encoding the recombinant protein may have the sequence SEQ ID NO: 7, which is as follows:AGCGTCACTGACGACTTTACCCTGACCAGCCCCTACTTGGGCACATG CTCGTACTGCCACCATACTGAACCGTGCTTCAGCCCTGTTAAGATCGA GCAGGTCTGGGACGAAGCGGACGATAACACCATACGCATACAGACTT CCGCCCAGTTTGGATACGACCATAGCGGAGCAGCAAGCGCAAACAAG TACCGCTACATGACGCGTGCCGGCCCCCACAGCAGGAACGGCGCCA AGCTTGTAGACGCGAATTCGAGCTCGGTACCCGGGGATCCGGTAACC ACCGTTAAAGAAGGCACCATGGATGACATCAAGATTAGCACCTCAGG ACCGTGTAGAAGGCTTAGCTACAAAGGATACTTTCTCCTCGCAAAATG CCCTCCAGGGGACAGCGTAACGGTTAGCATAGTGAGTAGCAACTCAG CAACGTCATGTACACTGGCCCGCAAGATAAAACCAAAATTCGTGGGA CGGGAAAAATATGATCTACCTCCCGTTCACGGTAAAAAAATTCCTTGC ACAGTGTACGACCGTCTGGCAGCAACAACTGCAGGCTACATCACTAT GCACAGGCCGGGACCGCACGCTTATACATCCTACCTGGAAGAATCAT CAGGGAAAGTTTACGCAAAGGCATGCGGCGGCGGCAGCCGCGGCGA TGGCGGCGGCGCTAGCTGCGTCGCCTATAAGAGCGACCAAACGAAG TGGGTCTTCAACTCACCGGACTTGATCAGACATGACGACCACACGGC CCAAGGGAAATTGCATTTGCCTTTCAAGTTGATCCCGAGTACCTGCAT GGTCCCTGTTGCCCACGCGCCGAATGTAATACATGGCTTTAAACACAT CAGCCTCCAATTAGATACAGACCACTTGACATTGCTCACCACCAGGAG ACTAGGGGCAAACCCGGAACCAACCACTGAATGGATCGTCGGAAAGA CGGTCAGAAACTTCACCGTCGACCGAGATGGCCTGGAATACATATGG GGAAATCATGAGCCAGTGAGGGTCTATGCCCAAGAGTCAGCACCAGG AGACCCTCACGGATGGCCACACGAAATAGTACAGCATTACTACCATC GCCATCCTGTGTACACCATCTTAGCCGTCGCATCAGCTACCGTGGCG ATGATGATTGGCGTAACTGTTGCAGTGTTATGTGCCTGTAAAGCGCGC CGTGAGTGCCTGACGCCATACGCCCTGGCCCCAAACGCCGTAATCCC AACTTCGCTGGCACTCTTGTGCTGCGTTAGGTCGGCCAATGCT

[0047] Another object of the invention is an expression vector in which a nucleic acid as defined above is inserted. Suitable expression vectors are known by the skilled person, and it may be for example selected among pET, such as pET-28a, pET-21a for His-tagged proteins, pGEX, pcDNA, pCMV or pYES, this list not being limitative.Another object of the invention is a cell transfected with an expression vector as defined above. Suitable cells are known by the skilled person, and it may be for example selected among such as BHK-21, E. coli, Pichia pastoris, Sf9, Sf21, HEK293, HeLa, COS, NIH-3T3, orCHO.

[0048] Another object of the invention is a virus or a viral vector expressing a recombinant protein of the invention as defined above. The virus or the viral vector may be any kind of virus or viral vector, for example it may be selected among retroviruses, adenoviruses, and alphaviruses. Among retroviruses, it may belong to the family of retroviridae, and more specifically to lentivirus such as HIV, SIV, deltaretrovirus such as HTLV, beta retrovirus such as JSRV, gammaretrovirus such as FeLV, MLV, alpharetrovirus such as ALV. Among adenoviruses, it may belong to genera of mastadenovirus, aviadenovirus, atadenovirus, siadenovirus or ichtadenovirus. Among alphaviruses, it may be for example selected among encephalitic alphaviruses such as Eastern Equine Encephalitis Virus (EEEV), Western Equine Encephalitis Virus (WEEV), Venezuelan Equine Encephalitis Virus (VEEV), Highlands J Virus (HJV), Arthritogenic Alphaviruses such as Chikungunya Virus (CHIKV), Ross River Virus (RRV), Mayaro Virus (MAYV), O'nyong-nyong Virus (ONNV), Sindbis Virus (SINV), or other alphaviruses such as Getah Virus, Semliki Forest Virus (SFV), or Barmah Forest Virus (BFV). Prefeably, it is Sindbis Virus (SINV). For example, it may be a strain of Sindbis Virus, such as the original Sindbis Strain, isolated in 1952, Ockelbo Virus, Karelian Fever Virus, Whataroa Virus, Middle Eastern Strains, African Strains, Australian Strains, Asian Strains.

[0049] Whatever the virus, it may be an oncolytic virus, i.e. a virus that may selectively infect, replicate in, and kill cancer cells while sparing normal, healthy cells. The oncolytic property of the virus may be due to a suicide gene, i.e. a gene engineered into the virus to enhance its ability to kill cancer cells or increase the safety of the therapy. Suicide genes may encode enzymes or proteins that convert a non-toxic prodrug into a toxic substance, selectively killing the cells infected by the virus or in the vicinity of the tumor. Suicide genes are known by the skilled person, and may be selected amongHerpes Simplex Virus Thymidine Kinase (HSV-TK), Cytosine Deaminase (CD), Nitroreductase (NTR), or Carboxypeptidase G2 (CPG2). Whether it contains or not a suicide gene, the oncolytic virus may be for example Talimogene Laherparepvec (T-VEC), Oncorine (H101), Reolysin (Pelareorep), Vaccinia Virus (e.g., Pexa-Vec), Newcastle Disease Virus (NDV) or Coxsackievirus A21 (CVA21).

[0050] The vector may be prepared by any method known by the skilled person to produce a pseudotyped virus, for example as described in Duverge, A., & Negroni, M. (2020) (

[0012] ). Briefly, a gene construct as defined above is transfected or co-transfected, optionally with a transfer vector containing the suicide gene, in a packaging cell as defined above, incubate the cells to allow the cells to express the viral components and assemble pseudotyped viral particles, and harvest the pseudotyped virus.

[0051] Another object of the invention relates to a virus as defined above, for use as a medicament.

