FR3036118B1 - RETROVIRAL PARTICLE COMPRISING AT LEAST TWO ENCAPSID NON-VIRAL RNA - Google Patents

Abstract

The present invention relates to a retroviral system for transferring non-viral RNAs into target cells and more particularly to a retroviral particle capable of delivering multiple RNAs. More particularly, it relates to retroviral particles comprising a protein derived from the Gag polyprotein, an envelope protein, optionally an integrase and at least two encapsidated non-viral RNAs, the encapsidated non-viral RNAs each having a dna RNA sequence. interest related to an encapsidation sequence, each encapsidation sequence being recognized by a binding domain introduced into the protein derived from the Gag polyprotein and / or in the integrase.

Description

The present invention relates to a retroviral system for transferring non-viral RNAs into target cells and more particularly to a retroviral particle capable of delivering multiple RNAs. The invention also provides compositions comprising these retroviral particles, kits for their production, related methods of manufacture, and uses of the particles and compositions.

Introducing multiple RNAs into a target cell is a major issue in research and development as in gene therapy. The use of virus-derived vectors has become a crucial method of gene transfer. Viral vectors are now divided into two main categories: integrative vectors, which integrate into the recipient genome, and non-integrative vectors, which usually form an extra-chromosomal genetic element.

Integrative vectors such as gamma-retroviral (RV) and lentiviral (LV) vectors are stably integrated. Non-integrative vectors, such as adenoviral vectors (ADV) and adeno-associated vectors (AAV) are rapidly cleared from rapidly dividing cells. Some factors influencing the choice of a particular vector include its packaging capacity, its range of target or host cells, its gene expression profile, its transduction efficiency and its ability to induce an immune response, which is particularly problematic if repeated transductions are needed.

Some applications require the use of non-integrative vectors such as in gene therapy or in many in vitro, ex vivo and in vivo applications. By way of examples, mention may be made of: • the induction of reprogramming of cells specialized in pluripotent cells, as well as the induction of the differentiation of stem or pluripotent cells into specialized cells, • the expression of antigens or proteins (toxic or not) simultaneously in a target cell, • the expression of genome engineering systems such as CRE, TALEN or CRISPR, or any other system requiring protein or RNA expression .

One way to introduce RNA into a target cell is to use viral vectors based on viruses belonging to the family Retroviridae (also referred to as the family of retroviruses or RNA viruses). Indeed, during its replication cycle, a virus of the family Retroviridae has the ability to convert its genome, consisting of RNA, into double-stranded DNA that will be integrated into the genome of the target cell. Specific examples of this family of retroviruses are gamma-retro viruses and lentiviruses.

The replicative cycle of a retrovirus comprises a first phase of recognition of the target (or host) cell via attachment to a membrane receptor. This recognition phase results, after membrane fusion, in the entry of the retrovirus into the host cell. The retro viral RNA is then copied to double-stranded DNA by reverse transcriptase, encoded by the retrovirus, and then integrated into the genome of the host cell. This viral genome is then transcribed by the host cell like all other genes in the cell. This genome encodes all the proteins and sequences allowing the manufacture of other viruses.

More particularly, three genes are common to all retroviruses: gag, pol and env.

Gag is a gene encoding a polyprotein, the proteins derived from this polyprotein by cleavage being structural proteins involved in the assembly of viruses during replication. These structural proteins are more specifically the matrix protein (MA), the capsid protein (CA) and the nucleocapsid protein (NC).

Pol is a gene encoding enzymes integrase, reverse transcriptase and protease.

Env is a gene encoding envelope glycoproteins.

These three genes thus make it possible to copy the retroviral RNA into double-stranded DNA, to integrate this DNA into the genome of the host cell and then to generate the structure of the neo-synthesized retro-viruses: envelope proteins, capsid proteins and nucleocapsid proteins. However, in order for the neo-synthesized retroviruses to be complete, it is necessary to encapsidate in each of these structures two copies of the retroviral RNA. This encapsidation of two copies of the retroviral RNA in the structure is carried out by the recognition by the nucleocapsid protein of a packaging sequence called Psi (for "Packaging Signal") carried by the copy of the retroviral RNA. .

When a viral vector derived from a retrovirus is used for gene therapy purposes, at least a portion of the coding regions of gag, pol and / or env is dissociated between the packaging system and the expression system of the sequence of interest. The coding sequences gag and pol, for example, are carried by the encapsidation plasmid bringing in trans the proteins necessary for the production of viral vectors. The encapsidation sequence as the sequence of interest are carried by independent systems, in order to make the retrovirus non-replicative.

However, in some cases, the integration of the RNA sequence of interest into the genome of the host cell in a random manner can interrupt an open reading phase and block the expression of important genes. In addition, for certain applications, such as cell reprogramming or stem cell differentiation, it is recommended that the transient expression of a sequence of interest be transiently achieved.

WO2007 / 072056 discloses a viral vectorization system comprising env, gag, optionally gaglpol and an RNA sequence containing a heterologous packaging signal which is recognized by a corresponding RNA binding domain associated with gag or gaglpol. . This system is described in the application as being non-integrative and allowing transient expression.

However, the effectiveness of such systems remains limited. In particular, for a target cell to express an RNA of interest conveyed by these systems, it is generally necessary to introduce into the cell several copies of this RNA of interest and consequently to use high multiplicities of infection ("multiplicity of infection" or MOI). The MOI is the ratio of the number of vectorization systems introduced to the number of cells to be infected. A strong MOI makes it possible to introduce into the cells several copies of the RNA of interest by allowing the same cell to undergo several infections. However, if it makes it possible to improve the level of expression of the RNA of interest conveyed, the use of strong MOI, because of the multiple infections undergone by the cell, also generates a certain toxicity.

There is therefore still a need for more efficient and less toxic viral vectorization systems.

The work of the inventors has made it possible to produce a vectorization system capable of delivering several RNAs of interest into the same cell in a single infection.

The present invention thus relates to a retroviral particle comprising a protein derived from the Gag polyprotein, an envelope protein, optionally an integrase and at least two encapsidated non-viral RNAs, the encapsidated non-viral RNAs each comprising a sequence of RNA of interest linked to an encapsidation sequence, each encapsidation sequence being recognized by a binding domain introduced into the protein derived from the Gag polyprotein and / or into the integrase.

The retroviral particle according to the invention makes it possible to introduce at least two non-viral RNAs, preferably 3, into a cell by a single infection.

By "protein derived from the Gag polyprotein" is meant any protein resulting from the cleavage of the Gag polyprotein. More particularly, it is a nucleocapsid protein, a matrix protein (for example, in retroviral particles derived from MoMuLV murine virus) or gammaretrovirus-specific p12 protein.

By "envelope protein" is meant any envelope protein, including pseudotyping envelope protein. By way of example, mention may be made of the Ampho envelope protein, the ecotropic envelope protein, the Moloney virus envelope protein of murine leukemia (MoMuLV), the envelope protein of the immunodeficiency virus. Feline (FIV), Harvey murine sarcoma virus envelope protein (HaMuSV), murine mammary tumor virus (MuMTV) envelope protein, Rous sarcoma virus envelope protein ( RSV), the measles virus envelope protein (also MV for measles virus), the Gibbon leukemia virus envelope protein (GalV) or the vesicular stomatitis virus envelope protein (VSV) -G). The envelope protein can thus be modified to target certain cell types or certain applications (use of surface receptors as envelope protein).

It is also possible to modify the envelope protein with an antibody, a glycolipid and / or a particular ligand in order to target a particular receptor and / or cell type.

Preferably, the envelope protein is the VSV-G protein.

By "integrase" is meant the enzymatic protein encoded by the pol gene, which allows the integration of the retroviral DNA into the DNA of the retrovirus-infected cell during the replication of said retrovirus.

The term "encapsidation sequence" denotes a motif (sequence and three-dimensional structure) RNA specifically recognized by a binding domain. Preferably, the encapsidation sequence is a rod-loop motif. More preferably still, the encapsidation sequence of the retroviral particle is the stem-loop motif of the bacteriophage MS2 RNA such as, for example, that resulting from the sequence ctagaaaacatgaggatcacccatgtctgcag (SEQ ID No. 1). The stem-loop motif and more particularly the stem-loop motif of bacteriophage MS2 RNA can be used alone or repeated several times, preferably from 2 to 25 times, more preferably from 6 to 18 times.

By "binding domain" is meant all or part of a protein specifically binding to the encapsidation sequence linked to the RNA sequence of interest. More particularly, it is a mutant protein or not, defined by a three-dimensional structure, specifically binding to the encapsidation sequence. Preferably, the link domain is a heterologous domain. More preferably, the binding domain is the Coat protein of bacteriophage MS2, PP7 phage or β-phage, the nun protein of HK022 prophage, the U1A protein or the hPum protein.

More preferably, the binding domain is the Coat protein of bacteriophage MS2.

By "each sequence" is meant that the encapsidation sequences may be identical or different, depending on whether these encapsidation sequences are recognized by an identical or different binding domain. Indeed, the retroviral particle according to the invention may comprise one or more binding domains. When several binding domains are introduced, these can be introduced into the Gag polyprotein and / or into the integrase.

The binding domain not only allows the recognition of the encapsidation sequence but also the encapsidation of the RNAs carrying the encapsidation sequence in the particle (or, in this case, a non-viral RNA linked to a DNA sequence). encapsidation).

By "encapsidation" is meant the packaging of an RNA in the viral capsid of a viral particle. It will be noted that in the present invention, the packaging of the non-viral RNAs is carried out by the recognition of a non-viral packaging signal, in particular other than Psi.

By "sequence of interest" is meant a sequence encoding a function of interest to the user. More particularly, the sequence of interest carried by each of the two encapsidated non-viral RNAs may be identical or different. With the aid of the particles according to the invention, it is possible to transfer identical or different non-viral RNAs into target cells.

Since the packaged RNAs are non-viral, these RNAs do not display the recognition sites of the proteins encoded by the pot gene.

Indeed, these non-viral RNAs do not enter the early stages of the replicative cycle of retroviruses, namely: the copy of single-stranded viral RNA into double-stranded DNA by reverse transcriptase; The maturation of the double-stranded viral DNA by recognition of the LTR ends by the integrase and maturation of the cytoplasmic preintegration complex in nuclear pre-integration complex.

