WO2012168480A1 - Agents and methods for producing hiv-capsid derived non-infectious adjuvants - Google Patents

Abstract

The invention relates to non-infectious HIV-capsid derived agents and their use as adjuvants and vaccines.

Description

AGENTS AND METHODS FOR PRODUCING HIV-CAPSID DERIVED NON-INFECTIOUS ADJUVANTS

FIELD OF THE INVENTION

The present invention relates to adjuvants for prophylactic and therapeutic vaccines.

In particular, it relates to HIV-capsid derived non-infectious agents and their use as adjuvants.

BACKGROUND OF THE INVENTION HIV, or Human Immunodeficiency Virus, is the causative agent of the Acquired Immunodeficiency Syndrome (AIDS). This pandemic agent is responsible for over a million deaths worldwide each year.

HIV is a RNA virus that is duplicated in a host cell using the reverse transcriptase enzyme to produce DNA from its RNA genome. The DNA is then incorporated into the host's genome by an integrase enzyme. The virus thereafter replicates as part of the host cell's DNA.

Antiretroviral drugs are medications for the treatment of infection by retroviruses, primarily HIV. Different classes of antiretroviral drugs act on different stages of the HIV life cycle. Combination of several (typically three or four) antiretroviral drugs is known as highly active anti-retro viral therapy (HA ART).

HIV binds to CD4 cell surface molecules (entry into the cell also requires binding to co-receptors CXCR4 and CCR5). This step can be inhibited by fusion or entry inhibitors. HIV is uncoated inside the cell and reverse transcriptase copies genomic RNA into DNA, making errors at a frequency of about one per replication cycle. Reverse transcriptase inhibitors were the first class of HIV inhibitors to be used as drugs.

Viral DNA can integrate into DNA and become a part of the cellular genome. This step makes the infection irreversible, and may mean that eliminating the virus from an infected individual is not possible. Integrase inhibitors are designed to block this step of infection.

The virus uses cellular machinery to synthesize viral proteins. Several of these are long amino acid chains which must be cleaved by a specific viral protease before new viral particles can become active. Protease inhibitors block viral maturation at this step.

RNA viruses rapidly mutate. The rapidly changing virus makes therapy difficult since resistant viruses emerge at a high frequency.

Moreover, antiretroviral drugs are expensive and treatments are long. Thus, there is a risk of non-compliance with the administration regimens which severely limits their efficacy.

Therefore, there is a great need for preventive strategies such as vaccination. Earlier clinical trials using HIV envelope-based subunit vaccines elicited antibodies that reacted with the gpl20 envelope protein but were not neutralizing antibodies (NAbs), and vaccination failed to show protection against ΗΓ infection. The failure of these trials promoted a shift to the development of HIV vaccines that focus on eliciting T cell responses. However, the disappointing outcome from a recent clinical trial of a T-cell-based vaccine regimen, the STEP trial conducted by Merck and HIV Vaccine Trials Network (HVTN), dealt another setback to AIDS vaccine development.

Yet another possibility would be to focus on eliciting an HIV-specific response in DCs. Indeed, dendritic cells (DC) play a major role in detecting and initiating the response to pathogens, thus linking the innate response to adaptive immunity. However, DCs are not activated by HIV-1 because they are largely resistant to infection with HIV-1 (Granelli-Piperno et al., 2004, PNAS, 101:7669, PMID 15128934). Current vaccination strategies therefore rely on non-HIV vectors (Modified Ankara vaccines, Pox, adenovirus) that possess non-HIV-derived adjuvant effects and may thus divert the immune system from inducing an HIV- specific response.

Thus, there is still a need in the art for a method for specifically activating DC against HIV in the absence of cell infection.

Previous experiments have shown that DC resistance to HIV-1 could be circumvented; thereby leading to an activation of the innate immune response (Manel et al., 2010, Nature, 472:286). In this setting, productive viral infection of the DC is required to induce innate immune activation. Specifically, an interaction between newly synthesized viral Capsid and cellular Cyclophilin A is required. However, there is a risk that the infection would then spread to other cell types. Thus, there is a need for agents that would drive the innate immune activation in the absence of productive viral infection.

SUMMARY OF THE INVENTION

The invention relates to a replication-defective retroparticle comprising a capsid protein, wherein said capsid protein binds to cyclophilin A and wherein said replication-defective particle is capable of activating dendritic cells (DCs) without infecting said DCs. The invention relates to a replication-defective particle comprising a capsid protein,

wherein said capsid protein carries at least one mutation in the cyclophilin- binding loop compared to residues 83 to 98 of the HIV-2 capsid polypeptide having the sequence set forth in SEQ ID No: l,

and wherein said replication-defective particle is capable of activating dendritic cells (DCs) without infecting said DCs. The invention relates to a replication-defective particle comprising a capsid protein,

wherein said capsid protein binds to cyclophilin A with an increased affinity compared to the capsid polypeptide having the sequence set forth in SEQ ID No : 1,

wherein said replication-defective particle comprises a Vpx protein, viral RNA and a reverse transcriptase.

The invention also relates to a vector comprising a nucleic acid encoding a replication-defective particle as defined above.

In another aspect, the invention also relates to a pharmaceutical composition comprising a replication-defective particle and/or a vector as defined above and a pharmaceutically acceptable carrier or excipient. In yet another aspect, the invention also relates to a replication-defective particle and/or a vector and/or a pharmaceutical composition as defined above for use in a method of treatment.

Also provided is a method of treatment, comprising administering a replication- defective particle and/or a vector and/or a pharmaceutical composition to a patient in need thereof. The invention further relates to an in vitro method for inducing the activation of dendritic cells in the absence of infection comprising the step of exposing said DC to a replication-defective particle and/or a vector and/or a pharmaceutical composition as defined above.

The invention also relates to a polypeptide comprising an amino acid sequence having the sequence as set forth in any of SEQ ID No:2 to SEQ ID No:7 and SEQ ID No:53 to SEQ ID No:61.

DETAILED DESCRIPTION OF THE INVENTION

Definitions As used herein, the term "cyclophilin A" or "CYPA" refers to the human protein having the sequence as set forth under accession number NP_066953.1, encoded by the human gene "PPIA" (Genbank accession number NM_021130.3). Cyclophilin A is a cellular peptidylprolyl isomerase which was shown to play a role in HIV infection. Cyclophilin A is known to bind to the N-terminal domain of HIV-1 capsid (Price et al., 2009, Nature Structural and Molecular Biology, 16:1036, PMID 19767750). Binding of HIV to CypA occurs through a cyclophilin binding-loop of the HIV capsid. Examples of such HIV cyclophilin-binding loop are given in figures 2 and 3. As used herein, the expression "binds to cyclophilin A" refers to a protein which binds to cyclophilin A with the same affinity as the wild-type capsid from which it is derived, or with an increased affinity compared to the wild-type capsid from which it is derived. Cyclophilin A has a low affinity for HIV-2 capsid (Kd of 91 μΜ) and a higher affinity for HIV-1 capsid (Kd of 5 μΜ). Insertion of alanine 88 residue of HIV-1 capsid in HIV-2 capsid increases the affinity of HIV-2 capsid to cyclophilin A (8 μΜ) (Price et al., see above). Typically, the affinity of a capsid protein for cyclophilin A can be measured by any method known to those skilled in the art including but not limited to Biacore analysis; immunoprecipitation analysis or isothermal titration calorimetry according to the protocol described in Price et al,.(Nature Structural and Molecular Biology, 16: 1036, PMID 19767750).

For instance, the binding affinity of the capsid protein can be measured by measuring the levels of cyclophilin A in viral particles, as described in the Examples below (see Figure 26). Briefly, viral particle supernatants are recovered, lysed, resolved on SDS-PAGE gels and the amount of cyclophilin A is measured by Western blot.

The affinity of the capsid protein for cyclophilin is deemed to be increased compared to affinity of HIV-2 capsid to cyclophilin A when it is increased by a factor of more than 2, preferably more than 3, even more preferably more than 5, 10, 15, 20, 25, or 28.

Typically, a mutant HIV-2 capsid according to the invention has a similar affinity for cyclophilin A as the capsid protein from the HIV-2 ROD isolate as defined in SEQ ID No: l, preferably a higher affinity for cyclophilin A than the capsid protein having SEQ ID No: l, even more preferably an affinity in the same range as the affinity of HIV-1 capsid for cyclophilin A.

According to one embodiment, the capsid protein of the replication-defective particle of the binds to cyclophilin A with an increased affinity compared to the capsid polypeptide having the sequence set forth in SEQ ID No : 1, as measured by measuring the levels of CypA in viral particles.

As used herein, the term "replication-defective retroparticle" refers to a retroviral particle which comprises all the necessary components to enter into permissive cells, but which is not replicative, i.e. which does not lead to the production of other new infectious particles after entering the cell. It is not capable of self- replication within the target cell. According to the invention, a replication- defective particle can be devoid of viral nucleic acid. Alternatively, it may contain a nucleic acid molecule.

As used herein, the "retroviral" has its general meaning in the art. It refers to the class of viral agents whose genome is encoded by a single-stranded RNA in the particle and whose replication cycle requires a reverse-transcription step into double- stranded DNA and irreversible integration in the cellular genome. Retroviral particles are composed of structural proteins that form the core of the particle ("Gag"), enzymatic proteins that perform essential functions such as reverse-transcription and integration ("Pol"), a nucleic acid, generally a single- stranded RNA molecule, that contains the retroviral genome, and a lipid bilayer derived from the cellular membrane of the cells that produced the particle. "Gag" is composed of different sub-units, at least matrix, capsid, nucleocapsid. "Pol" is composed of different sub-units, at least reverse-transcriptase, integrase, protease. "Env" refers the envelope proteins anchored into the lipid bilayer of the viral particles. Env recognizes a receptor at the surface of target cells, and allows the fusion of the lipid membrane of the viral particle with the lipid membrane of the cell. This allows the delivery of the viral core into the target cell cytoplasm.