[0052] The medicament may be administered in the form of a pharmaceutical composition. A pharmaceutical composition of the presently disclosed subject matter comprises, consists essentially of, or consists of at least one active ingredient (e.g., the virus of the presently disclosed subject matter) and at least one pharmaceutically acceptable diluent and / or excipient. As used herein, the term “pharmaceutically acceptable” refers to physiologically tolerable, for either human or veterinary application. Similarly, “pharmaceutical compositions” include formulations for human and veterinary use. The term “pharmaceutically acceptable carrier” also refers to a chemical composition with which an appropriate compound or derivative can be combined and which, following the combination, can be used to administer the appropriate compound to a subject. In some embodiments, a pharmaceutically acceptable diluent and / or excipient is pharmaceutically acceptable for use in a human.

[0053] The pharmaceutical compositions of the presently disclosed subject matter can in some embodiments consist of the active ingredient alone, in a form suitable for administration to a subject, or the pharmaceuticalcomposition can in some embodiments comprise or consist essentially of the active ingredient and one or more pharmaceutically acceptable carriers, one or more additional active ingredients, or some combination of these. The additional active ingredient can be present in the pharmaceutical composition in the form of a physiologically acceptable ester or salt, such as in combination with a physiologically acceptable cation or anion, as is well known in the art. The additional active ingredient may be any therapeutic ingredient that is known to the person skilled in the art for treating or preventing the pathology of interest, for example CAR-T Cell Therapy in the treatment of cancer. It can be administered before, during or after the virus of the invention. When two or more active ingredients are to be administered, they can be administered in the same pharmaceutical composition or in separate pharmaceutical compositions. When administered in separate pharmaceutical compositions, they can be administered simultaneously or one can be administered first. The amount of time between administration of the different compounds can vary and can be determined by one of ordinary skill in the art. For example, the two compounds could be administered up to 10 minutes apart, up to 30 minutes apart, up to 1 hour apart, etc. In some embodiments, one or more of the compounds can be administered more than once. In some embodiments, a compound is administered at least twice. In some embodiments, a compound is administered at least five times. In some embodiments, the method is useful for low dose treatment. In some embodiments, the method is useful for short-term treatment.

[0054] The formulations of the pharmaceutical compositions described herein can be prepared by any method known or hereafter developed in the art of pharmacology. In general, such preparatory methods include the step of bringing the active ingredient into association with a carrier or one or more other accessory ingredients, and then, if necessary or desirable, shaping or packaging the product into a desired single- or multi-dose unit.

[0055] For example, suitable formulations can include aqueous and nonaqueous sterile injection solutions that can contain antioxidants, buffers, bacteriostatics, bactericidal antibiotics, and solutes that render theformulation isotonic with the bodily fluids of the intended recipient. With regard to administering a composition of the presently disclosed subject matter, methods are well known to those skilled in the art and include, but are not limited to, parenteral administration, including injectable such as intravenous administration, intra-arterial administration, intramuscular administration, subcutaneous administration, intravitreous administration, including via intravitreous sustained drug delivery device, intracameral (into anterior chamber) administration, suprachoroidal injection, subretinal administration, subconjunctival injection, sub-tenon administration, peribulbar administration, transscleral drug delivery, intraocular injection, intravenous injection, intraparenchymal / intracranial injection, intra-articular injection, retrograde ureteral infusion, intrauterine injection, intratesticular tubule injection, intrathecal injection, intraventricular (e.g., inside cerebral ventricles) administration, administration via topical eye drops, and the like, intracerebral administration, oral administration, transdermal administration, administration by inhalation, nasal administration, topical administration, intravaginal administration, ophthalmic administration, intraaural administration, rectal administration. Administration can be continuous or intermittent. In some embodiments, a preparation can be administered therapeutically; that is, administered to treat an existing disease or condition. In some embodiments, a preparation can be administered prophylactically; that is, administered for prevention of a disease, disorder, or condition.

[0056] An effective dose of a composition of the presently disclosed subject matter is administered to a subject in need thereof. A “treatment effective amount” or a “therapeutic amount” is an amount of a therapeutic composition sufficient to produce a measurable response (e.g., a biologically or clinically relevant response in a subject being treated). Actual dosage levels of active ingredients in the compositions of the presently disclosed subject matter can be varied so as to administer an amount of the active compound(s) that is effective to achieve the desired therapeutic response fora particular subject. The selected dosage level will depend on the activity of the therapeuticcomposition, the route of administration, combination with other drugs or treatments, the severity of the condition being treated, and the condition and prior medical history of the subject being treated. However, it is within the skill of the art to start doses of the compound at levels lower than required to achieve the desired therapeutic effect and to gradually increase the dosage until the desired effect is achieved. The potency of a composition can vary, and therefore a “treatment effective amount” can vary. However, using the assay methods described herein, one skilled in the art can readily assess the potency and efficacy of a candidate compound of the presently disclosed subject matter and adjust the therapeutic regimen accordingly. After review of the disclosure of the presently disclosed subject matter presented herein, one of ordinary skill in the art can tailor the dosages to an individual subject, taking into account the particular formulation, method of administration to be used with the composition, and particular disease treated. Further calculations of dose can consider subject height and weight, severity and stage of symptoms, and the presence of additional deleterious physical conditions. Such adjustments or variations, as well as evaluation of when and how to make such adjustments or variations, are well known to those of ordinary skill in the art of medicine.

[0057] The medicament may be used to treat any disease, and more specifically any disease that may be prevented or treated by gene therapy. It may be for example, cancers such as glioblastoma, head and neck cancer, colon cancer, melanoma, non-small cell lung cancer (NSCLC), genetic disorders such as Duchenne Muscular Dystrophy or hemophilia, blood disorders such as Beta-Thalassemia, neurological disorders such as Alzheimer’s disease, eye disorders such as Retinitis Pigmentosa, infectious diseases such as HIV, cardiovascular disorders such as familial hypercholesterolemia or heart failure, autoimmune diseases such as rheumatoid arthritis, metabolic disorders such as lysosomal storage diseases or diabetes.Another object of the invention relates to a virus of the invention as defined above, for use in the treatment of a cancer, especially a501 positive cancers.

[0058] The terms “treatment” and “treating” as used herein refer to both therapeutic treatment and prophylactic or preventative measures, wherein the object is to prevent or slow down (lessen) the targeted pathologic condition, prevent the pathologic condition, pursue or obtain beneficial results, and / or lower the chances of the individual developing a condition, disease, or disorder, even if the treatment is ultimately unsuccessful. Those in need of treatment include those already with the condition as well as those prone to have or predisposed to having a condition, disease, or disorder, or those in whom the condition is to be prevented. The term “treating” refers any effect, e.g., lessening, reducing, modulating, ameliorating, reversing, or eliminating, that results in the improvement of the condition, disease, disorder, and the like, or ameliorating a symptom thereof.

[0059] 05 1, also known as the fibronectin receptor, is an integrin that binds to matrix macromolecules and proteinases. It is composed of as (ITGA5 / CD49e) and 01 (ITGB1 / CD29) subunits. It is the primary receptor for fibronectin.