These viral proteins (reverse transcriptase, integrase), encoded by the pol gene are therefore optional in the particle and the pol gene can therefore be either present or partially or completely deleted.

It will be noted that the retroviral particles according to the invention comprise a genetic material that is both viral and non-viral: the gag gene, which can be viral or chimeric. More particularly, the gag gene is chimeric when the binding domain (s) is / are introduced into it. Optionally, the pol gene, which may be viral or chimeric. Since the pol gene codes for several enzymes, the sequences relating to these enzymes can be totally or partially deleted, present and non-functional, or present and functional. More particularly, the pol gene is chimeric when the binding domain (s) is / are introduced into the integrase. At least two non-viral RNAs, each non-viral RNA bearing a sequence of interest and an encapsidation sequence. More particularly, these non-viral RNAs are devoid of any viral sequence.

More specifically, the retroviral particles according to the invention make it possible to introduce into target cells RNAs capable of inducing: transfer of one or more endogenous or exogenous coding sequences of interest from the target cell,

Transfer of one or more non-coding RNAs as RNAs capable of inducing an effect on gene expression, for example by means of shRNAs, miRNAs, sgRNAs, cellular RNA transfer, RNA messenger type or others (miRNA ...), genomic replicons of RNA viruses (HCV, etc.) or complete genomes of RNA viruses, • simultaneous expression of endogenous coding or non-coding sequences, exogenous to the target cell, • participate in the modification of the genome of the target cell by genome engineering systems, eg the CRISPR system. Advantageously, the retroviral particle according to the invention comprises a nucleocapsid protein, an envelope protein, optionally an integrase and at least two encapsidated non-viral RNAs, the encapsidated non-viral RNAs each comprising an RNA sequence of interest. linked to a packaging sequence, each encapsidation sequence being recognized by a binding domain introduced into the core protein and / or integrase.

By "nucleocapsid protein" is meant the NC structural protein encoded by the gag gene. When the binding domain is introduced into the nucleocapsid protein, then this protein is a chimeric protein derived from a chimeric gag gene.

When the binding domain is introduced into the integrase, then the integrase is a chimeric protein derived from a chimeric pol gene.

By "chimeric protein" is meant a recombinant protein comprising several different fused protein sequences.

According to a first embodiment, in the retroviral particle according to the invention, the binding domain is introduced into the nucleocapsid protein and the at least two encapsidated non-viral RNAs are distinguished by their RNA sequence.

This first embodiment allows an early and transient expression of the RNAs of interest, without associated integration event in the genome of the target cells.

Indeed, during the contacting of a particle according to this first embodiment with a target cell, the membrane of the retroviral particle and that of the target cell will merge and allow the release of the content of the particle in the cell. target. The RNAs are then released into the cytoplasm and the cellular machinery translates these RNAs directly into protein (s), that is to say without additional steps such as reverse transcription, translocation into the nucleus or integration into the nucleus. genome of the target cell.

More particularly, the at least two non-viral RNAs have the same encapsidation sequence. In this case, the at least two encapsidated non-viral RNAs are distinguished by their RNA sequence of interest, that is to say that the RNA sequences of interest included in the two packaged non-viral RNAs are different. . By "different" is meant sequences of interest that are not identical or that have a difference that does not result from a spontaneous or unintentionally chosen mutation by the manipulator.

Alternatively, the at least two non-viral RNAs have two different encapsidation sequences. These at least two encapsidation sequences are then recognized by at least two different binding domains, at least one of these domains being introduced into the nucleocapsid protein. In this case, the at least two encapsidated non-viral RNAs may comprise identical or different sequences of interest.

It is possible to package at least two non-viral RNAs, preferably three non-viral RNAs.

According to a particular embodiment of the first embodiment, a second binding domain is introduced into the nucleocapsid protein of the retroviral particle according to the invention.

In this case, at least two encapsidated non-viral RNAs carry different encapsidation sequences, each encapsidation sequence corresponding respectively to the first and second binding domain introduced into the nucleocapsid protein.

More particularly, when three non-viral RNAs are encapsidated: at least two of the encapsidated non-viral RNAs have the same encapsidation sequence corresponding to the first binding domain and are distinguished only by their RNA sequence of interest; Packaged non-viral RNA can carry an identical or different encapsidation sequence. When the encapsidation sequence is different, it may correspond to a second binding domain introduced into the nucleocapsid protein. Other binding domains can also be introduced into the core protein.

In addition to the binding domain (s) introduced into the nucleocapsid protein, it is also possible to introduce a binding domain into the integrase. Integrase is then a chimeric protein derived from a chimeric pol gene.

Preferably, when the binding domain is introduced into the integrase, the sequence of the integrase is mutated at the C-terminal domain in order to insert the sequence of the binding domain. More preferably still, the sequence of the integrase is mutated at the level of the C-terminal domain so as to introduce that of the protein "Coat" bacteriophage MS2.

In this case, the at least two encapsidated non-viral RNAs may carry different encapsidation sequences, each encapsidation sequence corresponding to the binding domains introduced into the nucleocapsid protein and into the integrase, respectively.

More particularly, when three non-viral RNAs are encapsidated: at least two of the encapsidated non-viral RNAs have the same encapsidation sequence corresponding to the first binding domain and are distinguished only by their RNA sequence of interest; Packaged non-viral RNA carrying a different encapsidation sequence, corresponding to a second binding domain introduced into the integrase. Other binding domains can also be introduced into the integrase.

Advantageously, the retroviral particle according to the invention is a lentiviral particle.

Preferably, in such a lentiviral particle: the binding domain is the bacteriophage MS2 Coat protein; the non-viral RNA packaging sequence is a stem-loop sequence of MS2; the nucleocapsid protein is the protein. of nucleocapsid (NC) of HIV belonging to the chimeric Gag polyprotein, the NC sequence being mutated at the second finger of zinc to insert the sequence of the protein "Coat" bacteriophage MS2.

Advantageously: The envelope protein is the VSV-G protein encoding the envelope protein of vesicular stomatitis virus. The encapsidation sequence comprises from 2 to 25 repetitions of the stem-loop sequence of MS2, preferentially from 6 to 18 repetitions of the stem-loop sequence, more preferably still from 10 to 14, for example 12 repetitions. Preferably, the stem-loop sequence is as follows: ## STR1 ##

An example of a lentiviral particle according to the invention, called MS2RLP, is described in more detail in the examples which follow.

According to a second embodiment, the invention relates to a retroviral particle comprising a protein derived from the Gag polyprotein, preferentially a core protein, an envelope protein, an integrase and at least two packaged non-viral RNAs, the RNAs. packaged non-viral proteins each comprising an RNA sequence of interest linked to a packaging sequence, each encapsidation sequence being recognized by a binding domain introduced into the integrase and, optionally, by a binding domain introduced into a protein derived from the Gag polyprotein, preferentially a nucleocapsid protein.

This second embodiment allows a transient expression of the RNA of interest, without associated integration event in the genome of the target cells.

Preferably, in this second embodiment, the sequence of the integrase is mutated at the C-terminal domain in order to insert the sequence of the binding domain.

Advantageously, the binding domain is a heterologous domain. More particularly, the binding domain is the Coat protein of bacteriophage MS2, PP7 phage or β-phage, the nun protein of prophage HK022, the U1A protein or the hPum protein.

Preferably, the sequence of the integrase is mutated at the level of the C-terminal domain so as to introduce that of the "Coat" protein of bacteriophage MS2.

The at least two non-viral RNAs may or may not have the same encapsidation sequence.

Similarly, the at least two encapsidated non-viral RNAs may or may not have the same RNA sequence of interest.

Preferably, the at least two encapsidated non-viral RNAs are distinguished by their RNA sequence of interest, that is to say that the RNA sequences of interest included in the two packaged non-viral RNAs are different.

More particularly, the at least two non-viral RNAs have the same encapsidation sequence, which is recognized by the binding domain introduced into the integrase.

It is possible to package at least two non-viral RNAs, preferably three non-viral RNAs.

According to a particular embodiment of the second embodiment, a second binding domain is introduced into the integrase of the retroviral particle according to the invention.

In this case, at least two encapsidated non-viral RNAs carry different encapsidation sequences, each encapsidation sequence corresponding respectively to the first and second binding domain introduced into the integrase.

More particularly, when three non-viral RNAs are encapsidated: at least two of the encapsidated non-viral RNAs have the same encapsidation sequence corresponding to the first binding domain, these two non-viral RNAs possibly being distinguishable by their DNA sequence. interest, the third packaged non-viral RNA may carry an identical or different encapsidation sequence. When the encapsidation sequence is different, it may correspond to a second binding domain introduced into the integrase. Other binding domains can also be introduced into the integrase.

In addition to the binding domain (s) introduced into the integrase, it is also possible to introduce a binding domain into the nucleocapsid protein.

The nucleocapsid protein is then a chimeric protein derived from a chimeric gag gene.

In this case, the at least two encapsidated non-viral RNAs may carry different encapsidation sequences, each encapsidation sequence corresponding to the binding domains introduced into the integrase and the nucleocapsid protein, respectively.

More particularly, when three non-viral RNAs are encapsidated: at least two of the non-viral encapsidated RNAs have the same encapsidation sequence corresponding to the first binding domain introduced into the integrase, these non-viral RNAs possibly being distinguishable by their sequence of RNA of interest, the third packaged non-viral RNA carries a different encapsidation sequence, corresponding to a second binding domain introduced into the nucleocapsid protein. Other binding domains can also be introduced into the core protein.

Advantageously, the retroviral particle according to the invention is a lentiviral particle.

When the retro viral particle is a lentiviral particle, it is possible to express transiently RNAs of interest, without associated integration event in the genome of the target cells, in particular quiescent cells.

Indeed, apart from its role in the integration reaction itself, the integrase (IN) participates in different steps of the retrovirus replicative cycle such as the morphogenesis of the viral particle, the reverse transcription and the nuclear import of the virus. preintegration complex (PIC).