According to the present invention, the retroparticle may be a particle of any type of retrovirus including, but not limited to gammaretroviruses (such as Murine Leukemia Virus, MLV, and Feline Leukemia Virus, FLV) and lentiviruses (such as SIV, FIV, BIV, EIA, and CAEV). A retroparticle according to the invention may also be a chimeric particle comprising components from different retroviruses.

As used herein, the term "viral core" has its general meaning in the art and refers to a retroviral particle which is devoid of membrane or envelope.

As used herein, the term "capsid" has its general meaning in the art. It refers to the viral protein which forms the protein shell of the retroparticle. Unless otherwise specified, the term capsid is used in the present document to designate a retroviral capsid, preferably a lentiviral capsid, and even more preferably a HIV capsid. HIV capsid proteins are processed from the Gag polyprotein and contain a cyclophilin- binding loop.

A reference wild-type capsid is the sequence as set forth in SEQ ID No: l, which corresponds to the capsid protein of HIV-2 and is obtained from the Gag polyprotein of HIV-2 isolate ROD. The amino acid residues described herein are preferred to be in the "L" isomeric form. However, residues in the "D" isomeric form can be substituted for any L- amino acid residue, as long as the desired functional property of DC activation is retained by the polypeptide. NH3 refers to the free amino group present at the amino terminus of a polypeptide. COOH refers to the free carboxy group present at the carboxy terminus of a polypeptide.

In keeping with standard polypeptide nomenclature, J. Biol. Chem., 243:3552-59 (1969), abbreviations for amino acid residues are shown in the following Table of Correspondence:

All amino-acid residue sequences are represented herein by formulae whose left and right orientation is in the conventional direction of amino-terminus to carboxy- terminus . As used herein, the term "mutation" has its general meaning in the art. It refers to a change in amino acid sequence when compared to a sequence of reference, such as a so-called wild-type sequence, and encompasses mutations by insertion, deletion, and/or substitution.

Throughout the present document, the amino acid numbering is with respect to the first residue of the capsid protein of HIV-2 ROD isolate (SEQ ID No: 1). Typically, the mutation can be a combined replacement and insertion, such as the P86HA mutant, in which the "P" residue of the HIV-2 capsid at position 86 is replaced by two residues, namely H and A.

Alternatively, the mutation can be a single substitution. For example, I85A refers to a capsid protein carrying A in replacement of the I residue at position 85 of the HIV-2 capsid protein.

As used herein, the terms "mutated capsid" or "mutant capsid" refer to a capsid protein which comprises at least one mutation compared the capsid protein of the HIV-2 ROD isolate.

Typically, a mutant capsid according to the invention is the amino acid sequence having the sequence as set forth in SEQ ID No: 8, which corresponds to SEQ ID No: l in which the mutation P86HA has been introduced.

Another mutant capsid protein according to the invention is the amino acid sequence having the sequence as set forth in SEQ ID No:9, which corresponds to the capsid of SIVmac239 in which the cyclophilin-binding loop has been replaced by SEQ ID No:2. Another mutant capsid protein according the invention is the amino acid sequence having the sequence as set forth in SEQ ID No 63 which corresponds to capsid of HIV-1 in which the cyclophilin-binding loop has been replaced by a sequence which reproduces the cyclophilin-binding loop of HIV-2 P86HA.

Any retroviral capsid homologous to HIV capsid, in which a cyclophilin-binding loop is present or can be introduced, may be used to generate a mutant capsid. Typically, the capsid may be derived from HIV-2, HIV-1, SIVmac239, or other lenti viruses or retroviruses.

As used herein, the term "Gag" has its general meaning in the art and refers to the Gag polyprotein, which is processed during maturation to MA (matrix protein, pl7); CA (capsid protein, p24); SP1 (spacer peptide 1, p2); NC (nucleocapsid protein, p7); SP2 (spacer peptide 2, pi) and p6.

An example of Gag polyprotein according to the invention is the sequence as set forth in SEQ ID No: 10.

Another example of Gag polyprotein is the sequence as set forth in SEQ ID No: 11, which is the Gag polyprotein of SIVmac239 in which the cyclophilin-binding loop of the capsid has been replaced with the mutated cyclophilin-binding loop of SED ID No:2.

As used herein, the term "Pol" refers to the polyprotein which is processed during maturation to viral enzymes reverse transcriptase, integrase, and protease.

An exemplary Pol sequence is the aminoacid sequence as set forth in SEQ ID No: 12 which corresponds to the Pol protein of HIV-2 ROD isolate.

Another example is the aminoacid sequence as set forth in SEQ ID No: 13, which corresponds to the Pol protein of SIVmac239 isolate. "Tat" refers to the viral protein that induces transcription from the integrated genomic LTR. An exemplary Tat sequence is the aminoacid sequence as set forth in SEQ ID No: 14, which corresponds to the Tat protein of HIV-2 ROD isolate. Another example is the aminoacid sequence as set forth in SEQ ID No: 15, which corresponds to the Tat protein of SIVmac239 isolate.

"Rev" refers to the viral protein that prevents splicing of the genomic RNA allowing proper incorporation in the viral particle of the RNA.

An exemplary Rev sequence is the aminoacid sequence as set forth in SEQ ID No: 16 which corresponds to the Rev protein of HIV-2 ROD isolate.

Another example is the aminoacid sequence as set forth in SEQ ID No: 17, which corresponds to the Rev protein of SIVmac239 isolate.

As used herein, the term "Vpx" refers to the HIV-2 protein that allows to bypass the block to reverse-transcription in myeloid cells. An exemplary Vpx protein is the HIV-2 protein having the sequence as set forth in SEQ ID No: 18.

Another example is the aminoacid sequence as set forth in SEQ ID No: 19, which corresponds to the Vpx protein of SIVmac239 isolate.

As used herein, the term "exposure" has its general meaning in the art. It refers to the incubation of target cells with a preparation containing the viral particles.

As used herein, the term "delivery" has its general meaning in the art. It refers to the penetration of the viral core into the target cell cytoplasm following exposure. As used herein, the term "infection" has its general meaning in the art. If refers to progression of the viral cycle at least up to the irreversible integration. If the integrated DNA is functional, expressed and codes for all viral elements, infection can be productive, leading to generation of progeny infectious particles, leading to viral "replication". Infection can also be non-productive, if the integrated DNA is not functional, or not expressed, or is mutated in essential viral elements. In the latter case, "replication" does not occur.

Typically, infection can be monitored using GFP as a reporter. For instance, the GFP gene can be inserted into the vector encoding the replication-defective particle under the control of a viral promoter, such as in place of the nef gene.

As used herein, the term "activation" or "maturation" refers to the triggering of an innate immune response in dendritic cells. The skilled person in the art knows how to assess this response by standard methods in the art.

Typically, DC activation can be assayed by monitoring the expression of CD86 by flow cytometry according to the protocol described in the Examples below. DC activation can be further confirmed by monitoring the cell surface expression of CD83 and/or CD38 or by measuring the amount of IFN released by said DCs into the culture medium. Suitable methods for measuring the amount of IFN released into the culture medium include measuring the IFN acitivity of the cell culture supernatants using a recombinant HL116 cell line, as described in the Examples below. The term "adjuvant" as used herein means a substance that helps or enhances the pharmacological effect of a drug or of a vaccine or increases the ability of an antigen to stimulate the immune system.

The term "therapy", "treatment" or "treat" as used herein means reversing, alleviating, inhibiting the progress of, or preventing the disorder or condition to which such term applies, or one or more symptoms of such a disorder or condition. They encompass both prophylactic and therapeutic treatments.

As used herein, the term "patient" can include human patients as well as animals. In this respect, the diagnostic and therapeutic methods can be performed in the veterinary context, i.e., on domestic animals, particularly mammals (e.g., dogs, cats, etc.) or agriculturally-important animals (e.g., horses, cows, sheep, goats, etc.) or animals of zoological importance (apes, such as macaques, gorillas, chimpanzees, and orangutans, large cats, such as lions, tigers, panthers, etc., antelopes, gazelles, and others). In a preferred embodiment, said patient is a mammalian, preferably a primate, even more preferably a human patient.

The inventors have discovered a replication-defective retroparticle comprising a capsid protein, which is capable of activating dendritic cells (DCs) without infecting said DCs.

Hence, the invention relates to a replication-defective retroparticle comprising a capsid protein, wherein said capsid protein binds to cyclophilin A and wherein said replication-defective particle is capable of activating dendritic cells (DCs) without infecting said DCs.

The invention relates to a replication-defective retroparticle comprising a capsid protein, wherein said capsid protein carries at least one mutation in the cyclophilin-binding loop compared to the wild type capsid protein, wherein said replication-defective particle is capable of activating dendritic cells (DCs) without infecting said DCs. In one embodiment, said mutation in said cyclophilin-binding loop is compared to residues 83 to 98 of the HIV-2 capsid polypeptide having the sequence set forth in SEQ ID No: l. The invention also relates to a replication-defective particle comprising a capsid protein,

wherein said capsid protein binds to cyclophilin A with an increased affinity compared to the capsid polypeptide having the sequence set forth in SEQ ID No : 1,

wherein said replication-defective particle comprises a Vpx protein, viral RNA and a reverse transcriptase.

In one embodiment, said capsid protein carries at least one mutation in the cyclophilin-binding loop compared to residues 83 to 98 of the HIV-2 capsid polypeptide having the sequence set forth in SEQ ID No: 1.