[0060] “a501 positive cancers” refers herein to cancer expressing integrin at their plasma membrane at a higher level than a healthy cell. The level may be 1.5 times more, or 2 times more, or 3 times more, even more than 3 times more, than a healthy cell.

[0061] Advantageously, the cancer may be selected in the group comprising glioblastoma, head and neck cancer, colon cancer, melanoma, non-small cell lung cancer (NSCLC).

[0062] As explained above, the virus, in the form of a medicinal product or pharmaceutical composition, may be administered systemically, for example by routes of administration oral (PO), intravenous (IV), intramuscular (IM), subcutaneous (SC), transdermal or inhalation, or locally, for example intracranial.'Another object of the invention relates to a method of in vitro neutralization of at least one cancer cell expressing integrin a5 1 at its plasma membrane, by administration of a virus of the invention as defined above. This in vitro method takes place advantageously in a sample containing at least one cancer cell expressing integrin a5 1.

[0063] “neutralization” refers herein to the fact to render the cancer cell incapable of continuing its harmful activities, such as uncontrolled growth, invasion of nearby tissues, and spreading to other parts of the body (metastasis). Neutralization may be made for example by killing cancer cells, inhibiting growth and proliferation, inducing apoptosis.

[0064] Another object of the invention relates to a method for producing a virus of the invention as defined above, comprising the steps of:

[0065] a) expressing an expression vector as defined above in a producer cell, thereby obtaining the virus in a production medium, and b) keeping the medium of viral production non-acidic through the addition of a base or buffer in the latter.

[0066] “ non acidic” refers herein to a neutral pH, i.e. 7, or a basic (alkaline) pH, i.e. greater than 7.

[0067] The buffer or the base may be any one commonly used for viral production. It may be for example Phosphate-Buffered Saline (PBS), HEPES buffer, Tris Buffer, citrate buffer, acetate buffer, DMEM or RPMI with buffers, sodium bicarbonate (NaHCO3) or sodium hydroxide.

[0068] This invention is further illustrated by the following examples with regard to the annexed drawings that should not be construed as limiting.

[0069] Brief description of the figures

[0070] - Figure 1 : represents plasmids used to produce gRNA for vectors carrying a transgene encoding either nRFP (Nuclear Red Fluorescent Protein) or nGFP (Nuclear Green Fluorescent Protein). The sequence carries the LTRs, the 3' of which is deleted to allow auto-inactivation of the provirus. The fluorescent proteins are made nuclear by fusion with histone2B to address the cell nucleus. The fluorescent protein is linked to histone 2B by a flexible linker (SEQ ID No: 8: GSGGGGSGGGGSGGGGSGGGGLACRNSISSLSDP), the protein sequence of which is shown in the diagram. The gRNA also enables expression of the puromycin resistance gene pac. The plasmid also carries the necessary PBS, Psi and RRE sequences in cis on the gRNA. The fluorescent proteins are under the control of a ubiquitous eEF-1a promoter, while pac is under the control of a ubiquitous PGK promoter.

[0071] - Figure 2 : represents schematic representation of the modified EGPs. Protein m168-RGD is derived from the m168 envelope protein and has two flexible linkers, i.e. CGGS and GGA flanking a RGD peptide inserted in place of the ZZ domain (cyan) from amino acid 70 of E2 protein. Protein m168-PHSRN, also derived from m168, has a ZZ domain substitution by peptide PHSRN (SEQ ID No: 2) flanked by flexible linkers (RAG and GAK). The proteins are schematised from N-ter to C-ter.

[0072] - Figure 3 : represents the transduction of cells with lentiviral vectors pseudotyped with mono-functionalised EGPs. U87MG cells were transduced with lentiviral vectors pseudotyped with different protein types. Expression of the nGFP transgene was analysed 3 days after transduction. The results presented show data from 3 replicates of the experiment.

[0073] - Figure 4 : represents schematic representation of doubly functionalized EGP of SinV. The m168-PHSRN protein used as the basis for construction of the other EGPs has the PHSRN peptide (SEQ ID NO: 2) flanked by flexible linkers (RAG and GAK). The RP1 protein has a second insertion consisting of the RGD peptide flanked by flexible linkers (CGGS and GGA). The insertion substitutes amino acids 274 to 280. The RP2 protein has the same insertion substituting amino acid 213 to 234. The RP2 and RP3 proteins have insertion of the same motif substituting amino acids 213 to 234 and 221 to 224 respectively. The proteins are schematised from N-ter to C-ter.

[0074] - Figure 5 : represents transduction of U87MG cells with lentiviral vectors pseudotyped with the different functionalized dual EGPs (i.e. m168(black chart), RP1 (white chart), RP2 (hatched chart) and RP3). Cells were transduced with equal quantities of pseudotyped lentiviral vectors. The percentage of nGFP+ cells corresponds to the percentage of transduced cells. Expression of the nGFP transgene was analysed 3 days after transduction. The results presented show data from 3 repetitions of the experiment.

[0075] - Figure 6: represents: western blot quantification of the relative expression of a5, [31, aV and [33. (A) Relative expression levels of the different integrin of interest. U87MGs are defined as having the reference expression level of 1. Measured values are normalised by tubulin expression measured in U87MG (black chart), U87MG CRISPR a5 (white chart) and HCC1954 (hatched chart) cells. n=2 (B) Western blot of the different integrins of interest for each cell type used (HCC1954, HEK293T, U87CRISPR and U87WT). The HEK293 cells presented here were not used in the experiment despite their characterisation.

[0076] - Figure 7 represents: Comparison of the transduction efficiency of mono-functionalised versus bi-functionalised GPEs. Transduction of U87MG, U87MG CRISPR a5 and HCC1954 cells with m168-RGD-pseudotyped lentiviral vectors (indicated RGD in the diagram, black charts) and RP2 (white charts). The percentage of nRFP+ cells corresponds to the percentage of transduced cells. Expression of the nRFP transgene was analysed 3 days after transduction. The results presented show data from 3 repetitions of the experiment.

[0077] - Figure 8 represents: Evaluation of the impact of a5[31 inhibitors on the transduction efficiency of m168-RGD and RP2 pseudotyped vectors. Transduction of U87MG cells with lentiviral vectors pseudotyped by m168-RP2 (RGD, black charts) or by RP2 (white charts). Cells were transduced with equal amounts of pseudotyped lentiviral vectors, one condition was performed without addition (U87MG), the second was performed by adding 10pM of each inhibitor at the time of transduction (U87MG + Fr248 / K34c). The percentage of nRFP+ cells corresponds to the percentage of cells transduced by each vector. Expression of the nRFP transgene was analysed3 days after transduction. The results presented show data from 3 repetitions of the experiment.