More particularly, in lentiviruses, the integrase contains nuclear localization sequences (NLS) allowing its localization in the nucleus by the PIC. Therefore, when the encapsidation of the non-viral RNAs is carried out by a binding domain carried by an integrase of a lentivirus, the packaged non-viral RNAs will be transported into the nucleus of the target cell. Indeed, when placing a lentiviral particle in contact with this second embodiment with a target cell, the membrane of the particle and that of the target cell will fuse and allow the release of the capsid content in the cell. target. The RNAs are then supported by the integrase which, via the PICs, will allow the import of RNA into the nucleus. This management is particularly interesting for certain applications, such as expression in quiescent cells. In the case of retroviral particles, other than lentiviruses, integrase does not contain these NLS and is therefore localized in the cytoplasm. It is however possible to add in this type of integrase, the NLS sequences in order to induce a nuclear localization of the integrase, and thus of the RNAs supported by this integrase.

This support is also particularly useful for a CRISPR system, which uses RNA guides that hybridize specifically to the genome of the target cell. Once hybridized, these RNA guides guide an endonuclease (Cas9), which will allow the modification of a specific locus of the genome of the target cell.

More particularly, in such a lentiviral particle: the binding domain is the bacteriophage MS2 "Coat"protein; the non-viral RNA packaging sequence is a stem-loop sequence of MS2; the integrase is a protein; a chimeric enzyme whose sequence is mutated at the C-terminal domain in order to insert the sequence of the bacteriophage MS2 coat protein.

Indeed, the integrase (IN) is composed of 3 distinct functional domains, each one indispensable to ensure a complete integration reaction. The N-terminal domain contains a zinc finger motif which stabilizes the repl folded structure and increases the catalytic activity of the enzyme. The central domain of ΙΊΝ contains the DDE amino acid motif to which the catalytic activity of the enzyme is attributed. This core domain is also involved in the recognition of the nucleotide sequence conserved at each end of the retroviral DNA. The C-terminal domain is the least conserved among the retrovirus family. It possesses DNA binding activity and is indispensable for maturation reactions of the 3 'strand transfer ends. Apart from its role in the integration reaction itself, IN participates in different steps of the replicative cycle of retroviruses such as morphogenesis of the viral particle, reverse transcription and nuclear import of the preintegration complex.

As described by Petit et al. (1999, J. Virol, P5079-5088), the insertion of an exogenous C-terminal sequence of ΙΊΝ does not disrupt the steps of production and transduction of target cells while the same insertion into N-terminal does not not allow the detection of a transduction event.

The retroviral particle was therefore modified to contain the bacteriophage MS2 "Coat" protein in fusion with the integrase protein (see Figure I). The encapsidation plasmid p8.74, carrying the pol gene coding for the integrase protein, is modified in order to insert the sequence coding for the Coat protein into the C-terminal of the integrase by assembly PCR. The plasmid p8.74 is linearized by PCR then the Coat sequence, previously amplified by PCR, is cloned at the C-terminal level of the integrase, either directly end-to-end or with the addition of a linker.

Advantageously: The envelope protein is the VSV-G protein encoding the envelope protein of vesicular stomatitis virus. The encapsidation sequence comprises from 2 to 25 repetitions of the stem-loop sequence of MS2, preferentially from 6 to 18 repetitions of the stem-loop sequence, more preferably still from 10 to 14, for example 12 repetitions. Preferably, the stem-loop sequence is as follows: ctagaaaacatgaggatcacccatgtctgcag (SEQ ID No. 1). The invention also relates to compositions comprising particles according to the invention.

More particularly, the compositions according to the invention are concentrated compositions. Advantageously, the compositions are also purified. These compositions may be concentrated and purified according to the process described in Application WO2013 / 014537.

Typically, the compositions according to the invention comprise less than 30% of DNA contaminants and less than 45% of protein contaminants relative to the crude supernatant. More particularly, the compositions according to the invention comprise less than 30% of DNA contaminants and less than 2% of protein contaminants relative to the crude supernatant.

The term "crude supernatant" is intended to mean the supernatant of cell culture (s), comprising retroviral particles according to the invention, after clarification. Such a clarification step is more particularly described hereinafter in the processes for producing the particles according to the invention. When the recovery of the supernatant is carried out in several times, the crude supernatant then corresponds to all the supernatants collected, pooled together and then clarified.

Optionally, the compositions according to the invention comprise less than 1% of DNA contaminants and less than 1% of protein contaminants relative to the crude supernatant. The invention also relates to particle production kits according to the invention and to the methods of manufacturing these kits.

More particularly, the invention relates to a kit for producing particles according to the invention, comprising: (i) an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted a encapsidation sequence, (ii) a packaging plasmid encoding a protein derived from the Gag polyprotein and / or a chimeric integrase, comprising a binding domain making it possible to recognize a packaging sequence, and, (iii) a envelope plasmid encoding an envelope protein.

Such a kit may also include instructions for using the plasmids contained in the kit.

More specifically, when the kit is intended to produce particles according to the first embodiment, this kit comprises: (i) an expression plasmid comprising at least two different non-viral RNA sequences, each RNA sequence comprising a sequence of interest for which upstream or downstream of this sequence of interest is inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each comprising a sequence of interest upstream or downstream of which is inserted an encapsidation sequence, (ii) a packaging plasmid encoding a chimeric nucleocapsid protein having a binding domain for recognizing the encapsidation sequence, and (iii) an envelope plasmid encoding a protein envelope.

The at least two non-viral RNA sequences are different because the sequences of interest are different and / or the encapsidation sequences are different.

Preferably, the kit producing particles according to the first embodiment comprises: (i) an expression plasmid comprising at least two sequences of interest, for which upstream or downstream of each of these sequences is inserted a sequence of encapsidation, or alternatively a first and a second expression plasmid each having a sequence of interest upstream or downstream of which is inserted an encapsidation sequence, the sequences of interest being different and the encapsidation sequences being identical (ii) a packaging plasmid encoding a chimeric nucleocapsid protein having a binding domain for recognizing the encapsidation sequence, and (iii) an envelope plasmid encoding an envelope protein.

Advantageously, the kit produces lentiviral particles.

Preferably: in the expression plasmid, the encapsidation sequence comprises from 2 to 25 repetitions of the stem-loop sequence of MS2, preferentially from 6 to 18 repetitions of the stem-loop sequence, more preferably still from 10 to 14, for example 12 repetitions. Advantageously, the stem-loop sequence is as follows: ctagaaaacatgaggatcacccatgtctgcag (SEQ ID No. 1). In the encapsidation plasmid, the nucleocapsid protein is the nucleocapsid protein (NC) of HIV belonging to the Gag polyprotein, which NC is mutated at the level of the second finger of zinc in order to insert the sequence of the protein Coat. Bacteriophage MS2. In the envelope plasmid, the coat protein is the VSV-G protein encoding the envelope protein of vesicular stomatitis virus.

Optionally, the kit comprises a second encapsidation plasmid encoding a chimeric integrase comprising a binding domain making it possible to recognize a packaging sequence. The second binding domain may be the same or different from the binding domain of the chimeric nucleocapsid protein.

Several binding domains can be introduced into each of the chimeric proteins.

Alternatively, when the kit aims to produce particles according to the second embodiment, this kit comprises: (i) an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted an encapsidation sequence, (ii) a packaging plasmid encoding a chimeric integrase having a binding domain for recognizing the encapsidation sequence, and, (iii) an envelope plasmid encoding an envelope protein .

Methods of manufacturing kits according to the invention are also proposed. Typically, a method of manufacturing a kit according to the invention comprises: (i) the preparation of an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted a encapsidation sequence (ii) the preparation of a packaging plasmid encoding a protein derived from the Gag polyprotein and / or a chimeric integrase, comprising a binding domain making it possible to recognize an encapsidation sequence, and (iii) ) preparing an envelope plasmid encoding an envelope protein.

More specifically, when the method relates to a kit which aims to produce particles according to the first embodiment, it comprises: (i) the preparation of an expression plasmid comprising at least two different non-viral RNA sequences, each an RNA sequence comprising a sequence of interest for which upstream or downstream of this sequence of interest is inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each comprising a sequence of interest in upstream or downstream of which is inserted an encapsidation sequence, (ii) the preparation of a packaging plasmid coding for a chimeric nucleocapsid protein comprising a binding domain making it possible to recognize the encapsidation sequence, and, iii) preparing an envelope plasmid encoding an envelope protein.

The at least two non-viral RNA sequences are different because the sequences of interest are different and / or the encapsidation sequences are different.

Preferably, this method comprises: (i) the preparation of an expression plasmid comprising at least two sequences of interest, for which upstream or downstream of each of these sequences is inserted an encapsidation sequence, or alternatively d a first and a second expression plasmid each comprising a sequence of interest upstream or downstream of which is inserted an encapsidation sequence, the sequences of interest being different and the encapsidation sequences being identical, (ii) preparing a packaging plasmid encoding a chimeric nucleocapsid protein having a binding domain for recognizing the encapsidation sequence, and (iii) preparing an envelope plasmid encoding a protein envelope.

Alternatively, when the method relates to a kit that aims to produce particles according to the second embodiment, this method comprises: (i) the preparation of an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted an encapsidation sequence, (ii) the preparation of a packaging plasmid coding for a chimeric integrase comprising a binding domain making it possible to recognize the encapsidation sequence, and, (iii) preparing an envelope plasmid encoding an envelope protein. The invention also relates to processes for manufacturing the particles according to the invention.

Such a method comprises a step of co-transfection of cells with: (i) an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted an encapsidation sequence, ( ii) a packaging plasmid coding for a protein derived from the Gag polyprotein and / or a chimeric integrase comprising a binding domain making it possible to recognize a packaging sequence, and, (iii) an envelope plasmid coding for a protein of the envelope, and the recovery of the supernatant of the transfected cells comprising the particles.

The cells used in the particle manufacturing processes according to the invention are producing cells, ie cells, which once transfected with the plasmids carrying the genetic material necessary for the formation of retro viral particles. , allow the formation of said particles. As an example of producing cells, mention may be made of HEK293T.

By "co-transfection step" is meant a transfection step in which the transfection is performed by contacting the producer cells with all the plasmids of the particle manufacturing process.