The inventors have observed that the viral DNA of particles that satisfy these criteria does not integrate into the cellular genome. Hence, the particles are able to activate DCs, but they are replication-deficient and therefore non-infectious.

For additional safety, the particle can be devoid of integrase or integrase activity. Hence, in a preferred embodiment, said particle does not contain an integrase or integrase activity. In one embodiment, said capsid protein carries a mutation in the cyclophilin- binding loop at the residue P86 wherein the amino acid numbering is with reference to the HIV-2 capsid polypeptide as set forth in SEQ ID No: l.

In one embodiment, said mutation is selected from the group consisting of:

- P86HA, P86HV, P86HG, P86HL, P86HM, P86HI, P86HT, P86HD,

- P86KA, P86KV, P86KG, P86KL, P86KM, P86KI, P86KT, P86KD, - P86RA, P86RV, P86RG, P86RL, P86RM, P86RI, P86RT, P86RD,

- P86YA, P86YV, P86YG, P86YL, P86YM, P86YI, P86YT, P86YD,

- P86FA, P86FV, P86FG, P86FL, P86FM, P86FI, P86FT, P86FD,

- P86WA, P86WV, P86WG, P86WL, P86WM, P86WI, P86WT, P86WD, - P86AM, P86PI, P86AA and P86QA.

In one embodiment, said mutation is selected in the group consisting of P86HA, P86RA, P86AM, P86HV, P86PI, P86AA and P86QA. In a preferred embodiment, said mutation is P86HA.

In the "P86HA mutant", the wild-type P residue of the HIV-2 capsid protein is replaced by amino acids HA. The same applies, mutatis mutandis, for the other mutants.

Several amino acids positions in the cyclophilin-binding loop are known to vary in known lentiviral capsid sequences. These aminoacids can thus vary accordingly in the invention and can be mutated to amino acid with similar biochemical properties or with known other amino acids at this position, as long as they maintain the properties of the agent (the ability to produce non-replicative retroviral particles that activate innate immunity).

According to one embodiment of the invention, the capsid further comprises a mutation selected from the group consisting of:

- I85A, I85G, I85L I85M, I85V, I85T, I85Y, I85H, I85F, I85S, I85N,

- L89A, L89V, L89G, L89I, L89M, L89T, L89Y, L89H, L89P,

- P90A, P90R, P90T, P90S, P90T, P90Y, P90H,

- A91I, A91V, A91G, A91L, A91M,

- L94I, L94A, L94V, L94G, L94M,

- E96D, - R98S

and combinations thereof;

wherein the amino acid numbering is with reference to the HIV-2 capsid polypeptide as set forth in SEQ ID No: l.

Without wishing to be bound by theory, it is believed that mutant capsid proteins according to invention are able to induce DC activation in the absence of infection due to an increased interaction of such mutated capsid protein with cyclophilin A. Indeed, addition of cyclosporin A, a drug that disrupts the interaction between capsid and cyclophilin restores the ability of the mutant capsid to infect DCs.

The proline residue at position 88 of the HIV-2 capsid protein (SEQ ID No: l) is the substrate of the peptidylprolyl isomerase activity of cyclophilin A. Thus, typically, according to the invention, the mutant capsid protein has a P at position 88. In other words, it does not comprise a mutation at residue 88. The inventors have observed that mutants that do not comprise a P at position 88 generally do not bind cyclophilin A.

In one embodiment, the mutant capsid protein comprises the amino acid sequence as set forth in SEQ ID No.2.

SEQ ID No:2 corresponds to the sequence of the cyclophilin-binding loop of the P86HA mutant, compared to the sequence of the capsid protein of HIV-2 ROD isolate. This mutant has been shown to be particularly effective in activating DCs without infecting them.

In another embodiment, the mutant capsid protein comprises the amino acid sequence as set forth in any one of sequence SEQ ID No:3 to SEQ ID No:7. These sequences correspond to the cyclophilin-binding loop of other isolates, in which the P86 residue has been replaced by residues HA. Table 1 : Examples of cyclophilin-binding loop sequences according to the invention

In another embodiment, the mutant capsid protein comprises the amino acid sequence as set forth in any one of sequence SEQ ID No:53 to SEQ ID No:61 These sequences correspond to the cyclophilin-binding loop which are further illustrated in the Example section.

Table 2: Further examples of cyclophilin-binding loops according to the invention

Wild- Genbank SubSequence of mutant Mutant Mutant name type accession type the cyclophilin binding SEQ

Isolate number loop ID

of

original

strain

HIV-2 X05291 A HPIRAGPLPAGQLREPR No :53 P86RA ROD

HIV-2 X05291 A HPIQAGPLPAGQLREPR No :54 P86QA ROD

HIV-2 X05291 A HPIAAGPLPAGQLREPR No: 55 P86AA ROD

HIV-2 X05291 A HPIAMGPLPAGQLREPR No: 56 P86AM ROD

HIV-2 X05291 A HPIHVGPLPAGQLREPR No: 57 P86HV ROD

HIV-2 X05291 A HPIPIGPLPAGQLREPR No 58 P86PI

ROD

HIV-1 M19921 B HPVHAGPIEPGQMREPR No 59 A92E

NL4- 3

HIV- M19921 B HPVHAGPIAPDQMREPR No 60 G94D

1

NL4-3

HIV- M19921 B HPIHAGPLPAGQLREPR No 61 V86I- 1 IAP91LPA-

NL4-3 M96L

HIV-2 X05291 A HPIHAAPLPAGQLREPR No:62 PG86HAA ROD

Capsids comprising a cyclophilin-binding loop having the sequence as set forth in SEQ ID NO:62 are disclosed as comparative examples. In one embodiment of the invention, the mutant capsid protein consists in the amino acid sequence as set forth in SEQ ID No: 8 (which corresponds to the HIV- 2 mutant capsid P86HA, see examples).

In another embodiment, the mutant capsid protein comprises the amino acid sequence as set forth in any one of sequence SEQ ID No:2 to SEQ ID no:7 and SEQ ID NO:53 to SEQ ID No:61 in the context of a SIVmac239 backbone instead of HIV-2.

In one embodiment of the invention, the capsid protein consists in the amino acid sequence as set forth in SEQ No:9.(SIVmac239 QP APQQ85 IH AGPLP A mutant, see examples).

In another embodiment, the mutant capsid protein comprises the amino acid sequence as set forth in any one of sequence SEQ ID No:2 to SEQ ID no:7 and SEQ ID NO:53 to SEQ ID No:61 in the context of a HIV-1 backbone instead of HIV-2. According to the invention, the non-capsid components of the particle can be derived from HIV-2, HIV-1, or SIVmac239,or any other retrovirus or lentivirus, provided that they also contain a Vpx protein. Other proteins may be required for the retroviral particle to induce DC activation. These components may be part of the replication-defective retroparticle itself (i.e. these components act in "cis"), or they may delivered by another particle by trans- complementation (i.e. they act in "trans"). According to one embodiment of the invention, the replication-defective particle also comprises one or several of the following proteins: Gag, Pol, Vpx and an envelope protein that allows delivery of the viral core to dendritic cells.

Without wishing to be bound by theory, it is believed that Vpx allows to bypass the block to reverse-transcription in myeloid cells. This block is mediated at least by the cellular protein SAMHD1 (Laguette et al., Nature, 2011, PMID 21613998). Vpx inhibits SAMHD1 by inducing the degradation of SAMHD1.

In one embodiment, the replication-defective retroparticle of the invention further comprises the Vpx protein.

In another embodiment, the replication-defective retroparticle of the invention does not comprise the Vpx protein. In this case. Vpx can be introduced into the dendritic cells by other means than that the viral particle containing the mutated capsid. For example, Vpx can be introduced by trans-complementation, before or during exposure to the immunogenic capsid (Goujon et al.. Gene Therapy, 2006. 13:991. PMID 1652548 1 ). Alternatively, other methods than introducing Vpx can be used to bypass the block to reverse-transcription in myeloid cells, such as inhibition of SAMHD 1 by small molecules compounds or small interfering RNA. The invention is not limited to the delivery method of the viral core described in the Examples below through the use of a replication-defective retroparticle. Any method to deliver the viral core into the cellular cytoplasm known to those skilled in the art is possible, including liposomes, chemical treatment, electroporation or fusogenic envelope proteins.

Suitable envelope proteins according to the invention include, but are not limited to the G glycoprotein from the Vesicular Stomatitis Virus (VSV), or an HIV envelope protein or any lenti viral or retroviral envelope protein.

According to one embodiment of the invention, the replication-defective retroparticle comprises an envelope protein selected from the group consisting of the G glycoprotein from VSV, the HIV-1 envelope protein and the HIV-2 envelope protein.

In one aspect, the invention relates to any means for delivering a mutant capsid protein as defined above to a dendritic cell wherein said capsid binds to cyclophilin A and wherein said mutant capsid protein is capable of activating dendritic cells (DCs) without infecting said DCs.

Said means for delivering a mutant capsid protein according to the invention may thus be used in any of the methods, composition or use described herein and relating to a replication-defective retroparticle according to the invention. In one embodiment of the invention, the replication-defective particle also comprises a single-stranded ribonucleic acid molecule which can, upon entering the dendritic cell, be reverse-transcribed to a double- stranded desoxyribonucleic acid molecule. In one embodiment, the replication-defective particle comprises a single- stranded ribonucleic molecule having a "psi" encapsidation sequence. The "psi" signal refers to a stretch of nucleotides in the viral RNA molecule that allows its loading, from the cellular cytoplasm, into the core of the particle. This sequence has been defined as ACAAACCACGACGGAGTGCTCCTAGAAAG (SEQ ID No:20) in Griffin et al, Journal of Virology, 2001, 75: 12058.