[0078] - Figure 9 represents: the schematic representation of doubly functionalized EGP of SinV. The RP2 protein was used as the basis for the construction of the RGD2x variant. The PHSRN motif of RP2, inserted into the 70-78 region, was substituted by an RGD motif flanked by flexible links. The resulting construct, RGD2x, is therefore functionalized at two points by RGD. The proteins are schematised from N-terto C-ter.

[0079] - Figure 10 represents: the comparison of the transduction efficiency of the two RGD2x versus RP2 bi-functionalised GPEs. Transduction of U87MG, U87MG CRISPR and HCC1954 cells with m168-RGD-pseudotyped lentiviral vectors (indicated by RGD 2x in the diagram, black charts) and RP2 (white charts). Cells were transduced with equal amounts of pseudotyped lentiviral particles. The percentage of nRFP cells corresponds to the percentage of transduced cells. Expression of the nRFP transgene was analysed 3 days after transduction. The results presented show data from 3 repetitions of the experiment.

[0080] - Figure 11 represents: the evaluation of the sensitivity of U87MG cells to induction of apoptosis by the suicide gene iCasp9. Relative proportion of transduced cells (nGFP+ / iCasp9+) to non-transduced cells (nRFP+ / iCasp9-). (A) Graphical representation of the ratio of iCas9+ / iCas9-cells. The squares curve indicates the culture condition without dimerization inducer. The circles curve indicates the condition with 1pM AP1903. The results shown are from 3 replicates. (B) Composite fluorescence microscopy images of the green and red fields of one replicate of the experiment.

[0081] Examples

[0082] Example 0: References used for the delimitation of Sindbis virus sequences

[0083] In order to maintain consistency with the reference framework commonly used in the existing Sindbis virus literature, region 71-74 as cited in original article Morizono et al., 2005 ([2]) and in the following example 1corresponds to region 70-78 of SEQ ID NO: 1. SEQ ID NO: 1 delimitation is consistent with a numbering that starts at the first amino acid of the E2 protein, as was done in previous studies. The latter is broader, as it accounts for certain deleted sequences.

[0084] For the second region, 213-234 from SEQ ID NO: 1, it corresponds to a portion of the region described as 318-354 in the following example 1. The 213-234 reference system is used in the description above to specifically cover the E2 protein in the context of this patent application (SEQ ID NO: 1), whereas the 318-354 reference is used in the example based on a numbering that starts at the first amino acid of the E1 protein. The latter is broader, as it accounts for certain deleted sequences. This is the same regarding region 372-379 in RP1 and 328-331 in RP3, which are based on a numbering that starts at the first amino acid of the E1 protein.

[0085] Example 1 : Development of lentiviral vectors for targeting a5B1+ cell Material and methods

[0086] Structural analysis

[0087] To identify flexible loop regions that are candidates for exogenous peptide insertion into the E2 of SinV coding sequence, the structures of the Sinv 6IMM, and 3MUU MCPs (cryo-electromicroscopy and X-ray diffraction crystallography, respectively) were analysed using PyMOL software on MacOS (Molecular Graphics System, Version 2.0 Schrodinger, LLC.). The b-factor, or atomic displacement factor, of the loops present at the E2 apex was studied for each structure. The most flexible protein sequences identified were then used for the insertion of exogenous peptides.

[0088] Plasmids

[0089] The sequence encoding the three surface proteins E1, E2 and E3 of SinV was obtained from the plasmid pm168 (Addgene plasmide # 34886). The PCR product amplified from pm 168 with an oligo hybridising 5‘ to the ZZ and bearing the Mui restriction site and an oligo hybridising 3’ to the ZZ and bearing the Hindi 11 restriction site was digested with Mlul and Hindi 11 and ligated to the DNA fragment shown below, yielding to pm168-PHSRN.5’- CGCGTGCCGGCCCCCACAGCAGGAACGGCGCCA -3’ (SEQ ID NO: 9)

[0090] 3’- ACGGGGGGGGGTGTCGTCCTTGCCGCGGTTCGA -5’ (SEQ ID NO: 10)

[0091] The coded sequence is RAGPHSRNGAK (SEQ ID NO: 6)

[0092] This fragment encodes PHSRN framed by flexible links between restriction sites allowing insertion. pm168-PHSRN was then modified by inserting the sequence shown in below between amino acids I372 and V379 to generate pCMV-RP1, between amino acids K318 and C354 to generate pCMV-RP2, and between amino acids E328 and C331 to generate pCMV-RP3. This fragment encodes RGD framed by flexible links between the Pael and Nhel restriction sites allowing insertion.

[0093] 5’- CGGCGGCGGCAGCCGCGGCGATGGCGGCGGCG -3’ (SEQ ID NO: 11)

[0094] 3’- GTACGCCGCCGCCGTCGGCGCCGCTACCGCCGCCGCGATC -5’ (SEQ ID NO: 12)

[0095] The coded sequence is : CGGGSRGDGGGA (SEQ ID NO: 13). To generate the plasmid carrying the 2x RGD dual functionalisation, we used plasmid RP2 from which the PHSRN coding fragment between Mlul and Hindi 11 had been excised and replaced by the RGD coding fragment flanked by flexible linkers (see below). This fragment encodes RGD framed by flexible links between restriction sites allowing insertion in place of the PHSRN motif of RP2.

[0096] 5’- CGCGTGGCGGCGGCAGCCGCGGCGATGGCGGCGGCA -3’ (SEQ ID NO: 14)

[0097] 3’- ACCGCCGCCGTCGGCGCCGCTACCGCCGCCGTTCGA -5’ (SEQ ID NO: 15)

[0098] The coded sequence is : RGGGGGSRGDGGGK (SEQ ID NO: 16). pHCMV-G (Zufferey et al., 1997 ([7])) encodes the VSV G protein under the control of the CMV promoter. pSDY-nRFP-Puro was obtained by modifying pSDY-dCK-Puro (Rossolillo et al., 2012 ([8])) by replacing the coding sequence for human deosycitosine kinase with that of the RFP fusedto histone 2B using the unique BamHI and Xhol sites present in pSDY-dCK-Puro. Next, pSDY-nRFP-Puro was modified by substituting GFP for RFP to obtain pSDY-nGFP-Puro. pSDY-nRFP-Puro and pSDY-nGFP-Puro, which are used to synthesise the genomic RNA of viral particles (Figure 1), were obtained by modifying pSDY-dCK-Puro (Rossolillo et al, 2012 ([8])) by replacing the coding sequence for human deosycitosine kinase with that of RFP or GFP respectively, fused to histone 2B using the unique BamHI and Xhol sites present in pSDY-dCK-Puro. pCMVAR8.91 (Zufferey et al., 1997 ([7])) encodes the HIV-1 Gag and Pol genes (short named p8.91).