More specifically, when the manufacturing method aims to produce particles according to the first embodiment, it comprises a step of co-transfection of cells with: (i) an expression plasmid comprising at least two non-viral RNA sequences different, each RNA sequence comprising a sequence of interest for which upstream or downstream of this sequence of interest is inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each having a sequence of interest upstream or downstream of which is inserted an encapsidation sequence, (ii) a packaging plasmid encoding a chimeric nucleocapsid protein having a binding domain for recognizing the encapsidation sequence, and, (iii) ) an envelope plasmid encoding an envelope protein, and recovering the supernatant of the transfected cells comprising the particles.

The at least two non-viral RNA sequences are different because the sequences of interest are different and / or the encapsidation sequences are different.

Preferably, this process for producing particles comprises a step of co-transfection of cells with: (i) an expression plasmid comprising at least two sequences of interest, for which upstream or downstream of each of these sequences is inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each comprising a sequence of interest upstream or downstream of which is inserted an encapsidation sequence, the sequences of interest being different and the sequences of encapsidation being identical, (ii) a packaging plasmid coding for a chimeric nucleocapsid protein comprising a binding domain making it possible to recognize the encapsidation sequence, and (iii) an envelope plasmid coding for a protein of envelope, and recovery of the supernatant of the transfected cells comprising the particles.

Alternatively, when the manufacturing method aims to produce particles according to the second embodiment, this method comprises a step of co-transfection of cells with: (i) an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted an encapsidation sequence, (ii) a packaging plasmid coding for a chimeric integrase comprising a binding domain making it possible to recognize the encapsidation sequence, and, (iii) a envelope plasmid encoding an envelope protein, and recovering supernatant from the transfected cells comprising the particles.

Preferably, all the particle manufacturing processes according to the invention are carried out in accordance with the processes described in patent application WO2013 / 014537.

More particularly, these particle manufacturing processes are performed on producer cells grown in serum-free medium and no sodium butyrate induction is performed.

Advantageously, the supernatant is collected in several times, for example between 3 and 6 times, at specific time intervals, such as at time intervals of the order of the half-life of the retro viral particles. Typically, the recovery of the supernatant is carried out in 4 to 5 times, at time intervals of the order of 6 to 18 hours, preferably 8 to 16 hours.

The particle manufacturing processes of the invention also include a step in which the supernatant is clarified.

Preferably, the clarification of the supernatant is carried out by centrifugation.

More preferably still, this method of manufacturing particles according to the invention further comprises a step in which the supernatant is concentrated and / or purified.

Preferably, the concentration and purification is carried out by frontal ultrafiltration on centrifugation units.

Advantageously, the supernatant undergoes one or more additional purification steps. These purification steps are preferably carried out by tangential ultrafiltration and / or diafiltration. More preferably still, the supernatant undergoes a tangential ultrafiltration step followed by a diafiltration step. Tangential ultrafiltration is advantageously carried out on polysulfone hollow fiber cartridges.

Optionally, the composition may then undergo an ion exchange chromatography step, in particular anion exchange chromatography. The eluate resulting from this chromatography step is then recovered and then concentrated again by frontal ultrafiltration on central centrifugation units. A composition resulting from such a process has less than 1% of DNA contaminants and less than 1% of protein contaminants relative to the crude supernatant. The invention also relates to compositions obtainable by any of the particle manufacturing processes according to the invention.

Typically, these compositions comprise less than 30% of DNA contaminants and less than 45% of protein contaminants relative to the crude supernatant. More particularly, the compositions according to the invention comprise less than 30% of DNA contaminants and less than 2% of protein contaminants relative to the crude supernatant.

Optionally, the compositions according to the invention comprise less than 1% of DNA contaminants and less than 1% of protein contaminants relative to the crude supernatant.

Finally, the invention relates to the use of a particle according to the invention, or a composition according to the invention for transducing cells. The use of the particles and compositions according to the invention is particularly advantageous for transducing primary cells and immortalized lines, in order to modify them transiently. The cells may be mammalian cells or other eukaryotes. The invention will be better understood from the following examples given by way of illustration, with reference to the figures, which represent respectively: FIG. 1 is a diagram illustrating the modification of the lentiviral encapsidation plasmid p8.74 in order to insert a binding domain in the integrase sequence, this encapsidation plasmid being used for the production of lentiviral particles according to the invention; Figure II shows different expression plasmid constructs carrying the luciferase RNA sequence of interest (Fig. 11a), a fluorescent reporter (Fig. Mb) or CRE (Fig. Ile), these plasmids expression being used for the production of lentiviral MS2RLP particles according to the invention;

FIG. III shows a diagram illustrating the modification of the lentiviral encapsidation plasmid p8.74 in order to insert a binding domain into the nucleocapsid sequence (Ilia) and the resulting plasmid construct (IIIb), this encapsidation plasmid being used for the production of lentiviral particles MS2RLP according to the invention;

Fig. IV is a construction scheme of an envelope plasmid;

Figure V shows different patterns of expression plasmid constructs bearing the luciferase RNA sequence of interest (Fig. Va), a fluorescent reporter (Fig. Vb) or CRE (Fig.

Vc);

Fig. VI is a construction scheme of a packaging plasmid, which encapsidation plasmid is used for the production of integrative lentiviral (ILV) vectors;

FIG. VII is a construction diagram of a packaging plasmid, this encapsidation plasmid being used for the production of integration-deficient lentiviral vectors (IDLV); Figure VIII illustrates the kinetics of expression of ZsGreen1 in HCT116 cells transduced by lentiviral vectors (ILV, IDLV) and lentiviral MS2RLP particles according to the invention;

Figure IX illustrates the kinetics of expression of luciferase in HCT116 cells transduced by lentiviral vectors (ILV, IDLV) and lentiviral MS2RLP particles according to the invention;

Figure X illustrates the dose effect of luciferase expression after Foreskin fibroblast transduction by ILV vectors and MS2RLP lentiviral particles according to the invention;

Figure XI illustrates the impact of the purity (production with or without serum) of a composition of lentiviral particles MS2RLP according to the invention on the transduction of fibroblasts Foreskin;

FIG. XII illustrates the impact of the purity (production with or without serum) of a composition of lentiviral particles MS2RLP according to the invention, on the transduction of HCT116 cells;

Figure XIII illustrates the impact of the presence of BX795 during Foreskin fibroblast transduction by lentiviral MS2RLP particles according to the invention and ILV vectors;

Figure XIV the impact of the presence of BX795 during the transduction of HCT116 cells by lentiviral particles MS2RLP according to the invention and ILV vectors;

FIGS. XV, XVI and XVII illustrate the kinetics of expression of luciferase by bioluminescence in vivo, in the mouse, after injection of a purified or non-purified suspension of MS2RLP particles according to the invention and of a purified suspension of ILV vectors Fig. XV presents photographs of each animal at given times, Figs. XVI and XVII being graphs showing the luciferase expression measurements at these same times;

FIG. XVIII is a diagram illustrating the quenching of fluorescence in HCT116-Lox-dsRed-Lox cells transduced with ILV vectors, pCre plasmids or MS2RLP particles according to the invention, allowing the expression of Cre recombinase;

FIG. XIX is a diagram illustrating the quantification of the number of AND Cre copies in HCT116-Lox-dsRed-Lox cells after transduction by ILV vectors, pCre plasmids or MS2RLP particles according to the invention;

Figure XX illustrates the transfer kinetics of several RNAs in HCT116 cells with MS2RLP particles according to the invention; and,

Figure XXI shows a comparison of the transfer kinetics of several RNAs in HCT116 cells with MS2RLP particles according to the invention or IDLV vectors.

EXAMPLE 1 Construction of lentiviral particles MS2RLP I. Material & Methods 1. Plasmid Constructs

1.1 Plasmids for the production of lentiviral narticles MS2RLP • Plasmid for expression of a sequence of interest: The expression plasmid carries a promoter-sequence of interest-polyA expression cassette (see FIG. or not an intron or RNA stabilizing sequence. In order to transport the mRNAs in the lentiviral particles, 12 repeats of the stem-loop motif of the MS2 RNA (ctagaaaacatgaggatcacccatgtctgcag, SEQ ID No. 1) were inserted into an expression cassette downstream of the reporter gene. The promoter used may be that of CMV (Figure 11a) or EF1 (Figures 11b and 11c) but other promoters may be used. The sequence of interest may be a DNA encoding a reporter protein such as native Firefly luciferase (Figure 11a), a fluorescent protein (Figure 11b) green (ZsGreenl), red (mCherry) or blue (mtBFP), or a cDNA coding a protein, for example CRE protein (Figure Ile). The sequence of interest may also be that of a shRNA, a miRNA, a sgRNA, an LncRNA or a circRNA. Packaging Plasmid: The lentiviral particle was modified to contain within the nucleocapsid protein, instead of the second Zn finger domain, the sequence of the bacteriophage MS2 coat protein. The encapsidation plasmid p8.74 (FIG. III), carrying the genes encoding the structure and function proteins (Gag, Pol), used for the production of MS2RLP particles is modified according to the strategy illustrated by FIG. 11a: this plasmid p8.74 is used to generate, by assembly PCR, a plasmid lacking the second zinc finger of the p8.74AZF nucleocapsid protein. The second zinc finger is substituted by the Coat protein of MS2 phage by Hpal cloning, to generate the plasmid p8.74AZF-MS2-Coat. This gives the construction illustrated in Figure lllb. The coding sequence Pol can be deleted or mutated in certain functional elements, for example the sequence encoding the reverse transcriptase (RT) or the integrase (IN) without altering the function of the MS2RLPs. Envelope Plasmid (pENV): This plasmid carries the gene encoding an envelope protein, which may be VSVG encoding the envelope protein of vesicular stomatitis virus (Figure IV).