Alternatively, in another embodiment of the invention, the replication-defective particle does not comprise a single- stranded ribonucleic acid molecule. However, such a single- stranded ribonucleic acid molecule can be delivered to the exposed dendritic cell by a separate particle, to which the DC are simultaneously exposed (a process referred to as "trans-complementation").

In one embodiment of the invention, the replication-defective particle also comprises mutations in other proteins than the capsid protein and which are known to inactivate the retrovirus of interest, such as the HIV virus, including but not limited to, mutations in the conserved enzymatic residues of the integrase sequence, which can be mutation of D64, D116, El 52 or mutations in the Long Terminal Repeat region of the viral nucleic acid that disrupt the irreversible integration capacity (see Leavitt et al., JBC, 1993, 268:2113, PMID 8420982 and Esposito et al., EMBOJ, 1998, 17:5832, PMID 9755183).

The invention also relates to a vector comprising a nucleic acid encoding a replication-defective particle as defined above.

In one embodiment, said vector essentially comprises the nucleic acid molecule as set forth in SEQ ID No:21 (encoding a replication-defective HIV-2 retroparticle) or SEQ ID No:22 (encoding a replication-defective SIV retroparticle).

The inventors have found that, surprisingly, the mutated capsid protein according to the invention, when incorporated into a replication-defective particle as defined above, was able to activate dendritic cells in the absence of infection and of subsequent viral production. These criteria meet the requirements for an adjuvant.

Thus, the invention also relates to a pharmaceutical composition comprising a mutated capsid protein, a replication-defective particle and/or vector as defined above and a pharmaceutically acceptable carrier or excipient.

Suitable pharmaceutical compositions can be formulated for delivery by oral, nasal, transdermal, parenteral, or other routes by standard methodology. In this respect, the excipient can include any suitable excipient (e.g., lubricant, diluent, buffer, surfactant, co-solvent, glidant, etc.) known to those of ordinary skill in the art of pharmaceutical compounding (see, e.g., "Handbook of Pharmaceutical Excipients" (Pharmaceutical Press), Rowe et al., 5th Ed. (2006)). In one embodiment, said pharmaceutical composition is an adjuvant.

In another embodiment, said pharmaceutical composition is an immunizing composition. In one embodiment, said pharmaceutical composition further comprises another adjuvant. Said other adjuvant may be any pharmaceutically acceptable adjuvant known in the art. Suitable adjuvants include, but are not limited to metal salts and in particular aluminium hydroxide salts; oils; squalene, Freund's adjuvant, water emulsions, Toll-like receptor ligands and in particular CpGs; GM-CSF; saponins etc.

The replication-defective particle and/or vector and/or pharmaceutical composition of the invention may be used in order to induce an innate immune response, by activation of dendritic cells, toward any type of antigen. This innate immune response may serve in protecting the organism against any type of disease, including, but not limited to: any type of infectious diseases including viral infections, bacterial infections, fungal infections and parasitic infections; allergy; cancer, autoimmune diseases etc. The invention also relates to a mutated capsid, a replication-defective particle and/or a vector and/or a pharmaceutical composition as defined above for use in a method of treatment.

In particular, the agents of the invention can be useful as therapeutic vaccines in patients who are infected with HIV, and as anti-HIV prophylactic vaccines in healthy subjects. The agents of the invention can also be useful in immunotherapies of cancer.

In one aspect, the invention relates to a replication-defective particle and/or a vector and/or a pharmaceutical composition as defined above for use in a method for treating viral infections, particularly lentiviral infections, more particularly retroviral infections, even more particularly HIV infection.

The invention also relates to a method for treating a subject having or at risk of having a viral infection, particularly a lentiviral infection, more particularly retroviral infection, even more particularly HIV infection comprising the step of administering to said patient a therapeutically effective amount of a replication- defective particle and/or a vector and/or a pharmaceutical composition as defined above.

In another aspect, the invention relates to a replication-defective particle and/or a vector and/or a pharmaceutical composition as defined above for use in a method for treating cancer.

The invention also relates to a method for treating a subject having or at risk of having a cancer comprising the step of administering to said patient a therapeutically effective amount of a replication-defective particle and/or a vector and/or pharmaceutical composition as defined above. The invention further relates to an in vitro method for inducing the activation of dendritic cells in the absence of infection comprising the step of exposing said DC to a replication-defective particle and/or a vector and/or a pharmaceutical composition as defined above.

In one aspect, the invention also relates to a polypeptide comprising the amino acid sequence as set forth in SEQ ID No:2 (mutant cyclophilin-binding loop of the mutant capsid protein HIV-2 P86HA), or any of SEQ ID No:3 to 7 and 53 to 61. The invention also relates to a polypeptide consisting essentially of, or consisting of the sequence as set forth in SEQ ID No:8 (HIV-2 mutant capsid P86HA) or SEQ ID No:9 (SIVmac293 mutant capsid QP APQQ85 IH AGPLP A) or SEQ ID No:61 (HIV-1 mutant capsid V86I-IAP91LPA-M96L). FIGURE LEGENDS

Figure 1. Genetic organization of HIV-1 and HIV-2. HIV-1 specifically encodes vpu. while HIV-2 specifically encodes vpx. CA indicates the vi al capsid. Figure 2. Location of the Cyclophilin A (CypA) binding peptide bond in HIV - 1 and HIV -2 capsid proteins. Numbering refers to the first amino acid of the capsid subunit.

Figure 3. Mutations and chimeras introduced in HIV-1 and HIV-2 capsids to alter CypA affinity. 1: HIV-1 WT; 2: HIV-1 P90A; 3: HIV-1 HA87P; 4: HIV-1 loops binding CypA HIV-2; 5: HIV-2 WT; 6: HIV-2 G87A 7: HIV-2 P86PA: 8: HIV-2 P86HA; 9: HIV-2 loops binding CypA HIV- 1 .

Figure 4. V iral titrations on GHOST cells indicates that HIV -2 P86HA generates functional particles. Viral supernatants were produced by transient transfection of 293FT cells with each viral vector plasmid in combination with a VSV-G expression plasm id. Supernatants containing viral particles were titrated on GHOST cells according to standard procedures.

Figure 5. HIV-2 P86HA is not infectious in primary human monocyte- derived dendritic cells (MDDCs). Viral supernatants were used to expose MDDCs obtained at day 4 of differentiation in the presence of GM-CSF and IL-4 starting from adult blood CD 14+ monocytes. GFP expression was measured 48 hours later by flow cytometry. Serial dilutions of viral supernatants were used and plotted according the MOI calculated from GHOST cells. GFP (green fluorescent protein ) is encoded instead of the net^" gene of HIV-2 and thus acts as a reporter of productive infection.

Figure 6. HIV-2 P86HA activates MDDCs in the absence of productive infection. Day 4 MDDCs were exposed to HIV-2 WT or P86H A viral supernatants. 48 hours later GFP and CD86 expression were measured by flow cytometry. Upregulation of CD86 is a hallmark of MDDCs activation (also referred to as maturation in human DCs ).

Figure 7. Cyclophilin A is required for MDDCs activation by HIV -2 1*86 II A and for blocking its infection. Experiment was carried as previously described. At the time of exposure. 2 μΜ Cyclosporin A (CsA) was added (bottom part ). CsA disrupts the interaction between CypA and the viral capsid. In the presence of CsA, HIV-2 P86HA infection is restored and its abil ity to mature MDDCs is suppressed.

Figure 8. Vpx is required for MDDCs activation by H IV -2 P86H .

Experiment was carried as previously described. Vpx was disrupted by mutating its initiation codon. At the time of exposure. 2 μΜ Cyclosporin A (CsA) was added (bottom part ). To restore vpx. treatment was trans-complemented in indicated wells by co-exposure with SIVVLP(G ) particles. SIVVLP(G ) are SIVmac25 1 virus-like particles pseudotyped with VSV-G that do not encapsidate genomic RNA and thus provide a mean to introduce the vpx protein in cells. We have previously shown that SIVVLP(G ) do not activate MDDCs.

Figure 9. Reverse transcriptase activity is required for MDDCs activation by HIV -2 P86HA. Experiment was carried as previously described. Vpx was disrupted by mutating its initiation codon. At the time of exposure. 2 μ Μ Cyclosporin A (CsA) and/or 25 μ Μ Azidothymidine (AZT) was added as indicated. Figure 10. Reverse transcriptase activity is required for MDDCs activation by HIV-2 P86HA: viral titration. Experiment was carried as previously described except that different vi al doses were used. Multiplicity of infection is based on GHOST titration. Figure 11. HIV-2 P86HA induces type I interferon production by MDDCs. Experiment was carried as previously described. MDDCs were also treated with a preparation of poly(I:C) (pIC), a known adjuvant that induces type I I F expression. 48 hours after exposure, MDDCs supematants were harvested and type I I F production was measured using a the HL1 1 6 bioassay.

Figure 12. Genomic RNA is required for MDDCs activation by HIV-2 P86HA. Experiment was carried as previously described. To prevent genomic RNA incorporation, the encapsidation signal ("psi") of HIV-2 RNA was removed

("DM" mutation according to PMID 11711596) resulting in the Λ psi mutation. At the time of exposure. 2 μΜ Cyclosporin A (CsA ) was added (bottom part).

Figure 13. Sequence of SIVmac239 capsid sequence of the corresponding P86HA mutation. 1 : HIV- 1 WT; 2: HIV-2 WT; 3: HIV-2 P86HA: 4: SIVmac239 WT; 5: SIVmac239 QP APQQ85 IH AGPLP A

Figure 14. Introduction of the QPAPQQ85IHAGPLPA mutation in SIV mac239 results in a similar phenot pe as HIV-2 P86HA. VSV-G pseiidotyped SIVmac239 supernatants were produced by 293 FT transfection. The e periment was carried out as previously described for H IV-2. SIVmac239 WT is less infectious on human cells that HIV-2. SIVmac239 QP APQQ85 IH AGPLP A is not infectious and activates MDDCs. Addition of CsA restores infection by STVmac239 QP APQQ85 IH AGPLP A and blocks activation.