[0099] Cellular cultures

[0100] HEK-293T (ATCC® CRL-3216™), U87MG (ATCC® HTB-14™), HCC1954 (ATCC® CRL-2338™) cells were sourced from the American Type Culture Collection (ATCC). U87MG CRISPR a5 were generated by Dr Dontenwill's team through a KO of integrin a5 using the CRISPR / Cas9 technique from U87MG.

[0101] HEK-293T, U87MG and U87MG CRISPR a5 were grown in DMEM (Gibco™) while HCC1954 were grown in RPMI (Gibco™). DMEM and RPMI were supplemented with 10% fetal bovine serum (SVF) and 1% PenStrep (100 U / ml pennicillin / 100 mg / ml streptomycin) (Gibco™) for routine cultures.

[0102] Standard culture conditions were in an incubator at 37°C and 5% CO2 with a humid atmosphere maintained.

[0103] Production of lentiviral vector

[0104] To produce pseudotyped vectors, HEK-293T cells were cotransfected with 3 plasmids: (1) an envelope plasmid (pCMV-RP-1, -2, or -3) encoding one of the modified GPEs described previously, (2) p8.91 , and (3) the pSDY-nRFP-Puro plasmid. Transfection was performed using linear PEI (PolyEthylenelmine) (MW = 25000). Cell culture medium was replaced with fresh PenStrep-free medium 6h post-transfection. Viral supernatants were harvested 48h post-transfection and purified using sucrose pads (20%). A further step can be introduced at this level to avoid acidification of the pH of the medium, as described below. The viral particles produced were assayed using the Fujirebio p24 ELISA assay kit.Test transduction of the enveloppes

[0105] All transductions were carried out without polybrene by placing a constant quantity of pseudotyped viral particles (100ng of p24) in contact with 3.105cells of each cell type in 6-well plates. Each well contained 2mL of medium adapted to the cell type, supplemented with SVF and devoid of PenStrep. The pseudotyped vectors were left in contact with the cells for 8 h, after which the culture media were replaced with fresh medium. Fluorescent markers were observed at least 3 days post-transduction. Image capture and processing to determine transduction efficiency of pseudotyped viral vectors

[0106] After transduction the cells were fixed with absolute ethanol (Harlow and Lane, 2006) and then stained with DAPI (4',6-Diamidino-2-Phenylindole, Dilactate). Confocal fluorescence microscopy images were captured using a ZOE™ Fluorescent Cell Imager (BioRad®) in blue field (excitation: 355 / 40nm; emission: 433 / 36nm) and red field (excitation: 556 / 20nm; emission: 615 / 61 nm). The images were captured in TIFF format with a magnification of 175x and a field of view of 0.70mm2.

[0107] Cells were counted using FIJI imaging software (Schindelin et al., 2012 ([9])) complemented by the Stardist Cell Detection Plugin (Schmidt et al., 2018 (

[0010] )). No images were retouched for cell count processing. On the other hand, the contrast and saturation of some images could be improved so that cells are clearly rendered visible to the human eye and can be presented in this manuscript. Transduction efficiency calculations were carried out by analysing duos of images, one blue + one red, captured in 5 different fields of view per replica. The blue field of view was used to determine the total number of cells present in the image, using DAPI staining, while the red field of view was used to determine the number of cells transduced by expression of the transgene encoding nRFP. The two-way ANOVA statistical analyses presented were performed using GraphPad Prism software (Version 9.5.1 for MacOS).

[0108] Blocking transduction by adding inhibitors of a5B1RGD- and RP2-pseudotyped vectors were generated as described previously. The vectors were used in equal amounts (100 ng p24) to transduce 3,105 U87MG cells in the presence of 10pM of two a5 1 -specific inhibitors: K34c (Cat. No.: HY-150124) and FR248 (disclosed in Heckmann et al., 2008 (

[0014] , as molecule 44b). The cells were left in contact with the viruses and inhibitors for 8 hours. The results were analysed by imaging as described above.

[0109] Creation of U87MG cell lines labelled with nGFP and nRFP

[0110] In order to set up co-cultures and to have a visual marker for cell identification, two U87MG cell lines expressing different nuclear fluorescent proteins were created. One expressing nRFP and the other nGFP. These two lines were constructed by transduction using VSV-G pseudotyped lentiviral vectors. They carried either a pSDY-nRFP-Puro or a pSDY-nGFP-Puro plasmid. 3.105U87MG cells were transduced and selected 24h posttransduction with 1pg / ml puromycin added to the culture medium for 7 days. Cells were isolated by limiting dilution to obtain two monoclonal lines: U87-nGFP and U87-nRFP.

[0111] Transduction and apoptosis induction assay on U87MG

[0112] U87-nGFP cells were transduced using an RP2 pseudotyped vector produced following the lentiviral vector production protocol previously described. For this experiment the pSDY used did not carry a nuclear fluorescent marker but did allow expression of iCasp9. Transduced U87-nGFP cells were mixed 1:1 with non-transduced U87-nRFP cells at 24 h post-transduction. At 6 h post-culture, the media were replaced with fresh medium supplemented with 1pM AP1903, the iCasp9 dimerisation inducer (CAS number : 195514-63-7), or as described in Xiaoou Zhou et al. (2015) (

[0013] ).

[0113] Image capture and processing for the apoptosis induction assay

[0114] Two images of the same field, one in red field (excitation: 556 / 20nm; emission: 615 / 61 nm), the other in green field (excitation: 480 / 17nm; emission: 517 / 23nm) were captured using the ZOE™ Fluorescent Cell Imager at to, t24h, t48h and t120h. The relative proportion of U87-nGFP andU87-nRFP was measured by counting cells using FIJI imaging software (Schindelin et al., 2012 ([9])) complemented with the Stardist Cell Detection Plugin (Schmidt et al., 2018 (

[0010] )). The data presented were processed using GraphPad Prism software (Version 9.5.1 for MacOS), and are presented as the mean of 3 replicates ± standard deviation (SD).

[0115] Western Blot

[0116] Cells were harvested after detachment without trypsin. The cells were lysed in a solution of Laemmli's reagent and the proteins were then deposited on a pre-cast acrylamide denaturing gel from BioRad. After migration, the proteins are transferred to a PVDF (Polyvinylidene fluoride, GE Healthcare) nitrocellulose membrane. The non-specific binding sites are then blocked for 1h at room temperature in a solution of 0.1% PBS-Tween and 5% milk, then the membranes are incubated overnight with different primary antibodies (1:1000) directed against the proteins of interest, a5, [31, aV and [33. After three washes in 0.1% PBS-Tween, a secondary antibody (1:10000) coupled to the enzyme HRP (‘horseradish peroxidase’) is then applied 30 minutes at room temperature to the membranes. After three washes, the proteins were visualised using BioRad's ELC (Enhanced Chemiluminescence) technique. The intensity of the bands was assessed using Fiji software.