1.2 Plasmids for the Production of ILV Integral Lentiviral Vectors • Plasmid for Expression of a Sequence of Interest: The expression plasmid carries a promoter-sequence expression cassette of interest (FIG. V). This plasmid may contain other elements such as the WPRE (Woodchuck Hepatitis Virus Posttranscriptional Regulatory Element) or the cPPT / CTS sequence. Viral pathogenicity is eliminated by substitution of regions of the viral genome required for retroviral replication by the transgene. Packaging Plasmid: The encapsidation plasmid p8.74, carrying the genes encoding the structure and function proteins (Gag, Pol) is used for the production of integrative lentiviral vectors (FIG. VI). • Envelope plasmid (pENV): This plasmid is identical to the envelope plasmid used for the production of lentiviral particles MS2RLP (FIG. IV). 1.3

IDLV integration: mutant D64L

The 3 plasmids are the same as those described in point 1.2 except for the encapsidation plasmid which comprises a D64L mutation of the pol sequence encoding Ntegrase (Figure VII). This mutation on a single amino acid induces inhibition of integration (Nightingale et al., 2006 and Apolonia et al., 2007). The D64L mutation on integrase (IN) was introduced into the p8.74 encapsidation plasmid by site-directed mutagenesis. The mutation was verified by sequencing. 1.4 plasmids pLuc and oCre

These expression plasmids are similar to those used for the production of ILV and IDLV vectors (Figure Vb for pLuc and Figure Vc for pCre), used directly in transfection for control. 2. Productions of lots

After transfection of the plasmids onto cells, the supernatants are harvested and used raw or concentrated / purified according to the method described in application WO 2013/014537. 2.1 Production of lentiviral vectors and lentiviral particles

Produced in 10-stage CellSTACK (6360 cm2, Corning) with HEK293T cells (ATCC, CRL-11268), grown in Dulbecco's Modified Eagle's Medium (DMEM, Gibco, Paisley, UK) supplemented with 1% penicillin / streptomycin and 1 % of ultraglutamine (PAA) at 37 ° C in a humid atmosphere at 5% CO2. For the serum impact study experiment, DMEM is supplemented with 10% FCS. In particular, no induction of sodium butyrate is carried out. For each lot (MS2RLP, ILV and IDLV), the transfection mixture is composed of the following three plasmids:

One of the expression plasmids described above, according to whether a particle (MS2RLP) or a vector (ILV, IDLV), p8.74AZF Coat (MS2RLP), p8.74 (ILV) or p8.74 mutated is formed. at position D64L (IDLV), and

pENV carrying the VSV-G envelope 24 hours after standard calcium phosphate transfection, the culture supernatant is replaced by fresh and unsupplemented DMEM medium. The cells are incubated at 37 ° C / 5% CO2. After changing the medium, the supernatant is collected four times (32h, 48h, 56h and 72h post transfection). Each recovery is clarified by centrifugation for 5 min at 3000 g before being microfiltered on a 0.45 μm filter (Stericup, Millipore). All recoveries are then mixed to form the crude supernatant. 2.2 Concentration and Durification of Lentiviral Vectors and Lentiviral Darticles

The vectors and particles are concentrated and purified according to one of the two following methods: • The method P1 aims to carry out a frontal ultrafiltration of the supernatant on central centrifugation units. • The P2 process aims to achieve tangential ultrafiltration and diafiltration of the supernatant. The crude supernatant is concentrated and purified by tangential ultrafiltration via the use of polysulfone hollow fiber cartridges. The supernatant is diafiltered continuously for diavolumes against DMEM or TSSM buffer. After the diafiltration, the retentate is recovered and then concentrated again by frontal ultrafiltration on central centrifugation units. 3. Titration 3.1 Titration of functional particles by oPCR,

The HCT116 cells (ATCC, CCL-247) are inoculated in 96-well plate in 100 μl of DMEM supplemented with 10% F VF, 100 μg / ml Stretomycin, 100 U / ml Penicillin and 2 μM L-GIn and then incubated 24h at 37 ° C. / 5% C02. Six serial dilutions are made for each vector as well as for an internal standard. The cells are transduced by these serial dilutions in the presence of Polybrene® 8 μg / ml (Sigma) and then incubated for three days at 37 ° C./5% CO 2. For each series of samples, a non-transduced cell well is added to control. The cells are then trypsinized and the titre (Transduction Unit / mL) is determined by qPCR after extraction of the genomic DNA using the Nucleospin tissue gDNA extraction kit (Macherey-Nagel). The titer obtained (TU / mL) by qPCR is normalized by the internal standard whose title was previously determined by FACS. 3.2 Quantification of physical particles by ELISA P24

The p24 capsid protein is detected directly on the viral supernatant using and following the recommendations of the HIV-1 kit p24 ELISA kit (Perkin Elmer). The captured p24 protein is complexed with a biotinylated polyclonal antibody and then detected by streptavidin conjugated with HRP peroxidase. The resulting complex is detected by spectrophotometry after incubation with the ortho-phenylenediamine-HCl substrate (OPD) producing a yellow color which is directly proportional to the amount of captured p24 protein. The absorbance of each well is quantified at the Synergy H1 Hybrid plate reader (Biotek) and calibrated against the absorbance of a standard range of p24 protein. The viral titre expressed as physical particles per mL is calculated from the concentration of p24 protein obtained knowing that 1pg of p24 protein corresponds to 104 physical particles. 4. Monotransduction

This example aims to develop the conditions of monotransduction by lentiviral vectors or lentiviral particles produced above. 5. Kinetics of expression of ZsGreenl

The HCT116 cells seeded the day before at 25,000 cells / cm 2 in 24-well plate (Corning) are transduced at 100,000 PP / cell in the presence of 8 μg / ml of Polybrene®. At 8h and 24h post-transduction, the cells are trypsinized and analyzed by flow cytometry to measure the percentage of fluorescent cells. Each experiment is performed in triplicate. 6. Kinetics of luciferase expression 6.1 HCT116 cells

HCT116 cells (ATCC, CCL-247) are seeded in 6 or 24-well plates and incubated 24h at 37 ° C / 5% CO 2. The transduction by the ILV, IDLV or MS2RLP particles is carried out in the presence of 4 μg / ml Polybrene. The transduction supernatant is removed 4 hours later and replaced with fresh supplemented culture medium. From 4h to 24h posttransduction, the cells are recovered and the expression of luciferase is analyzed using the OneGIo Luciferase assay kit (Promega) following the recommendations of the supplier and using the Synergy H1 Hybrid plate reader (Biotek) . This test is done in triplicate. In parallel, the HCT116 cells are transfected with the plasmid pLuc. For this, 50000 HCT116 cells are transfected with 2.6 μg of plasmid pLuc using PEI PRO® (Polyplus Transfection). 6.2 Foreskin Fibroblasts

Foreskin fibroblasts, seeded the day before at 10 000 cells / cm 2 in 6-well plate (Corning) are transduced in the presence of 4 μg / ml of Polybrene®. Cells were transduced at 200,000 and 500 OOOPP / cell with MS2RLP-Luc, or at MOI5 with an integrative lentiviral vector Luc.

The cells transduced by the MS2RLP-Luc are trypsinized 8 hours posttransduction and passed in 96-well plate, opaque black background in 100 μl of complete medium. The cells transduced by the ILV Luc are trypsinized 24 h post-transduction and passed into a 96-well plate, opaque black background in 100 μl of complete medium. 3 minutes before reading with the luminometer, 100 μl of One Glo Luciferase reagent (Promega) are added per well to be analyzed. 6.3 In the mouse

The kinetics of expression of luciferase by in vivo bioluminescence were performed.

Three groups of male Balb / c mice were compared, one group received a purified integrative lentiviral vector suspension rLV-EF1-Luc (n = 2), a second group received a suspension of MS2RLP-Luc particles (n = 4) produced according to method P2, a last group received a suspension of MS2RLP-Luc particles (n = 4) produced according to method P1. An uninjected animal served as a negative control for background determination.

The measurements are made from 5h to 24h (Figure XV A), and from 48h to 168h (Figure XV B) after systemic injection. Each measurement of luciferase expression was performed 15 minutes after the intraperitoneal injection of 300 mg / kg of D-luciferin (Promega) with an Andor camera, and the Solis image acquisition software. The acquisition time is 5 minutes and the resulting images have been processed and standardized with the ImageJ software. Figure XV shows the images of each animal at given times. The graphs (Figures XVI and XVII) represent the measurements of luciferase expression, in relative units of luminescence (RLU) at these same times. 7. Impact of particle purity for transduction 7.1 Foreskin Fibroblasts

Foreskin fibroblasts seeded the day before at 10,000 cells / cm 2 in CelIBind (Corning) 24-well plate are transduced in the presence of 4 μg / ml of Polybrene®. The cells were transduced at 500,000 PP / cell with two types of supernatant quality: batch produced in the presence of serum and then obtained according to the P2 vs. method. batch product without serum then obtained according to the method P2. At 8h post-transduction, the cells are trypsinized and passed into a 96-well plate, opaque black background in 100 μl of complete medium. 3 minutes before reading on the luminometer, 100 μl of One Glo Luciferase reagent (Promega) are added per well to be analyzed. 7.2 HCT116 cells

The cells seeded the day before at 25000 cells / cm 2 in CelIBind 24-well plate (Corning) are transduced in the presence of 4 μg / ml Polybrene®. The cells were transduced at 100000 PP / cell with two types of supernatant quality: batch produced in the presence of serum and then obtained according to the method P1 vs. lot produced without serum then obtained according to the method P1. At 8h post-transduction, the cells are trypsinized and passed into a 96-well plate, opaque black background in 100 μl of complete medium. 3 minutes before reading with the luminometer, 100 μl of One Glo Luciferase reagent (Promega) are added per well to be analyzed. 8. Impact of BX795 8.1 Foreskin Fibroblasts

Foreskin fibroblasts seeded the day before to 10 000 cells / cm 2 in 6 well plate CelIBind (Corning) are transduced in the presence of 4 μg / ml of Polybrene®. Cells were transduced at 500000 PP / Cell for MS2RLP particles, and MOI for ILVs. At 8h post-transduction, the cells are trypsinized and passed into a 96-well plate, opaque black background in 100 μl of complete medium. 3 minutes before reading to the luminometer. 100 μL of One Glo Luciferase reagent (Promega) is added per well to be analyzed. 8.2 HCT116 cells

HCT116 cells seeded the day before at 25,000 cells / cm 2 in CelIBind 6-well plate (Corning) are transduced in the presence of 4 μg / ml of Polybrene®. Cells were transduced at 100000 PP / Cell for MS2RLP particles, and MOI for ILVs. At 8h post-transduction, the cells are trypsinized and passed into a 96-well plate, opaque black background in 100 μl of complete medium. 3 minutes before reading to the luminometer 100 μL of One Glo Luciferase reagent (Promega) are added per well to be analyzed.