Figure 15. HIV-2 P86HA vector and protein sequences encoded by said vector.

Figure 16. SIV mac239 QPAPQQ85IHAGPLPA vector and protein sequences encoded by said vector.

Figure 17. HIV-2 P86HA generates viral cDNA, but does not generate 2LTR circles viral DNA and does not integrate in the target cell genome. MDDCs were infected with HIV-2 WT or HIV-2 P86HA in the absence of the presence of CsA. AZT or Raltegravir. alone or in combination. Total cellular DNA was harvested 24 hours after infection. Quantity of late RT viral cDNA, 2LTR circles viral DNA and integrated viral DNA relative to beta-blogin DNA was measured by real-time PGR. H IV-2 P86HA produces AZT-sensitive viral cDNA. but does not lead to detectable 2LTR circles or to integrated DNA.

Figure 18. DC activation by HIV -2 WT and HIV -2 P86HA is maintained in the absence of integration. (A) MDDCs were infected with H IV-2 WT or HIV-2 P86HA in the absence or the presence of CsA. AZT or Raltegravir. SIVVLP(G ) were also added to increase infectivity. 48 hours later. GFP and CD86 expression were analysed. (B) Quantification of experiments performed as in (A) in 4 independent donors. ** p < 0.01 ; * p < 0.05. Figure 19. DC activation by HIV -2 WT and HIV -2 P86HA is maintained when the integrase catalytic site is disrupted. (A) MDDCs were infected with HIV-2 WT, HIV-WT D116A, HIV-2 P86HA, or HIV-2 P86HA D116A. SIVVLP(G ) were also added to increase infectivity. 48 hours later. GFP and CD86 expression were analysed. (33) Quantification o experiments performed as in (A) in 6 independent donors.

Figure 20. Activation before integration by SIVmac239 is conserved in macaque DCs. MDDCs from macaques were infected with SIVmac239 WT or SrVmac239 QPAPQQ85IHAGPLPA, alone or in the presence of CsA or RAL. 48 hours later. GFP and CD 86 expression were analysed. Data is representative of 4 independent macaque samples.

Figure 21. Activation before integration is conserved in HIV-1 when the cyclophilin-binding loop is capsid is engineered as in HIV-2 P86HA. (A) MDDCs were infected by H IV- 1 WT or HIV- 1 V86I-I AP 1 LPA-M96L. alone or in the presence of SIVVLP(G ) and/or CsA. GFP and CD86 expression were analysed 48 hours later. Data is representative of 3 independent donors.

Figure 22. Activation before integration is also obtained with HIV -1 mutant described to have a CsA -dependent infectivity. MDDCs were infected by H1V- 1 WT. HIV- 1 A92E or HIV- 1 G94D. alone or in the presence of SIVVLP(G ). CsA and/or RAL. GFP and CD86 expression were analysed 48 hours later. Data is representative of 3 independent donors.

Figure 23. Mutations P86RA, P86QA, P86AA, P86AM, P86IIV and P86PI leads to CsA-dependent infectivity of HIV -2 on GHOST cells. Titer of H IV-2 WT. P86I IA. P86RA. P86QA. P86.AA. P86AM, P86HV and P86PI was determined on GHOST cells in the presence of the absence of CsA. Data is shown in infectious units per ml. Data is an average of 3 independent viral productions.

Figure 24. DC infection and activation by HIV -2 P86RA, P86QA, P86AA,

1*86 AM, P86HV and P86PI. (A) MDDCs were infected by HIV-2 WT. P86HA. P86RA. P86QA. P86AA. P86AM, P86HV and P86PI. in the absence or presence of CsA or RAL. SIVVLP(G) were also added to increase infectivity. GFP and CD86 expression were analysed 48 hours later. (B) Quantification of experiments performed as in (A) in 4 independent donors. Virus dilutions used were absolute volumes of 100 μ ΐ (10²), 33 μ ΐ (10^L5) and 10 μ ΐ (10¹).

Figure 25. DCs are not activated by HIV -2 PG86HAA but are infected. (A)

Viral particles of HIV-2 WT. HIV-2 P86HA and HIV-2 PG86HAA were analysed by western blotting for Gag and CypA. (B) Quantification of the ratio of viral CypA to viral caps id within particles based on western blot analysis. (C) MDDCs were infected by H IV-2 WT. H IV-2 P86HA or HIV-2 PG86HAA in the absence of the presence of CsA. GFP and CD86 expression were analysed 48 hours later. Data representative of 3 independent donors. Figure 26. P86HA mutation in HIV -2 leads to increased affinity to CypA. (A)

293FT cells were transfected with plasmid coding for HIV - 1 . HIV-2 WT and H IV-2 P86HA. 48 hours later, supernatants were harvested, filtered at 0.45 μ Μ and centrifugated to recover the virus pellet. Virus pellet and cel ls were lysed and analysed by SDS-PAGE followed by western blotting against Gag and CypA simultanesouly (Cells, top panel ). total act in (Cells, bottom panel ). Gag alone (Virus, top panel ) or CypA alone (Virus, bottom panel ). Pr55 and p24/27 of Gag are indicated. (B) Quantification of the ratio of viral CypA to viral capsid within particles based on western blot analysis. The invention may be better understood by reference to the following non- l imiting examples, which are provided as exemplary of the invention. The following examples are presented in order to more fully illustrate the preferred embodiments of the invention and should in no way be construed, however, as l imiting the broad scope of the invention. EXAMPLES

Methods Constructs.

All amino acids numbering refers to the first amino acid of the capsid sub- unit of Gag.

"HIV- 1 WT^" is a GFP reporter H IV- 1 vector derived from the molecular clone NL4-3. It has mutations env vpu vpr vif net^" and encodes GFP in net^*.

"HIV-2 WT^" is a GFP reporter H IV-2 vector derived from the molecular clone ROD9. It has mutations env net^" . w ith the GFP open reading frame in place of nef.

"HIV- 1 P90A" is derived from HIV- 1 WT by mutation of residue P at position 90 in capsid to A.

"HIV- 1 HA87P" is derived from HIV- 1 WT by mutation of residue H at position 87 in capsid to P and by removing residue A at position 88.

"HIV- 1 A92E" is derived from H IV- 1 WT by mutation of residue A at position 92 in capsid to E.

"HIV- 1 G94D" is derived from H IV- 1 WT by mutation of residue G at position 94 in capsid to D.

"HIV- 1 V86I-IAP91 LPA-M96L^" is derived from HIV- 1 WT by mutation of residue V at position 86 in capsid to I. of residues IAP at position 91 in capsid to LPA and of residue M at position 96 in capsid to L.

"HIV-2 G87A" is derived from HIV-2 WT by mutation of residue G at position 87 in capsid to A.

"HIV-2 P86PA^" is derived from HIV- WT by inserting residue A at position 87 in capsid.

"HIV-2 P86HA^" is derived from HIV-2 WT by mutation of residue P at position 86 in capsid to H and by adding residue A at position 87.

"HIV-2 WT Dl 16 A" is derived from HIV-2 WT by mutation of residue D at position I 16 in integrase to A. and by adding residue A at position 87. "HIV-2 P86HA D116A" is derived from HIV-2 P86HA by mutation of residue D at position I 1 6 in integrase to A.

"HIV-2 P86RA" is derived from HIV-2 WT by mutation of residue P at position 86 in capsid to R and by adding residue A at position 87.

"HIV-2 P86QA" is derived from HIV-2 WT by mutation of residue P at position 86 in capsid to Q and by adding residue A at position 87.

^•'HIV-2 P86AA" is derived from HIV-2 WT by mutation of residue P at position 86 in capsid to A and by adding residue A at position 87.

"HIV-2 P86AM" is derived from HIV-2 WT by mutation of residue P at position 86 in capsid to A and by adding residue M at position 87.

"HIV-2 P86HV^" is derived from HIV-2 WT by mutation of residue P at position 86 in capsid to H and by adding residue V at position 87.

"HIV-2 P86PI" is derived from H IV-2 WT by inserting residue I at position 87 in capsid.

"HIV-2 PG86HAA^" is derived from HIV-2 G87A by mutation of residue P at position 86 in capsid to H and by adding residue A at position 87.

"HIV- 1 loops binding CypA HIV-2^" was derived from HIV- 1 WT by exchanging the residues 68 to I 1 8 in capsid with the residues 67 to 1 16 from HIV-2 WT.

"HIV-2 loops binding CypA HIV-1" was deri ved from H IV-2 WT by exchanging the residues 67 to 1 16 in capsid with the residues 68 to 1 1 8 from HIV- 1 WT.

The "Δνρχ" mutation in "HIV-2 WT" and "HIV-2 P86HA" was generated by disrupting the start codon of vpx.

The "APsi" mutation in "HIV-2 WT^" and "HIV-2 P86HA^" was generated by deleting the genomic RNA encapsidation signal corresponding to the DNA sequence ACAAACCACGACGGAGTGCTCCTAGAAAG (SEQ ID No:20).

"SIVmac239 WT" is a GFP-reporter virus derived from the molecular clone SrVmac239. It encodes IRESGFP in net^*, and has mutations env net^" .

"SIVmac239 QP A PQQ851 H AGPLP A^" is a capsid-mutant of the parental

SrVmac239 WT vector, for which the capsid sequence QPAPQQ starting at position 85 has been mutated to HAGPLPA.

"CMV-VSVG" is a CMV-dnven expression vector coding for the vesicular stomatitis v irus G glycoprotein.