[0117] Results

[0118] Mono-functionalisation of m 168 for targeting a5B1

[0119] To date, virus functionalisations have largely turned to RGD and not to PHSRN. To test whether a GPE functionalization could be of interest compared with RGD, we inserted either the RGD or PHSRN peptide into the 71-74 region of the m168 variant, as shown in Figure 2 and detailed in Materials and Methods.

[0120] Transduction of U87MG cells (which strongly express a5[31) with LV vectors pseudotyped with these EGPs showed comparable transduction for the variant carrying the PHSRN motif and the m168 protein was devoid offunctional motifs. In contrast, the variant carrying the RGD motif gave an approximately twofold higher level of transduction.

[0121] Structural analysis of flexible loops at the apex of E2

[0122] In order to determine whether the PHSRN motif allows an increase in efficiency when used in combination with RGD compared to a EGP using RGD alone, it was necessary to identify other regions on the surface of E2 of SinV allowing insertion of a second peptide outside the 71-74 region. On the natural ligand of a5 1, fibronectine, RGD and PHSRN are separated by approximately 30A. To identify the flexible loops most likely to tolerate modifications, we studied the PDB:6IMM structure of the EGPs of SinV. We represented the structure by highlighting the b-factor. These factors can be seen as indicating the relative vibrational motion of the different parts of the structure. Atoms with a low b-factor belong to a part of the structure that is well ordered and not very mobile. Atoms with high b-factors generally belong to a more flexible part of the structure, indicating a more disordered structure.

[0123] As expected, the 71-74 region, already identified as the region of interaction between E2 and its receptor and described as tolerant to exogenous peptide insertions, shows a high b-factor compared with the rest of the structure.

[0124] Loop 209-216 is correctly exposed and has a high b-factor compared with the rest of the structure. It has a relatively short sequence of 7 amino acids. This site was selected and named flexible loop 1, or region 1 (R1). This loop is spaced approximately 25 A from the 71-74 region, which is close to the distance between PHSRN and RGD observed on fibronectine.

[0125] The 166-200 loop has no apparent b-factor. This region is too flexible for its structure to be determined by cryo-EM. As this loop is relatively large, two ways of inserting an exogenous peptide into it will be evaluated. Insertion will be achieved using a total deletion of the loop (region 2) or a deletion of the centre of the loop only (region 3). Region 3 is therefore a sub-part of region 2.To ensure that the loops identified as flexible really were, a second structure obtained by a different method, X-ray diffraction crystallography, was also studied. In this second structure (PDB: 3MUU) region 1 is also one of the loops with the highest b-factor at the apex of E2 and regions 2 and 3 could not be could not be represented because they were too flexible. The two methods therefore show that the regions identified are very flexible and may be candidates for the insertion of exogenous peptides.

[0126] Dual functionalisation of E2 for targeting a5B1 + cells

[0127] Once the various candidate regions had been identified, we chose to insert PHSRN into the 71-74 region and RGD into the three candidate regions identified, resulting in the constructs detailed in Figure 4. The constructs were designed to maintain an orientation close to the configuration found in fibronectin.

[0128] Interestingly, one protein, in particular the one carrying the insertion of the RGD motif in the second region (RP2 protein) showed a significantly higher capacity to transduce U87MG cells than the protein lacking integrins-specific interaction motifs (m168). The percentage increase in transduction is around 7-fold. This increase in functionality compared with m168 is therefore much greater and more significant than the results obtained with the protein mono-functionalized by RGD and the protein carrying the PHSRN motif in the 71-74 region but no insertion in region 2. In contrast, the other two variants, RP1 and RP3, show rates comparable to or even significantly lower than m168 (with undetectable levels for RP3).

[0129] In order to compare the degree of specificity conferred by dual RGD-PHSRN-functionalization versus single RGD-functionalization, we evaluated the percentage of cells transduced on cell populations with different integrin expression profiles, focusing particularly on a5 1. In addition to the U87MGs, used for the experiments described previously, we used the U87 CRISPR a5 KOs for integrin a5 (see Materials and Methods) and the HCC1954 a5pi -negative ductal carcinoma tumour line. Western blot characterization of the expression of a5 1 and aV 3, the two most studied intergins with RGD, in these cells shows (Figure 6 that U87MGs express a5in large quantities compared with the other two cell types. In contrast, 1 is expressed by all 3 cell types (Figure 6). But surprisingly, CRISPR a5 U87MGs express between 1.5 and 2 times more 01 than wild-type U87MGs. This indicates that other integrins with RGD dimers involving 01 can be formed in large numbers on the cell, in response to a decrease in a5 molecules. HCC1954 show a slightly lower level of 01 expression than U87MG. 01 is involved in the formation of a very large number of integrin dimers. In fact, many cells express it, so it is not surprising to find this protein in all the cell types used here. However, a5 is the limiting factor in the formation of the a501 dimer and its very low relative expression by U87MG CRISPR a5 and HCC1954 means that these cells can be classified as a501-. HCC1954 cells express a high level of aV compared with U87MG cells, which, together with the expression of 01, defines these cells as aV01+. 03 is expressed only by wild-type U87MG and CRISPR a5, and the expression profile of aV and 03 is similar between U87MG and CRISPR a5 cells. What emerges from this characterization is that, with aV01 expression only, HCC1954 constitute the most restricted RGD integrins spectrum of the three cell types and that, in addition to the expected absence of a5 expression by U87MG CRISPR cells there is an unexpected increase in 01 expression compared to wild-type cells.

[0130] Under our experimental conditions, use of the RP2 variant enabled transduction of around 40% of U87MG cells exposed to the vector. This rate is approximately 4 times higher than that obtained using the variant functionalized by an RGD motif (this time localized in the 71-74 loop, whereas in RP2 it is inserted in the region 2). We have compared a variant carrying only the RGD motif in R2 to the variant carrying only the PHSRN motif in the 71-74 loop to assess the contribution of the additional presence of RGD in relation to m168-PHSRN (Figure 3). From this point of view, adding the RGD motif to E2 increases transduction by more than 15 times.

[0131] In addition to efficiency, transduction specificity is extremely important and, in our case, transduction must be restricted as strictly as possible to a501+ cells. To assess this aspect, we need to compare the transductionrate of U87MG cells with that of cells not expressing a5 1. Classically, this comparison is made using cells that are different from U87MG cells and therefore have a different genetic background to the latter. When we made this comparison using HCC1954 as a5 1- cells, we found a 7.5-fold specificity. In relation to numerous studies described in the literature, here too we can compare the transduction of U87MG with that of the U87MG CRISPR a5 line, which has the same genetic background apart from the expression of a5.

[0132] This makes it possible to establish the specific contribution of a5 to transduction. In these cells, the transduction rate was reduced by about 2-fold compared with wild-type cells, with a specificity of about 2-fold. In sharp contrast, the presence of RGD alone resulted in transduction rates of around 10% for U87MG and HCC1954 cells, which increased by a factor of 2 for U87MG CRISPR cells. This unexpected increase may be due to the high level of pi expressed by these cells, which may lead to greater presentation of integrin with RGD on the cell surface.