9. In vitro expression of CRE recombinase

Firstly, HCT116 cells (ATCC, CCL-247) are transduced by the ILV-EF1-Lox-dsRed-Lox lentiviral vector to MOI20 in the presence of 4 μg / ml Polybrene. Six days later, this polyclonal population is cloned by limiting dilution by seeding the cells in 96-well plate at 0.5 cells per well. The monoclonal HCT116-lox-dsRed-lox-clone 14 line was selected by cytometry (MACS Quant VYB, Miltenyi), based on the level of expression of dsRed.

In a second step, the polyclonal and monoclonal cell lines were transduced by ILV-Cre vectors to MOI or MS2RLP-Cre particles at a dose of 25 Physical Particles (PP) / cell, in the presence of 4 μg / ml Polybrene. ®.

The polyclonal and monoclonal HCT116-Lox-dsRed-Lox cells transduced by MS2RLP-Cre particles or ILV-Cre vectors were incubated at 37 ° C / 5% CO 2 for 14 days. At 14 days post-transduction, the expression of dsRed is analyzed by flow cytometry. Non-transduced cells (NT) serve as controls. This experiment is done in triplicate. In parallel, the polyclonal and monoclonal HCT116-Lox-dsRed-Lox cells were transfected with the plasmid pCre. For this, 50,000 HCT 116-Lox-dsRed-Lox cells were transfected with 2.6 μg of the plasmid pCre using PEI P RO® (Polyplus Transfection).

After extraction of the DNA of the HCT 116-Lox-dsRed-Lox cells transduced by the ILV-Cre vectors or by the MS2RLP-Cre particles, or transfected with the plasmid pCRE, the number of copies integrated into the cells is measured by qPCR. by detecting a short sequence of Cre recombinase: gcatttctggggattgcttataacaccctgttacgtatagccgaaattgccaggatcagggttaaagat atctcacgtactgacggtgggagaat (SEQ ID NO: 2) with the following qPCR oligonucleotide pair: Q-CRE-F: 5'-GCATTTCTGGGGATTGCTTA-3 '(SEQ ID NO: 3 Q-CRE-R: 5'-ATT CTCCACCGT CAGTACG-3 '(SEQ ID NO: 4) II. Results 1. Levels of production of lentiviral particles

The first experiments consisted in comparing the respective production levels of MS2RLP particles, IDLV or ILV vectors. Lots of each type of particles were produced and the particle titers were measured by an Elisa P24 assay. The results are shown in Table 1 below:

Table 1: Titers of the ILV, IDLV-D64L and Particle Vectors

MS2RLP

These results show that the three types of particles have identical titers (PP / ml +/-) after concentration by the same ultrafiltration technique. The encapsidation modifications or the mutation of the integrase do not therefore lead to a production deficit of the lentiviral particles. 2. Kinetics of expression in the target cells.

A first series of tests was carried out with MS2RLP particles carrying a single type of RNA encoding a green fluorescent protein (ZsGreenl) characterized by a long half-life time. The obtained results

with MS2RLP particles are shown in Figure VIII and show that the transfer of RNA is detected early (as early as 8h post-transduction). The results obtained show a percentage of HCT116 positive cells comparable to 24 hours by the lentiviral particles MS2RLP or the lentiviral ILV or IDLV-D64L vectors. The percentage of positive cells is detectable as early as 8 hours for the MS2RLP particles but only after 24 hours for the ILV and IDLV-D64L vectors. These MS2RLP particles are therefore functional and have a capacity equivalent to that of an ILV or an IDLV-D64L to introduce a coding sequence into a cell population.

In the case of an integrative lentiviral system, the RNAs are retranscribed by RT. The cDNA is then translocated into the nucleus before it is integrated into the genome of the host cell. It is only from this stage that transcription and translation of the transgene take place. In the case of a non-integrative lentiviral system (mutation on ΓΙΝ), the RNAs are retrotranscribed by the RT. The cDNA is then translocated into the nucleus before being transcribed into RNA and then translated into protein.

Only MS2RLP particles allow early expression of the transgene due to the absence of all these intermediate steps. The delivered RNA is directly translated into protein by the cellular machinery.

A second series of tests was performed with MS2RLP particles carrying a single type of RNA encoding Luciferase (Luc) characterized by a short half-life time. The results obtained with MS2RLP particles are shown in Figure IX and show that the optimal activity of Luciferase is also detected early. The expression of Luciferase is transient because it gradually decreases from 8h to 24h. On the other hand, the expression of the fluorescence remains detectable and even stable on this type of kinetics. These results should consider the half-life of the reporter protein used. It is obvious that the detection of the reporter gene is proportional to this half-life time.

The longer it takes, the longer the protein-related activity will be

In FIG. X, it can also be observed that the expression intensity of luciferase varies according to the PP / cell ratio used during the transduction. This dose effect is shown in particular on Foreskin fibroblasts.

It is therefore important to be able to reach high doses of MS2RLP particles to be applied to the cells to be transduced to induce a significant level of expression capable of translating into biological activity. The concentration of particles appears as a key factor of success. In Foreskin fibroblasts, the expression of Luciferase induced with high dose particles is 500000 PP / cell, is equivalent to that obtained with conventional integrative lentiviral vectors at a multiplicity of infection (MOI) of 5. The determination of optimal dose for HCT116 transduction by MS2RLP particles was carried out under the same conditions as for Foreskin fibroblasts, by testing the doses of 50000, 100000 and 200000 PP / Cell. The dose retained is 100,000 PP / cell, for which the best Luciferase RNA transfer is obtained. These results show that the transduction method must be adapted according to the permissiveness of the target cells

These results are confirmed by in vivo injection experiments of MS2RLP-Luc particles which show a detection of luciferase expression 5 hours after injection. 3. Impact of production and purity of MS2RLP particle suspension on transduction efficiency.

Depending on the target cell type, it is important to be able to increase the particle dose in order to obtain a large number of cells expressing the RNAs transported. It is possible to increase the particle dose without affecting the viability of the target cells by using purified particles, as shown in WO 2013/014537. Indeed, the final purity of the batch of lentiviral particles is affected by the presence of fetal calf serum (FCS).

To evaluate the impact of the purity of the MS2RLP particle suspension on the expression of Luciferase, we compared the expression of Luciferase according to the purity of two types of supernatant lots: a supernatant concentrated from MS2RLP particles produced with serum, a supernatant concentrated from MS2RLP particles produced without serum.

This study was performed in both immortalized and permissive HCT116 cells (Figure XII) and in less permissive and delicate primary cells, Foreskin fibroblasts (Figure XI).

The results obtained show that purity of the lot has a significant impact on the efficiency of transduction of delicate primary cells at high dose. At 500,000 PP / cell, the effect of purity is considerable since the activity of Luciferase is increased by nearly 80% with the lot produced without serum (Figure XI). In the presence of FBS, there is, on the other hand, a decrease in the expression of luminescence.

On highly resistant immortalized HCT116 cells, the impact of purity is not detectable on Luciferase activity (Figure XII). In contrast, the batch produced in the presence of serum shows more heterogeneous results, which shows a much lower reproducibility than with the lot produced without serum. This result therefore justifies the choice to produce batches of particles in the absence of serum.

These experiments demonstrate on the one hand the need to concentrate these lentiviral particles to obtain an important level of expression compatible with the induction of a phenotype at the cellular and tissue level and on the other hand the impact of the purity of these lots. on the efficiency of gene transfer. 4. Identification of optimal conditions for gene transfer by MS2RLP particles.

To increase the efficiency of RNA transfer, the results obtained by Sutlu et al. (2012) to increase the transduction efficiency of Natural Killer NK cells were transposed. Specifically, it has been hypothesized that antiviral responses based on the Toll-like receptor (TLR) and / or RIG-1-like (RLR) receptor may constrain the efficiency of RNA transfer in certain cells. In order to test this hypothesis, an inhibitor of small signaling molecules TLR and RLR was used during contacting of cells with MS2RLP particles. The BX795 molecule is an inhib1 / ΙΚΚε complex inhibitor, which acts as a common mediator in the signaling pathways of RIG-I, MDA-5, and TLR3. The use of the BX795 molecule at a concentration of 6 μΜ considerably increases the transduction efficiency of HCT116 cells (Figure XIV) and Foreskin fibroblasts (Figure XIII). The observation times (8 h for MS2RLP particles, 24h for ILVs) take into account the functioning of each particle (Figures VIII and IX). A synergistic effect between MS2RLP particles and the use of BX795 in HCT 116 cells and in Foreskin fibroblasts can be noted, whereas BX795 induces a negative effect with ILV integrative lentiviral vectors in both cell types. This effect can be linked to a differential in the number of RNA molecules delivered by these two types of particles. 5. Study of gene transfer by MS2RLP Luciferase particles in vivo. The in vivo injection of MS2RLP-Luc particles is intended to demonstrate the kinetics of expression of a transgene, in this case luciferase, in the various organs of mice injected systemically. A suspension of integrating ILV-EF1-Luc purified vectors is used as a positive control. Measurements are made from 5 am and up to 7 days post injection. The intravenous administration generates a dilution of the injected suspension and therefore a diffusion of the bioluminescence signal which must be considered. The analyzes are therefore made on the whole animal to take into account the distribution of the signal in the entire body. A suspension of MS2RLP-Luc particles produced according to method P1 in the presence of serum and a suspension of MS2RLP-Luc particles purified according to process P2 are tested. The injection of the suspension of MS2RLP-Luc particles, according to the method P1 in the presence of serum in which the purification is less extensive, has led to difficulties in both sampling and injection. Indeed, this suspension of MS2RLP-Luc particles, obtained by the method P1 in the presence of serum, is characterized by a high viscosity brown suspension which is particularly difficult to collect with a 29G needle used for in vivo injections. This difficulty is found at the time of delivery of the suspension in the vein of the tail. The use of suspension of particles MS2RLP-Luc obtained by a method P1 in the presence of serum results in poor administration of the particles which are located in the caudal area of the animal (ie at the injection site) and do not pass not in the general circulation of the animal (Figure XV A), time H5 and H8 of the group "MS2RLP-Luc produced according to the method P1 in the presence of serum"). The suspension of MS2RLP-Luc particles obtained by a method P1 in the presence of serum can not therefore be used in the context of in vivo studies.