"pSIV3+" is a helper construct derived from the SIVmac251 molecular clone and has mutations psi env nef .

AH mutants were generated by overlapping PCR-mutagenesis using standard molecular biology methods. All plasmid D A was prepared with Invitrogen Hi Pure plasmid kit. Plasmid DNA did not induce dendritic cell maturation, and viral-producing cells were washed after DNA transfection.

Cells.

GHOST (GHOST X4R5), 293FT and HL l 16 cells were cultured in DM EM.

10% fetal bovine serum (FBS) (Gibco ). antibiotics and 2% HAT for HL l 1 6 cells.

Peripheral blood mononuclear cells were isolated from buffy coats from normal human donors using Ficoll-Paque PLUS. Macaque mononuclear cells were isolated from peripheral normal macaque blood using 90% Ficoll-Paque PLUS.

Human and macaque CD I 4 ^' cells were isolated by a positive selection with anti- human CD 14 magnetic beads ( M iltenyi ). Purity was at least 99%. CD I 4 ^' cells were cultured in RPMI medium. 10% FBS ( Biowest ). antibiotics and HEPES in the presence of recombinant human GM-CSF at 1 0 ng/ml and IL-4 at 50 ng/ml.

Fresh media was added at day 3, and cells were stimulated or infected at day 4.

Exposure to viral supernatants.

At day 4 of MDDC differentiation, cells were harvested, counted and resuspended in their own media at a concentration of one million per ml with 5 mg/ml. polybrene. and 100 μΐ was al iquoted in round-bottomed 96-well plates. For e posure. 50 μ ΐ of media or SIVVLP(G ) was first added. One-hundred microl itres of media or dilutions of various HIV- 1 -derived viral preparations were then added. Azidothymidine (AZT). Cyclosporin A (CsA) or Raltegravir (RAL ) were added respectively at 24 μΜ, 2 μΜ and 20 μ Μ.

48 hours after exposure, cel l culture supernatants were harvested and u Itraviolet-i rrad i ated to inactivate free vi rus. .Supernatants were subsequently used for a quantitative bioassay for IFNs. Cell-surface staining of activation markers (CD86, CD83, CD38) was also performed 48 h after exposure. Stimulation and DCs activation were positively controlled by treatment with poly(I:C).

Quantitative bioassay for IFNs.

.Supernatants were assayed for I F activity using a recombinant HL I 1 6 cell line, which carries a luciferase reporter controlled by the IFN-inducible 6- 1 6 promoter. In brief, the reporter cells were exposed to cell culture supernatants for 5 h and assayed for luciferase activities (Pro mega ), which were then translated to IFN activities by using a standard curve generated from a serial di lution of human IFNalpha-2a ( ImmunoTools ).

Virus production.

Viral particles were produced by transfection of 29 FT cells with 3 g DNA and 8 μΐ Trans IT-293 (Minis Bio ); for SIVVLP(G ). 0.4 μg CMV-VSVG and 2.6 μg pSIV3⁺; for HIV- 1 . H IV-2 or SIVmac239 derived VSV-G pseudotyped vectors. 0.4 μg CMV-VSVG and 2.6 μg HIV-derived vector DNA. One day after transfection. media was removed, cells were washed out once and fresh media was added. Viral supernatants were harvested one day later and filtered at 0.45μηι. Virus titer was measured by GHOST cell titration according to standard procedures.

Real-time PGR.

Total cellular DNA was harvested using a Machery-Nagel Nucleospin

Tissue kit. Real-time PGR analysis was performed in a Roche LightCycler 480 using Roche 480 SYBR Green I master reagent in 20 μ ΐ final volume per well according to manufacturer specifications. Each sample was measured in triplicate for all primers. For beta-blogin. primers were bglobin-f CCCTTGGACCCAGAGGTTCT (SEQ ID No:43 ) and bglobin-r CG AGC ACTTTCTTGCC ATG A (SEQ ID No:44). For Late RT. primers were hiv2-3'U3-fwd GAAGGGATGTTTTACCATTTAGTTA (SEQ ID No:45) and hiv2-psi-rev GTTCCAAGACTTCTCAGTCTTCTTC (SEQ ID NO:46). For 2LTR, primers were hiv2-3'U3-rev TAACTAAATGGTAAAACATCCCTTC (SEQ ID NO:47) and hiv2-R-fwd GTTCTCTCCAGCACTAGCAGGTA (SEQ ID NO:48). For integrated DNA two rounds of amplification were performed. For the first round, primers were alu l GCCTCCC A A AGTGCTGGG ATT AC AG (SEQ ID NO:49) and hiv2-r AAGGGTCCTAACAGACCAGGGTCT (SEQ ID NO:50). For the second round, 1 μΐ of first-round reaction was used as template, and primers were hiv2-t2 GC AGGT AG AGCCTGGGTGTTC (SEQ ID NO:51) and hiv2-r2 C AGGCGGCG ACT AGG AG AG AT (SEQ ID NO:52). Cycling conditions were lx 95°C 5'; 35x 95°C 10"-50°C 20"-72°C 30". Relative concentrations of Late RT, 2LTR and integrated viral DNA were calculated relative to beta-globin using the ACt method. Western Blotting

Cells were lysed in 80 μΕ of Lysis buffer (50 niM Tris Hcl pH8, 120 mM Nacl, 4 mM EDTA, 1% NP40 and lx EDTA-free protease inhibitors cocktail (Roche)). Lysates were cleared by centrifugation at 5500 g for 7 minutes at 4°C. Viruses supematants were filtered at 0.45 μ ιη. centrifugated at 16000 g for 1 hours 30 mill at 4°C. Virus pellets were lysed in 15 μ I of Lysis buffer. Cellular and viral protein lysates were resolved on 4%-20% Biorad precast SDS-PAGE gels and transferred on nitrocellulose membrane. Proteins were blotted with antibodies as follow: mouse anti-Gag/capsid mouse 183-H 12-5C: mouse anti-Vpx 6D2.6; mouse anti- act in MAB 1501 ; rabbit anti-CypA. ECL signal was recorded on the ChemiDoc XRS Biorad Imager. Data was analysed with the Image Lab software (Biorad ).

Results

HIV-2 was able to induce an innate immune response in dendritic cells (DCs ) while HIV- 1 was not. This has been attributed, at least, to the presence of the Vpx protein in HIV-2 (figure 1). In dendritic cells, the replication of HIV is strongly inhibited at the level of reverse transcription, and this is referred to as a restriction to repl ication. Vpx allows to overcome the restriction block to HIV reverse-transcription in DCs mediated at least by SAMHD I . a cellular protein that blocks H IV reverse-transcription in myeloid cells by inhibiting the completion of reverse-transcription (PMID 21613998). In the presence of Vpx. HIV-2 induced an innate immune response as shown by the expression of activation markers such as CD86, CD38 and CD83 and by the production of soluble type I interferon (IFN). In the presence of Vpx. HIV - 1 was able to induce such as response. This response further required productive infection of the DCs and an interaction between CypA and the viral capsid (figure 2). The affinity for CypA is different between HIV- 1 and H IV-2. suggesting that the sequence of the cyclophilin- binding loop in capsid ( region 2 in figure 3) may modulate the induction of the innate immune response. HIV- 1 is thought to bind more strongly to CypA than HIV-2. The interaction between CypA and capsid is strictly defined by residues in the cyclophilin-binding loop (region 2), as residues of this loop make direct contacts with CypA. Capsid residues outside of the loop may also indirectly modulate the interaction with CypA. by alteri ng the overall conformation of capsid (such as region I and region 3 in figure 3). To gain more insight in our understanding of the CypA-capsid dependent induction of DCs innate immune response, we generated a number of mutant vectors (figure 3).

In HIV- 1 . the peptide bond between G89 and P90 is the substrate of CypA. Mutant HIV - 1 P90A was generated to disrupt the CypA interaction. Mutant HIV- 1 HA87P was generated to confer an interaction with CypA more similar to that of H IV-2. Mutant "HIV- 1 loops binding CypA HIV-2^"* was generated to confer an interaction with CypA identical to that of HIV-2.

In H IV-2. the peptide bind between G87 and P88 is the substrate of CypA. Mutant G87A was generated to disrupt the CypA interaction. Mutants H IV-2 P86HA and H IV-2 P86PA were generated to confer an interaction with CypA more similar to that of HIV- 1 . Mutant "H IV-2 loops binding CypA HIV- Γ^" was generated to confer an interaction with CypA identical to that of HIV- 1 .

To determine if introduction of these mutations would affect overall infectivity of the viruses, supernatants containing viral particles were titrated on GHOST cells (figure 4). HIV- 1 WT and HIV-2 WT showed titers above 10e6 infectious units per milliliter (i.u./ml). Mutants HIV-1 P90A and HIV-2 G87A showed a slightly decreased titer in the same log decade of titer, indicating that the CypA interaction modulates infectivity. Mutations HIV - 1 HA87P had a titer below 10e6 i.u/ml. "H IV- 1 loops binding CypA H IV-2^". "HIV-2 loops binding CypA HIV- ! ^" and HIV-2 P86PA had titers below 10e5 i.u/ml. indicating that these mutations affected the overall functionality of the capsid. rendering the viruses defective. Interestingly, mutant HIV-2 P86HA had a titer above 1 0e6 i.u/ml. indicating that this virus was fully functional and did not have gross defect in structure or stabi lity.