[0133] With a view to application in gene therapy, we want to improve both efficiency (number of cells transduced) and specificity (proportion of a5 1 + cells transduced compared with a5 1- cells). To take account of these two factors combined, we extracted a value, called specificiency (Sp), for each GPE variant, taking into account a pair of transduced cells. To determine this value we followed the following rational: if GPE variant W transduced a % of X cells (which are the desired target) and b % of Y cells, which we do not wish to transduce, its specificity value for X vs Y cells (Sp-Wx / y) would be equal to a(a / b). In order to minimise variations due to imprecision of estimates for infrequent events, if the transduction values of non-target cells are low (<5%) and homogeneous between the different variants tested (verified by one-way ANOVA test) with a mean of the standard deviations greater than the mean of the values, the mean of these values, bMoy, is taken into account instead of the individual values for each variant. If the Z envelope protein variants transduce X and Y cells (under the same experimental conditions) in percentages c and d, its Sp will be Sp-ZX / Y=c(c / d). It is therefore now possible to calculate a preference index for one variant over another and for one cell type over another by dividing the Sp calculated for each MCP on identical cell types. So if we want to estimate the preference factor (PF) of variant W over Z for cells X over cells Y, this will be given by Sp-Wx / y / Sp-Zx / Y. Any PF value greater than 1 will indicate that W is more specific than Z, and the value will indicate by how many times. If the values are less than 1 , the specificity of variant Z is better than that of variant W. If we now apply these calculations to the RP2 variant in relation to RGD for U87MG vs HCC1954 cells, we obtain specificity values of 14.23 for the variant carrying RGD and 214.79 for RP2. This gives us a preference factor of 15.09 in favour of the RP2 variant (see Table 1 below). Specificiency values calculated and values used for the calculation for vectors armed with RP2 bi-functionalised or RGD mono-functionalised pseudotypes.

[0134] Table 1

[0135]

[0136] Effect of a5P1 -specific inhibition on RP2 transduction

[0137] In order to ensure that transduction occurs through an interaction involving a5 1, we transduced U87MG cells in the presence or absence of a5pi -specific inhibitors. Although this hypothesis was supported by the decrease in transduction observed when wild-type and CRISPR a5 U87MG were compared, we could not rule out a possible difference in the behaviour of the cells during transduction linked to the CRISPR / Cas9-induced KO of a5 for reasons other than the absence of a5. U87MG cells were therefore transduced in the absence or presence of 10pM K34c + 10pM Fr248 (Martinkova et al., 2010 (

[0011] )). Both molecules are specific inhibitors of a5pi. Two cases are expected here.In the absence of an inhibitor, transduction by an m168-RGD-pseudotyped vector will be able to exploit a5 1 but also any other RGD integrin dimer present on the cell, such as aV 3 present on U87MG. An RP2-pseudotyped vector, on the other hand, will only be able to transduce the cell via a5 1, the dimer for which it is specific thanks to the dual functionalisation. In the presence of an inhibitor, transduction by a vector carrying m168-RGD will therefore always be possible, whereas transduction by an m168-RP2 vector will not. The results of three independent experiments are presented in Figure 27 and clearly show that the addition of inhibitors exclusively affects transduction by RP2 pseudotyped vectors and not by RGD.

[0138] The magnitude of the decrease in transduction of U87MGs by soluble inhibitors (2.6-fold) is of the same order as that observed using CRISPR a5 U87MG cells (1.8). The RP2 pseudotype therefore possesses a significant capacity to specifically target the a5 1 dimer due to the presence of the PHSRN motif in the GPE.

[0139] Interest in targeting two distinct regions for a5B1

[0140] To study the effect of inserting two identical ligands in place of the RGD / PHSRN pair, we constructed the RGD2x variant (shown in Figure 9), which carries an RGD motif at position 71-74 and one at position R2. Transduction of U87MGs by vectors pseudotyped with these RGD2x EGPs showed values comparable to those observed with the variant carrying a single RGD. The specificiency of RGD2x using the same cells as above, U87MG vs HCC1954 is 16.45 whereas for RP2 it is 214.79 (Figure 10). The preference factor in this case for the use of RP2 is therefore 13.05. Importantly, the transduction levels of HCC1954 are comparable with both variants, as would be expected from transduction that does not rely on the use of a5pi to mediate entry. Specificity values calculated and values used for the calculation for vectors armed with RP2 or RGD bi-functionalised pseudotypes are shown in Table 2.

[0141] Table 2

[0142]

[0143] Elimination of iCasp9-induced GBM a561 + cells

[0144] The development of lentiviral vectors for gene therapy is ultimately aimed at the targeted delivery of suicide genes to the cancer cells to be eliminated. We therefore tested the ability of our RP2 pseudotyped lentiviral vector to allow depletion of a cell population exposed to this vector carrying the iCasp9 suicide gene in U87MG cells.

[0145] U87MG-nGFP cells (a U87MG line that we have modified to allow expression of the nuclear version of GFP, see materials and methods), were transduced with an LV vector armed with RP2 GPEs and carrying a transgene encoding iCasp9. No selection was applied after transduction of nGFP cells with the RP2 vector carrying iCasp9 to determine whether the vector was efficient enough to allow sufficient transduction, without counterselection of non-transduced cells, to observe a significant decrease in the cell population. After exposure to this vector, this cell population was mixed 50%-50% with a population of U87MG that we had generated expressing the nuclear version of RFP. This second population, not exposed to the RP2 vector carrying iCasp9, was not transduced and does not express iCasp9. Six hours after mixing, the cells were treated with AP1903, inducing dimerisation, and therefore activation, of caspase 9. As early as 24 h after treatment with AP1903, the green cell population decreased by more than 30% (Figure 11), a trend that was confirmed 96 h after treatment with a decrease in the nGFP+ population of around 50%. This specific decrease in nGFP cells was observed exclusively in the population treated with AP1903.

[0146] These results show for the first time that the gene encoding iCasp9 can therefore be used as a suicide gene for the elimination of GBM cells such as U87MG and that the doses of AP1903 inducer used do not inducesignificant toxicity in cells that are not sensitised to the inducer because they are not transduced.

[0147] Example 2: Treatment of glioblastomas with RP2 protein in an orthotopic model of mice

[0148] To evaluate the treatment of glioblastomas in the context of the orthotopic model of a5 1+ cells, 15 mice (Janvier Labs), were inoculated intracranially with 3x105U87MG cells.

[0149] First of all, tumour growth was monitored by positron emission tomography (PET) tomodensitometry 10 days after inoculation of glioblastoma U87MG cells. Out of the 15 mice 14 showed growth of tumour in the site of injection of the U87MG cells.