The purified suspension of MS2RLP-Luc particles could be normally administered and give rise to 818 ULR luciferase expression at 5 hours post-injection. This signal is always visible at 8 o'clock and then disappears at later times (Figures XV A, XVI and XVII), group "purified MS2RLP-Luc"). The bioluminescence signal is mainly found in the liver and spleen (Figure XV A), group "purified MS2RLP-Luc"). By comparison, at 5 hours, the luminescence measured in the animals which received the purified suspension of integrative viral vectors ILV-EF1-Luc is not significant since the reverse transcription and integration are abortive at these short times. On the other hand, at an early stage (5 hours post injection), the signal obtained with the purified suspension of MS2RLP-Luc particles is significant and has reached a level of expression twice as large as the signal obtained with the purified suspension of vectors. integrative viral ILV-EF1-Luc. The purified suspension of MS2RLP-Luc particles reaches, after 5 hours, a level of expression equivalent to that of the purified suspension of integrative viral vectors ILV-EF1-Luc after 168 hours (FIGS. XV B and XVI) group "ILV" -EF1-Luc purified "). The purified suspension of MS2RLP-Luc particles makes it possible to obtain, in vivo, an optimized transient expression equivalent to a level of expression resulting from an integrative lentiviral transduction mechanism. The expression is transient due to the non-integration of the transgene into the genome of the cells. The concentration / purification process thus makes it possible to obtain effective MS2RLP particle suspensions after injection in vivo. 6. Comparison of the functionality of the different particles

The HCT116-Lox-dsRed-Lox cells were previously generated by transduction with an integrative lentiviral vector carrying the following sequence: EF1-Lox-DsRed-Lox. After obtaining a polyclonal fluorescent line, a monoclonal line was obtained by limiting dilution. It should be noted that the number of copies of the integrated Lox-DsRed-Lox sequence has been quantified by qPCR in the polyclonal and monoclonal lines. The number of integrated copies averages 6 for the polyclonal line and 13 copies for the monoclonal line.

These two lines were transduced by MS2RLP particles carrying the coding sequence for the Cre recombinase enzyme. In parallel, these HCT116-Lox-dsRed-Lox cells were transduced by lentiviral ILV vectors bearing the same Cre sequence.

The polyclonal and monoclonal HCT116-Lox-dsRed-Lox cells transduced by the MS2RLP-Cre particles, the ILV-Cre vectors were incubated at 37 ° C / 5% CO2 for 14 days prior to cytometer analysis. Quantification results of the percentage of fluorescent cells by flow cytometry are shown in Figure XVIII. The results show a virtual disappearance of the fluorescence within the polyclonal and monoclonal populations brought into contact with the MS2RLP-Cre particles. Indeed, the percentage of fluorescent cells has increased from 100% to less than 10%. The result is much less effective with integrative vectors ILV-Cre (15% decrease) as well as in the transfection condition of plasmid pCre (no decrease).

The integration events after transfer of nucleic acids encoding CRE recombinase in HCT116 cells by MS2RLP-Cre particles, by ILV-Cre vectors, or by pCre plasmids were verified. For this purpose, we quantified the number of residual copies of the genome carried by the different types of particles and plasmid, in the HCT116-Lox-dsRed-Lox cells contacted with the ILV vectors, the MS2RLP particles or the plasmids encoding the CRE protein. For this purpose, the total DNAs of the wild-type and modified HCT116 cells were prepared 6 and 14 days after the transduction and the DNA coding for the CRE recombinase was measured by qPCR.

The results presented in Figure XVIII show that with integrative lentiviral vectors the number of integration events is 4 copies per cell at 6 days. This number is consistent with the multiplicity of infection (MOI) used. The transfection control shows a large number of DNA copies at 6 days, corresponding to the detection of the plasmid present in the cells, and then total disappearance of Cre copies at 14 days, because of the lack of integration. of the plasmid in the genome of the target cells.

With MS2RLP particles, no integration event is detected. Indeed, since the expression is not dependent on the reverse transcriptase nor on the LTR ends, these particles deliver RNA within the target cells that is either directly translated into proteins or driven by a protein in the nucleus of the cells. and does not generate any kind of DNA.

EXAMPLE 2 Transfer of several different RNAs by single transduction with MS2RLP I particles. Material & Methods 1. Mono- and tri-transduction

This example is carried out using MS2RLP particles or IDLV vectors allowing either the transfer of a single type of RNA allowing the expression of a single protein (ZsGreen1, mCherry or mtBFP), or the transfer of several types of RNAs allowing the expression of several different proteins (ZsGreenl + mCherry + mtBFP).

The MS2RLP particles and IDLV non-integrative lentiviral vectors are produced under the following conditions: the particles were produced by transfection of HEK293T cells with a single expression plasmid encoding either ZsGreen1, mCherry, or mtBFP (like this is described in Example 1), in addition to encapsidation plasmids and the envelope, the particles were produced by transfection of HEK293T cells with three expression plasmids respectively encoding ZsGreen1, mCherry and mtBFP added in equimolar amount, for a total amount of expression plasmids equivalent to the total amount of expression plasmids used for particle production with a single expression plasmid, in addition to encapsidation plasmids and the envelope.

The HCT116 cells were transduced by the MS2RLP particles either by transduction with the 3 types of MS2RLP ZsGreen1 particles, mCherry or mtBFP independently,

by mono-transduction with the particles produced with the three expression plasmids encoding ZsGreen1, mCherry and mtBFP simultaneously.

In parallel, HCT116 cells are transduced by IDLV vectors either by transduction with the 3 types of IDLV ZsGreen1 vectors, mCherry or mtBFP independently, by mono-transduction with the IDLV vectors produced with the three expression plasmids encoding ZsGreen1, mCherry. and mtBFP simultaneously.

The percentage of fluorescent cells was quantified by flow cytometry.

HCT116 cells (ATCC, CCL-247) are inoculated in 96-well plate and incubated 24h at 37 ° C / 5% CO 2. Transduction by MS2RLP particles or IDLV vectors is carried out in the presence of 4 μg / ml Polybrene. At different post-transduction times (8h, 24h and 48h), the cells are recovered and the percentage of cells expressing ZsGreen1, mCherry and mtBFP is quantified by cytometry (Macs Quant VYB, Miltenyi Biotec). An inhibitor of cellular defense mechanisms, BX795 (InvivoGen), is used at a concentration of 6μΜ. Each experiment is performed in triplicate. II. Results

1. Ability to transfer several different RNAs in a single transduction with MS2RLP particles The objective of the following experiments is to demonstrate the ability of MS2RLP particles to transfer several different RNAs within target cells, thereby reducing the number of transductions on the target cells. Thus, it becomes possible to transfer several different RNAs into a single transduction rather than having to transduce by RNA species. For this, we generated purified batches of MS2RLP particles, according to the optimum conditions defined in Example 1, that is to say purified batches, produced without serum, according to the method described in the application WO 2013/014537 . During the production phase of the MS2RLP particles, three types of plasmid coding for fluorescent reporters were used simultaneously in equimolar quantity (ZsGreen1, mCherry and mtBFP), with the encapsidation and envelope plasmids, in order to obtain batch of MS2RLP particles comprising several different types of RNA. The HCT116 cells were then transduced in the presence of BX795, and the transfer of the different RNAs is observed at 8h, 24h and 48h post-transduction.

Figure XX illustrates the proportion of tri-fluorescent cells in green, red and blue, after a single transduction of HCT 116 cells by MS2RLP particles at two different doses (100000 PP / cell and 300000 PP / cell), or after three simultaneous transductions of MS2RLP particles expressing ZsGreen1 or mCherry or mtBFP produced according to Example 1, at a final dose of 300,000 PP / cell.

Contrary to the results of Example 1 showing that it is possible to detect the expression of a single fluorescent or luminescent protein at 8 hours, the expression of the three fluorescences at 8 h post-transduction is not detectable, at a dose of 100000 PP / cell. On the other hand, 48% of cells are trifluorescent at 24 hours post-transduction. This proportion of trifluorescent cells increases, up to 60% at 48 hours post-transduction. The results therefore show that MS2RLP particles are able to transport and transfer at least 3 types of RNA in a single target cell transduction.

The determination of the optimum particle dose per cell is important because it is observed that the increase in the dose of MS2RLP particles results in a decrease in the number of trifluorescent cells (FIG. XX, 300,000 PP / cell), in particular 48 hours after the transduction. In fact, if the proportion of trifluorescent cells at 300,000 PP / cell is a little larger at 8 hours post-transduction (5% of trifluorescent cells) than at 100,000 PP / cell, the proportion of trifluorescent cells begins to decrease at 24 hours. -transduction passing from 48% to 40% of trifluorescent cells. This difference in transfer efficiency between the two doses used is all the greater at 48 h post-transduction: indeed, 43% of trifluorescent cells are observed for the dose of 300,000 PP / cell against 60% of trifluorescent cells for the dose of 100000 PP / cell.

In parallel, three simultaneous transductions of HCT116 cells with batches of MS2RLP particles produced individually according to Example 1 were carried out. The dose used is 100000 PP / cells for each type of fluorescence, ie a total dose of 300000 PP / cell. The results show that the percentage of trifluorescent cells obtained by three simultaneous transductions of particles carrying only one species of RNA is less important than that induced by a single transduction by particles carrying the 3 RNA species. Indeed, only 7% of trifluorescent cells are detected at 8h post-tranduction, then 20% of trifluorescent cells at 24 hours post-transduction, only 42% of the level obtained with the single transduction at 100000 PP / cell and 50% of the level. obtained with single transduction at 300000 PP / cell; and finally 23% of trifluorescent cells at 48 hours post-transduction, ie 38% of the percentage obtained with the single transduction at 100,000 PP / cell and 54% of the percentage obtained with the single transduction at 300,000 PP / cell. In conclusion, tri-transduction does not make it possible to achieve tri-color cell percentages as high as those obtained with the three simultaneous transductions.