Next, we focused on comparing HIV-2 WT with H IV-2 P86HA. We treated mo n oc y te -de ri ved DCs with viral supernatants to measure their infectivity on target dendritic cells. We used GFP expression from net^' as a marker of infection. The proportion of infected dendritic cells was measured by flow cytometry and normalized as function of input MOI based on GHOST titrations (figure 5). Whi le HIV-2 WT infected dendritic cells as predicted. H IV-2 P86HA had lost its infectious abil ity on DCs. At MOI 4 of HIV-2 P86HA. the proportion of GFP- positive cells was below 1%, while at MOI 0.6 of HIV-2 WT. the proportion of GFP-positive cells was above 30%. This indicates that the P86HA has lost at least 2 logs of infectivity on DCs compared to its W counterpart.

We then measured the induction of an innate immune response in DCs by measuring the proportion of CD86-expressing cells after exposure to viral particles (figure 6). Control cells showed a background activation (9% CD86+), and HIV-2 WT infected and activated cells (41% CD86+). After exposure to HIV- 2 P86HA. activation increased to 53% while GFP remained undetectable. This indicates that HIV-2 P86HA. despite the fact that it had lost its infectivity potential on DCs. has maintained its ability to induce an innate immune response. As such. HIV-2 P86HA satisfies criteria for being an adjuvant: it is a non- infectious entity that autonomously induces an innate immune response in DCs.

The mutation P86HA presumably reinforces the interaction between CypA and capsid. To test it^' this is the case, we measured the levels of CypA in viral particles of HIV- 1 . HIV-2 WT and H IV-2 P86HA (Figure 26 ). We produced viral particles from 293FT cells. Cellular and viral proteins were analysed by western hlotting. To measure the relative affinity of CypA to the capsids. we quantified the amount of capsid and CypA in the viral supernatants by image analysis and calculated the ratio of CypA over capsid. There was more than 3-fold more CypA per capsid in HIV-2 P86HA particles compared to HIV - 1 particles, and more than 28-fold more CypA per capsid in HIV-2 P86HA compared to HIV-2 WT particles. Intracellular CypA levels were not affected by the capsid sequences. To test if this increased interaction of P86HA capsid to CypA is indeed impl icated, we treated DCs at the time of infection with cyclosporine A (CsA), which disrupts the CypA- capsid interaction (figure 7). Upon treatment with CsA. DCs exposed to HIV-2 P86HA showed a decreased activation (23% with CsA compared to 53% without CsA ) and GFP expression could now be detected (53% GFP+). CsA did not show an effect on H IV-2 WT in this particular donor at this given high MOI. Thus the interaction between CypA and H IV-2 P86HA capsid drove an innate immune response and inhibits infection, and removing CypA by CsA treatment restored infectivity and diminished the extent of innate immune activation. It is thus shown that CypA acts on HIV-2 P86HA by disrupting the infectivity feature of the capsid. while enhancing its innate activating features.

Vpx is required for innate immune activation of dendritic cells by HIV- 1 and HIV-2. Vpx removes a restriction blocked at the level of reverse transcription imposed by SAMHD1. In the absence of Vpx. genomic RNA is not properly reverse-transcribed in DNA. To determine if Vpx was also required for activation by HIV-2 P86HA. we di srupted the vpx gene (figure 8). In the absence of vpx. HIV-2 WT Λ vpx and HIV-2 P86HA Δνρχ did not induce CD86 expression (8% CD86+ in control. 12% in HIV-2 WT Δνρχ and 10% CD86+ in H IV-2 P86HA ). To determine if the loss of activation and infection was indeed due to the lack of vpx. we complemented each mutant with SIVVLP(G ) which carry the vpx protein from SIVmac25 1 . In the presence of SIVVLP(G ). the abil ity of H IV-2 WT to infect and activate DC was restored (15% CD86+. 28% GFP+) and the abil ity of HIV-2 P86HA to activate without infection was also restored (35% CD86+, no GFP). Treatment with CsA decreased CD86 expression for each mutant, and restored infectivity for H IV-2 P86HA Δνρχ in the presence of SIVVLP(G) (32% GFP+). This confirmed that Vpx and CypA are required to block infectivity of H IV-2 P86HA and to induce innate immune activation.

The requirement for Vpx suggested that reverse transcription (RT) was requi ed. To test this possibility, we exposed DCs to HIV-2 WT and HIV-2 P86HA with CsA and/or AZT. an inhibitor of RT (figure 9). HIV-2 P86HA activated DCs in the absence of treatment (59% CD86+). As previously shown, in the presence of CsA. activation was decreased and GFP was detected. In the presence of AZT. activation was inhibited (59% to 8%). This observation was confirmed over a range of viral titration ( figure 10).

Innate immune activation of DCs by HIV- 1 and HIV-2 is marked by the expression of CD86 and by the expression of soluble type I IFN. To determine if HIV-2 P86HA also induced a type I IFN response, we measured the concentration of active IF secreted in the supernatant of DCs over the course of 48 hours after exposure (figure 11). Poly( I:C (-treated DC secreted more than 200 U/ml of IFN and served as a positive control. Unexposed DCs did not secrete IFN. In this particular donor. IFN could not be detected after exposure to HIV-2 WT. After exposure to HIV-2 P86HA. 2 U/ml of IFN were detected. CsA-treatment abrogated this induction type I IFN. indicating that HIV-2 P86HA induced a type I IFN response through its interaction with CypA.

A requirement for Vpx and reverse transcriptase activity in HIV-2 P86HA- induced innate immune response suggested that viral RNA may be required. To test this hypothesis, we deleted the encapsidation signal ("APsi" mutation ) (figure 12). Exposure of DCs to HIV-2 WT and HIV-2 P86HA did not lead to CD86 upregulation. Importantly, residual infectivity of the APsi could be observed by the expression of GFP in both HIV-2 WT and H IV-2 P86HA in the presence of CsA. suggesting that the APsi mutation had not affected overall viral particle structure.

Next, since innate immune activation of DCs was obtained with the P86HA in the absence of apparent infection (no GFP expression ), we tested whether integration was required. First, we measured by quantitative PGR the amount of viral DNA ( Late RT products ). 2LTR viral DNA circles (hallmark of nuclear entry) and integrated DNA (Figure 17), in the presence or absence of AZT (inhibitor of reverse-transcription ) or RAL (Rait or Raltegravir. inhibitor of integrase ). In the case of HIV-2 WT infection. Late RT. 2LTR circles and integrated DNA could be detected as expected. In the case of P86HA. Late RT was detected to similar amounts as HIV-2 WT. However, the amounts of 2LTR circles and integrated DNA were reduced to levels equivalent to the WT virus treated with RAL or AZT. This indicated that innate immune activation of DCs by the P86HA mutant occurred in the absence of viral integration and in the absence of 2LTR circles formation. This suggests that viral DNA is absent form the nucleus of DCs and remains in the cytosol.

Next, we directly tested whether integration was required for innate immune activation by the P86HA mutant. We treated DCs with CsA. AZT or RAL at the time of infection (Figure 18). In accordance with the real-time PGR results. RAL prevented GFP expression by H IV-2 WT. However. CD86 expression was not inhibited by RAL treatment of HIV-2 WT and HIV-2 P86HA. To confirm these results, we also generate mutant viruses in which integrase was rendered catalytically inactive by mutation residue Asp 1 1 6 to Ala. Accordingly. HIV-2 WT D116A was not infectious but maintained the ability to activate DCs ( Figure 19). Similarly. HIV-2 P86HA D116A remained non-infectious in DCs. like HIV-2 P86HA. and maintained the ability to activate DCs. In conclusion, these results demonstrate that innate activation of DCs is obtained in the absence of integration of the viral DNA upon infection by HIV-2 P86HA. Furthermore, inactivation of the integrase of HIV-2 WT maintains the ability to activate DCs. though to a lesser extent than HIV-2 P86HA.

The mutation P86HA in HIV-2 capsid represents one possibil ity of mutation that confers increase binding of capsid to Gyp A. In addition to residues HA in HIV- 1 capsid. different residues have been described and can be found in the HIV sequence databases. We thus tested whether other mutation at position P86 in HIV-2 capsid would also possess the abil ity to activate an innate response without infecting dendritic cells. We generated mutations P86RA. P86QA. P86AA. P86AM, P86HV. P86PI. Titers on GHOST cells were increased for all mutant capsids upon treatment with CsA. similarly to P86HA (Figure 23). In DCs. all mutants activated DCs in the absence of infection, similarly to P86HA. except P86RA that showed residual infectivity. Treatment with CsA restored infectivity of H IV-2 WT. P86I IA. P86AM, P86HV. P86PI. and to a lesser extent HIV-2 P86RA. P86QA and P86AA. In all cases. DC activation was inhibited by CsA (Figure 24 ). Treatment with Raltegravir inhibited infection by H IV-2 WT and by HIV-2 P86RA. but activation was maintained in all conditions. Overall. P86RA. P86QA. P86AA. P86AM, P86HV and P86PI activated similarly or better than P86HA (relative to their respective GHOST titer).

Next, we sought to determine if any virus with the mutation P86HA was able to activate DCs. or if the mutant virus could also be generated that had lost the ability to activate DCs. We thus generated the mutant PG86I IAA. in which Glycine at position 87 is replaced with an Alanine, thus disrupting the CypA binding site in capsid. Indeed. HIV-2 PG86HAA incorporated a decreased amount of CypA per capsid compared to HIV-2 P86HA (Figure 25). Upon DC infection. HIV-2 PG86HAA induced less activation than HIV-2 WT and H IV-2 P86HA. In addition infection was not blocked, unl ike HIV-2 P86HA. Furthermore, whi le treatment with CsA restored infectivity of H IV-2 P86HA. it did not affect infectivity of PG86HAA. Thus, not all mutants of capsid at position P86 induce DC activation in the absence of infection.