[0150] Five and 9 of the PET-positive mice were then injected with either PBS or PBS-containing 106particles of RP2 armed SinV oncolytic virus, respectively, on the next day (injection 1) and 7 days (injection 2) after. Five days after the injection 2, the size of the tumour was measured by PET.

[0151] As shown in Table 3, mice injected with PBS presented, on average, an increase of the cancer mass with respect to the mass detected at the moment of injection 1 of nearly 4-fold (3.84 ±1.86 fold). In addition, one mouse died between injections 1 and 2 for reason not determined. The cohort of mice treated with RP2 armed SinV oncolytic virus, displayed an average increase of only 44% (1.44 ± 0.55) of the size of the cancers, compared to the day before injection 1. The difference between the two cohorts (not taking into account the deceased mouse) was significant (p=0.0035), indicating a control of the expansion of the cancers of U87MG cells int the brain by injection of RP2 armed SinV oncolytic virus intravenously (in tail of the mice).

[0152] Table 3

[0153]

[0154]

[0155] Reference List

[0156] Feng, Y., and Mrksich, M. (2004). The synergy peptide PHSRN and the adhesion peptide RGD mediate cell adhesion through a common mechanism. Biochemistry 43, 15811-15821.

[0157] Morizono, K., Xie, Y., Ringpis, G.E., Johnson, M., Nassanian, H., Lee, B., Wu, L., and Chen, I.S.Y. (2005). Lentiviral vector retargeting to P-glycoprotein on metastatic melanoma through intravenous injection. Nat. Med. 11, 346-352.

[0158] Temple F Smith, Michael S Waterman, Comparison of biosequences, Advances in Applied Mathematics, Volume 2, Issue 4, 1981, Pages 482-489.

[0159] Saul B. Needleman, Christian D. Wunsch, A general method applicable to the search for similarities in the amino acid sequence of two proteins, Journal of Molecular Biology, Volume 48, Issue 3, 1970, Pages 443-453.

[0160] Pearson WR, Lipman DJ. Improved tools for biological sequence comparison. Proc Natl Acad Sci U S A. 1988 Apr;85(8):2444-8. Stephen F. Altschul, Warren Gish, Webb Miller, Eugene W. Myers, David J. Lipman, Basic local alignment search tool, Journal of Molecular Biology, Volume 215, Issue 3, 1990, Pages 403-410. Zufferey, R., Nagy, D., Mandel, R.J., Naldini, L., and Trono, D. (1997). Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat. Biotechnol. 75, 871-875.

[0161] Rossolillo, P., Winter, F., Simon-Loriere, E., Gallois-Montbrun, S., and Negroni, M. (2012). Retrovolution: HIV-Driven Evolution of Cellular Genes and Improvement of Anticancer Drug Activation. PLoS Genet.

[0162] 8, e1002904.

[0163] Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., etal. (2012). Fiji: An open-source platform for biological-image analysis. Nat. Methods 9, 676-682.Schmidt, U., Weigert, M., Broaddus, C., and Myers, G. (2018). Cell detection with star-convex polygons. In Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), (Springer Verlag), pp. 265-273.

[0164] Martinkova, E., Maglott, A., Leger, D.Y., Bonnet, D., Stiborova, M., Takeda, K., Martin, S., and Dontenwill, M. (2010). a5 1 integrin antagonists reduce chemotherapy-induced premature senescence and facilitate apoptosis in human glioblastoma cells. Int. J. Cancer 127, 1240-1248.

[0165] Duverge, A., & Negroni, M. (2020). Pseudotyping Lentiviral Vectors: When the Clothes Make the Virus. Viruses, 12(11 ), 1311. https: / / doi.org / 10.3390 / v12111311.

[0166] Xiaoou Zhou, Gianpietro Dotti, Robert A. Krance, Caridad A. Martinez, Swati Naik, Rammurti T. Kamble, April G. Durett, Olga Dakhova, Barbara Savoldo, Antonio Di Stasi, David M. Spencer, 2 Yu-Feng Lin, Hao Liu, Bambi J. Grilley, Adrian P. Gee, Cliona M. Rooney, Helen E. Heslop, and Malcolm K. Brenner. Inducible caspase-9 suicide gene controls adverse effects from alloreplete T cells after haploidentical stem cell transplantation. Blood (2015) 125 (26): 4103-4113.

[0167] Dominik Heckmann, Axel Meyer, Burkhardt Laufer, Grit Zahn, Roland Stragies, and Horst Kessler. Rational Design of Highly Active and Selective Ligands for the a5 1 Integrin Receptor. ChemBioChem 2008, 9, 1397 - 1407.

Claims

CLAIMS1. Recombinant protein having at least 90% of identity with the protein of sequence SEQ ID NO: 1 , said percentage of identity not including gaps due to possible insertions and / or deletions, characterized in that said protein comprises at least one substitution between amino acids 213 and 234, and / or at least one substitution between amino acids 70 and 78.

2. Recombinant protein according to claim 1, wherein said at least one substitution between amino acids 213 and 234 comprises a RGD tripeptide.

3. Recombinant protein according to claim 1 or 2, wherein said at least one substitution between amino acids 70 and 78 comprises a PHSRN peptide (SEQ ID NO: 2).

4. Recombinant protein according to any one of the preceding claims, said protein comprising a RGD tripeptide between amino acids 213 and 234 and a PHSRN peptide between amino acids 70 and 78.

5. Nucleic acid encoding a recombinant protein as defined in any one of the preceding claims.

6. Expression vector into which a nucleic acid according to claim 5 is inserted.

7. Cell transfected with an expression vector as defined in claim 6.

8. Virus expressing a recombinant protein as defined in any one of claims 1 to 4.

9. Virus according to claim 8, which is selected among retroviruses, adenoviruses, and alphaviruses.

10. Virus according to claim 8 or 9, which is an oncolytic virus.

11. Virus according to any one of claims 8 to 10, which is a strain of Sindbis virus.

12. Virus as defined in any one of claims 8 to 11 , for use as a medicament.

13. Virus as defined in any one of claims 8 to 11 , for use in the treatment of a cancer, especially a5 1 positive cancers.

14. Virus for use according to claim 13, wherein said cancer is selected in the group comprising glioblastoma, head and neck cancer, colon cancer, melanoma, non-small cell lung cancer (NSCLC).

15. Virus for use according to claim 13 or 14, wherein said virus is administered systemically or locally.

16. A method of in vitro neutralisation of at least one cancer cell expressing integrin a5 1 at its plasma membrane, by administration of a virus as defined in 8 to 11.

17. A method for producing a virus as defined in any one of claims 8 to 11, comprising the steps of:a) expressing an expression vector as defined in claim 6 in a producer cell, thereby obtaining the virus in a production medium,b) keeping the medium of viral production non acidic through the addidion of a base or buffer in the latter.

Images