The demonstration of the transfer capacity of different types of RNA in a single transduction of the same batch of MS2RLP particles represents a significant gain on the simultaneous transfer of these different RNAs into the target cells. This new property of the MS2RLP particles, which has the direct consequence of optimizing the number of transductions to be carried out, makes it possible not to compromise the viability of the target cells by both an excessive number of transductions as well as by putting the cells into contact with each other. with too much particulate matter.

2. Comparison with an IDLV vector

The same experiments as in point 1 above were reproduced with another type of particles in parallel: a deficient viral vector for integration by a mutation at position 64 on the sequence encoding Integrase (IDLV).

Figure XXI illustrates the proportion of tri-fluorescent cells in green, red and blue, after a single transduction of HCT116 cells with MS2RLP particles produced with the 3 ZsGreen1 expression plasmids, mCherry and mtBFP, in comparison with three HCT116 cell transductions with IDLV vectors expressing ZsGreen1 or mCherry or mtBFP produced according to Example 1, at a total dose of 300000 PP / cell.

The IDLV vectors are produced under the same conditions as the MS2RLP particles in order to allow single transduction at the same dose as that used for the MS2RLP particles. At 24 hours after the single transduction at a dose of 100000 PP / cell, the proportion of trifluorescent cells is slightly greater in the case of the use of IDLV vectors compared to that obtained with the MS2RLP particles (respectively 56% and 48%). %). This trend is confirmed 48 hours after the single transduction at a dose of 100000 PP / cell (80% and 60% respectively). At 24 hours post-transduction, the single transduction of HCT116 cells by IDLV vectors results in a percentage of tri-stained cells equivalent to that obtained after tri-transduction by IDLV vectors. This trend remains stable at 48 h post-transduction, with a minimal difference between the two transduction methods (71% for tri-transduction versus 81% for single transduction). Thus, unlike the results obtained with the MS2RLP particles, the IDLV vectors produced with 3 expression plasmids do not allow a significant increase in the percentage of tri-stained cells during their transduction, in comparison with a tri-transduction. This advantage of MS2RLP particles is probably explained by their ability to package more RNA molecules than IDLVs or ILVs (which strictly encapside only 2 molecules). Thus the expression of 3 different transgenes is facilitated because it does not require the transduction of the same cell by several lentiviral particles.

Note also that the percentage of trifluorescent cells at 8h post-transduction is not detectable with the IDLV vectors at 300,000 PP / cell in the case of the three transductions or in that of the monotransduction. At 300,000 PP / cell, the MS2RLP particles, on the other hand, make it possible to detect trifluorescent cells at 8h post-transduction (FIG. XX).

In the case of the three mono-transductions by IDLV vectors, the proportion of trifluorescent cells is slightly greater at 24 hours and 48 hours after the three transductions (60% and 70% trifluorescent cells respectively) than in the case of transduction. single particle MS2RLP at 100000 PP / cell (respectively 48% and 60% trifluorescent cells). It is important to note that this transfer difference is approximately 20% for a dose used for IDLV vectors 3 times higher than the optimal dose of MS2RLP particles.

In addition, IDLV particles, although deficient for integration, do not make it possible to overcome residual integrations in the genome of the target cells (Nightingale et al., 2006 and Apolonia et al., 2007). As a result, the probability of residual integration events increases with the IDLV particle dose used. The use of MS2RLP particles, by their nature, thus makes it possible to overcome this constraint by the total absence of residual integration, whatever the dose used.

In conclusion, the latter results show that only MS2RLP particles are capable of introducing more than 2 RNA species into target cells from a single particle composition at short post-transduction or injection times. -8 hours) without inducing an integration event.

Claims (21)

A retroviral particle comprising a protein derived from the Gag polyprotein, an envelope protein, optionally an integrase and at least two encapsidated non-viral RNAs, the packaged non-viral RNAs each having an RNA sequence of interest related to an encapsidation sequence, each encapsidation sequence being recognized by a binding domain introduced into the protein derived from the Gag polyprotein and / or into the integrase. 2. Retroviral particle according to claim 1, comprising a nucleocapsid protein, an envelope protein, optionally an integrase and at least two encapsidated non-viral RNAs, the packaged non-viral RNAs each having an RNA sequence of interest. linked to a packaging sequence, each encapsidation sequence being recognized by a binding domain introduced into the core protein and / or integrase. The retroviral particle of claim 2, wherein the binding domain is introduced into the core protein and the at least two encapsidated non-viral RNAs are distinguished by their RNA sequence. The particle of claim 3, wherein a second binding domain is introduced into the core protein. The particle of claim 3 or 4, wherein a binding domain is also introduced into the integrase. The retroviral particle of any one of claims 2 to 5, which is a lentiviral particle. The lentiviral particle according to claim 6, wherein: the binding domain is the sequence of the bacteriophage MS2 coat protein, the encapsidation sequence of the non-viral RNAs is a stem-loop sequence of MS2, the Nucleocapsid protein is the nucleocapsid protein (NC) of HIV belonging to the Gag polyprotein, which NC is mutated at the second finger of zinc in order to insert the sequence of the protein "Coat" bacteriophage MS2. 8. Retroviral particle according to claim 1 or 2, wherein the binding domain is introduced into the integrase. 9. Composition comprising particles according to any one of claims 1 to 8. 10. A kit for producing particles according to any one of claims 1 to 8, comprising: (i) an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted a encapsidation sequence, (ii) a packaging plasmid encoding a protein derived from the Gag polyprotein and / or a chimeric integrase, comprising a binding domain making it possible to recognize a packaging sequence, and, (iii) a envelope plasmid encoding an envelope protein. A particle production kit according to any one of claims 3 to 7, comprising: (i) an expression plasmid comprising at least two different non-viral RNA sequences, each RNA sequence comprising a sequence of interest for which upstream or downstream of this sequence of interest is inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each comprising a sequence of interest upstream or downstream of which is inserted a encapsidation sequence, (ii) a packaging plasmid encoding a chimeric nucleocapsid protein having a binding domain for recognizing the encapsidation sequence, and (iii) an envelope plasmid encoding a protein of envelope. The particle production kit according to claim 11, comprising: (i) an expression plasmid comprising at least two sequences of interest, for which upstream or downstream of each of these sequences is inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each comprising a sequence of interest upstream or downstream of which is inserted a packaging sequence, the sequences of interest being different and the encapsidation sequences being identical, ii) a packaging plasmid encoding a chimeric nucleocapsid protein having a binding domain for recognizing the encapsidation sequence, and, (iii) an envelope plasmid encoding an envelope protein. 13. A method of manufacturing a kit according to claim 10, comprising: (i) the preparation of an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted a encapsidation sequence (ii) the preparation of a packaging plasmid encoding a protein derived from the Gag polyprotein and / or a chimeric integrase comprising a binding domain making it possible to recognize an encapsidation sequence, and (iii) the preparing an envelope plasmid encoding an envelope protein. A method of manufacturing a kit according to claim 11, comprising: (i) preparing an expression plasmid comprising at least two different non-viral RNA sequences, each RNA sequence comprising a sequence of interest for which upstream or downstream of this sequence of interest is inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each comprising a sequence of interest upstream or downstream of which is inserted a encapsidation sequence, (ii) preparing a packaging plasmid encoding a chimeric nucleocapsid protein having a binding domain for recognizing the encapsidation sequence, and (iii) preparing a plasmid for envelope encoding an envelope protein. 15. A method of manufacturing a kit according to claim 12, comprising: (i) the preparation of an expression plasmid comprising at least two sequences of interest, for which upstream or downstream of each of these sequences is inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each comprising a sequence of interest upstream or downstream of which is inserted a packaging sequence, the sequences of interest being different and the encapsidation sequences being identical, (ii) the preparation of a packaging plasmid coding for a chimeric nucleocapsid protein comprising a binding domain making it possible to recognize the encapsidation sequence, and (iii) the preparation of an envelope plasmid encoding an envelope protein. 16. A method of manufacturing particles according to any one of claims 1 to 8, comprising a step of co-transfection of cells with: (i) an expression plasmid comprising at least one sequence of interest, for which upstream or downstream of this sequence is inserted an encapsidation sequence, (ii) a packaging plasmid encoding a protein derived from the Gag polyprotein and / or a chimeric integrase comprising a binding domain making it possible to recognize a packaging sequence , and (iii) an envelope plasmid encoding an envelope protein. and recovering the supernatant from the transfected cells comprising the particles. 17. A method of manufacturing the particles according to any one of claims 3 to 7, comprising a step of co-transfection of cells with: (i) an expression plasmid comprising at least two RNA sequences | different non-viral, each RNA sequence comprising a sequence of interest for which upstream or downstream of this sequence of interest is inserted an encapsidation sequence, or | alternatively a first and a second expression plasmid each having a sequence of interest upstream or downstream of which is inserted an encapsidation sequence, (i {) a packaging plasmid encoding a chimeric nucleocapsid protein comprising a binding domain for recognizing the encapsidation sequence, and (iii) an envelope plasmid encoding an envelope protein, and recovering the supernatant of the transfected cells comprising the particles. 18. The method of manufacturing the particles according to claim 17, comprising a step of co-transfection of cells with: (i) an expression plasmid comprising at least two sequences of interest, for which upstream or downstream of each of these sequences are inserted an encapsidation sequence, or alternatively a first and a second expression plasmid each comprising a sequence of interest upstream or downstream of which is inserted an encapsidation sequence, the sequences of interest being different. and the encapsidation sequences being identical, (ii) a packaging plasmid coding for a chimeric nucleocapsid protein comprising a binding domain making it possible to recognize the encapsidation sequence, and (iii) an encapsidation plasmid coding for envelope protein, and recovery of the supernatant from the transfected cells comprising the particles. 19. The method of manufacture according to any one of claims 16 to 18, wherein the supernatant is clarified. 20. The manufacturing method according to claim 19, wherein the supernatant is further concentrated and / or purified. 21. In vitro use of a particle according to any one of claims 1 to 8, or a composition according to claim 9 for transducing cells.