HIV-2 is related to SIVmac239. In the cyclophilin-binding loop. SrVmac239 has a different sequence from H IV-2 (figure 13). We wished to determine whether engineering a SIVmac239 with an analogous cyclophilin- binding loop to H IV-2 P86HA would lead to a SIVmac239 virus with simi lar property. We thus mutated the cyclophilin-binding loop of SIVmac239 WT to the HIV-2 P86HA mutations (QPAPQQ at position 85 to IHAGPLPA ). We used a GFP-reporter SIVmac239 virus, encoding IRES-GFP in nef. with a VSV-G pseudotyping envelope. We exposed DCs to HIV-2 P86HA. SIVmac239 W and SIVmac239 QP APQQ85 IH AGPLP A (figure 14 ). In control cells, a background activation of 2% was observed. Upon e posure to H IV-2 P86HA. an innate immune response was induced leading to 27% CD86+ cells. As previously shown, in the presence of CsA. CD86 expression was not induced and the virus now became infectious (63% GFP+ cells ). SIVmac239 WT is not well adapted to human cells, and accordingly only. 9% GFP+ cells were observed, expressing 25% CD86. In the presence of CsA. infection slightly increased to 6% and CD86 expression was inhibited. SIVmac239 QPAPQQ85IHAGPLPA was not able to infect DC and still lead to 9% CD86-ex pressing cells. In the presence of CsA. the virus infected 9% of the cells and CD86 expression was inhibited. Thus. SIVmac239 QP APQQ85 IH AGPLP A has lost i nfectivity while maintai ning its abil ity to induce an innate immune response. Thi s was not due to a defect in the viral particle preparation since addition CsA lead to actually more GFP+ cells that SIVmac239 WT. Next, we asked if the similar mechanism of activation was present in DCs from macaques instead of humans. We isolated monocytes from macaque peripheral blood and generated macaque m o n oc y te-de ri ved DCs. We infected the macaque DCs with SIVmac239 WT and SIVmac239 QPAPQQ85 IHAGPLPA and measured GFP expression and DC activation 48 hours later ( Figure 20 ). The viruses induced activation of the DCs as shown by upregulation of CD86. However SIVmac239 QPAPQQ85 IH AGPLP A was not infectious as shown by the absence of GFP expression. As in human DCs. CsA treatment restored infectivity of SIVmac239 QP A PQQ851 H AG PLP A and also reduced DC activation by SIVmac239 WT and SIVmac239 QP A PQQ851 H AG PLP A . Also similar to human DCs. inhibition of integration by treatment with Raltegravir blocked infection of SIVmac239 WT but preserved DC activation by SIVmac239 WT and SIVmac239 QPAPQQ85IHAGPLPA. Thus, similar to HIV-2 and to human DCs. the caps id mutation QP A PQQ851 H AG PL PA in the cyclophilin-binding loop in SIVmac239 leads to generation of noninfectious viral particles that are able to activate DCs. In addition, blocking integration of the WT virus maintains activation.

HIV- 1 was previously shown to activate DCs in the presence of Vpx. but only after integration and productive infection. We noted that the cyclophil in- binding loop sequence in capsid was different between HIV- 1 and H IV-2 P86HA. Thus, we asked if the strategy to activate DCs before integration could be generalized, by converting a wild-type HIV - 1 virus to a virus that could activate DCs before integration. We generated the mutation V86I-I AP9 1 LPA-M96L in HIV- 1 capsid that reproduced the cyclophil in-binding loop of HIV-2 P86HA. As expected, exposure of DCs by HIV- 1 WT of HIV- 1 V86I-IAP91LPA-M96L in the absence of Vpx did not lead to infection of activation of the DCs ( Figure 21). In the presence of Vpx. H IV- 1 WT infected and activated DCs. Activation was inhibited by CsA. as previously shown. In the case of HIV - 1 V86I-IAP91 LPA- M96L in the presence of Vpx. GFP expression was reduced to 8% compared to 51% for the WT virus. However, activation reached 57%. Addition of CsA to HIV- 1 V86I-IAP9 1 LPA-M96L inhibited activation to 21% and increased infection to 20%. Thus, like H IV-2 P86HA. HIV- 1 V86I-IAP9 1 LPA- 96L activates DCs but is not infectious, when the restriction to DC infection is removed by providing Vpx.

Mutations in HIV- 1 that allow the virus to escape CsA inhibition have been described. In particular, mutation A92E and G94D in capsid allow HIV- 1 to replicate in the presence of CsA. Thus, these mutations confer an i nfection phenotype comparable to HIV-2 P86HA. which is increased in the presence of CsA. We asked whether these mutations in HIV- 1 could, like HIV-2 P86HA. also lead to DC activation before integration (Figure 22 ). In the absence of Vpx. I I IV- I A92E and G94D did not infect and activate DCs. In the presence of Vpx. H IV- 1 A92E and G94D infected and activated DCs. In the presence of CsA. DC activation by these mutant viruses was inhibited. In the presence of Raltegravir. HIV- 1 A92E and G94D maintained DC activation, while HIV- 1 WT was not able to activate DCs. Thus. HIV- 1 A92E and G94D have the potential to activate DC in the absence of integration.

In conclusion, we demonstrated that HIV-2 P86HA has lost its infectious potential in DCs and has maintained the ability to trigger an innate immune response. This requi ed an interaction with cellular Cyclophilin A. the presence of viral RNA, the presence of Vpx. and the activity of reverse transcriptase. However the integrase activity was not required, and integrated D A was not detected upon infection with HIV-2 P86HA. Essentially similar results were obtained with H IV- 2 P86RA. P86QA. P86AA. P86AM, P86HV and P86P1. In contrast. HIV-2 P86PA had a decreased titer indicative of general defects in the particles, and HIV-2 PG86HAA was still infectious hut did not activate well DCs. By applying the same strategy of modification used for HIV-2 P86HA in SIVmac239, leading to the mutant QPAPQQ85IHAGPLPA, we obtained SIVmac239 mutant virus with similar properties to H IV-2 P86HA. We were able to extend our results to macaque DCs. In HIV- 1 . similar results were also obtained by applying the same strategy of mutation, generating HIV- 1 V86I-IAP91 LPA-M96L. when Vpx was also provided. Final ly, similar results were obtained with HIV- 1 mutants that were described to be CsA-dependent for infectivity ( HIV - 1 A92E and G94D). Finally. HIV-2 WT could also activate DCs when integration was blocked, but the potency was reduced compared to HIV-2 P86HA.

Thus, we demonstrate that any H IV- 1 . HIV-2 or SIVinac can activate DCs in the absence of infection and integration when 3 criteria are met: (i) when infectivity mediated by the caps id is CsA-dependent (ii) when a viral cDNA is efficiently reverse transcribed in DCs and (iii) when integration in DCs is blocked. Any viral repl ication step once these criteria are met is dispensable.

Claims

1. A replication-defective particle comprising a capsid protein,

wherein said capsid protein carries at least one mutation in the cyclophilin- binding loop compared to residues 83 to 98 of the HIV-2 capsid polypeptide having the sequence set forth in SEQ ID No: l,

and wherein said replication-defective particle is capable of activating dendritic cells (DCs) without infecting said DCs.

2. A replication-defective particle comprising a capsid protein,

wherein said capsid protein binds to cyclophilin A with an increased affinity compared to the capsid polypeptide having the sequence set forth in SEQ ID No : 1,

wherein said replication-defective particle comprises a Vpx protein, viral RNA and a reverse transcriptase.

3. A replication-defective particle according to claim 2, wherein said capsid protein carries at least one mutation in the cyclophilin-binding loop compared to residues 83 to 98 of the HIV-2 capsid polypeptide having the sequence set forth in SEQ ID No: 1.

4. A replication-defective particle according to any one of claims 1 to 3, wherein said capsid protein carries a mutation in the cyclophilin-binding loop at the residue P86, wherein the amino acid numbering is with reference to the HIV-2 capsid polypeptide as set forth in SEQ ID No: 1.

5. A replication-defective particle according to any of the above claims wherein the capsid protein does not comprise a mutation at position 88 wherein the amino acid numbering is with reference to the HIV-2 capsid polypeptide as set forth in SEQ ID No: 1.

6. A replication-defective particle according to any of the above claims wherein said capsid protein comprises the amino acid sequence as set forth in any one of SEQ ID No:2 to SEQ ID No:7 and SEQ ID No:53 to SEQ ID No:61.

7. A replication-defective particle according to any of the above claims, wherein said replication-defective particle does not comprise an integrase or integrase activity.

8. A replication-defective particle according to any of the above claims, further comprising a ribonucleic acid molecule.

9. A replication-defective particle according to any of the above claims, wherein the particle is derived from HIV-2 or HIV-1 or SIVmac239 virus.

10. A vector comprising a nucleic acid encoding a replication-defective particle as defined in any one of claims 1 to 9.

11. A pharmaceutical composition comprising a replication-defective particle according to any one of claims 1 to 9 and/or a vector according to claim 10 and a pharmaceutically acceptable carrier or excipient.

12. A replication-defective particle according to any one of claims 1 to 9 and/or a vector according to claim 10 and/or a pharmaceutical composition according to claim 11 for use in a method of treatment.

13. A replication-defective particle and/or vector and/or pharmaceutical composition according to claim 11 for use in a method for treating and/or preventing HIV infection.

14. A replication-defective particle and/or vector and/or pharmaceutical composition according to claim 11 for use in a method for treating and/or preventing cancer.

15. An in vitro method for inducing the activation of dendritic cells in the absence of infection comprising the step of exposing said DC to a replication-defective particle according to any one of claims 1 to 9 and/or a vector according to claim 10 and/or a pharmaceutical composition according to claim 11.

16. A polypeptide comprising an amino acid sequence having the sequence as set forth in any of SEQ ID No:2 to SEQ ID No:7 and SEQ ID No:53 to SEQ ID No:61.