Unit: Virologie moléculaire et vectorologie

Director: Pierre Charneau

Our research focuses on the study of the early steps of lentiviral replication, and particularly on the mechanisms of HIV-1 genome nuclear import. The understanding of these fundamental mechanisms is channelled towards the elaboration of efficient lentiviral gene transfer vectors. These new vectors open up numerous applications in gene therapy (developed in the context of collaborations), and also prove to be powerful tools for vaccination against AIDS itself as well as other viruses and certain tumours (projects developed in our laboratory).

The active nuclear import of lentiviral genomes, via routing and translocation of viral complexes through the nuclear pores of interphasic nuclei, accounts for their capacity to infect non-dividing cells. We have previously shown that HIV-1 genome nuclear import is governed by original mechanisms: the presence of two central cis-acting sequences within the HIV-1 genome (cPPT and CTS) is responsible for the synthesis of a three-stranded DNA structure (DNA flap) that functions as a cis-acting determinant for the nuclear import of viral DNA.

Fundamental Virology

The central DNA flap sequence present within the HIV-1 genome, and within all lentiviral genomes, is a cis-acting determinant of HIV-1 genome nuclear import. An important share of our projects seeks to understand by which precise molecular mechanism the DNA flap is implicated in HIV-1 genome nuclear import.

We have recently shown that central termination, which is the final step in reverse transcription leading to formation of the DNA flap, induces the maturation of the reverse transcription complex (RTC) into pre-integration complex (PIC), free from capsid and thus of size compatible for translocation through the nuclear pore. Moreover, we have identified by transmission and scanning electron microscopy (TEM and SEM), and Cryo SEM, the ultrastructure of intracellular HIV-1 complexes (coll. T. Allen and M.C. Prévost).

It is also possible that the DNA flap structure recruits viral and/or cellular protein ligands that have nuclear localisation signals (NLSs), thus permitting initiation of translocation of the DNA filament through the nuclear pore. We are using a number of experimental approaches to identify these shuttle proteins.

In order to visualise the early steps of HIV-1 infection in a quantitative manner, up to integration of the provirus in the cellular DNA, we have set up a system for the dynamic imaging and 4D tracking (coll. S. Shorte and J.C. Olivo-Marin) of HIV-1. Our labelling of HIV-1 integrase by a fluorescent marker enables the dynamic follow-up of the viral DNA within living infected cells. This work has provided us with entirely new data on the kinetic characteristics of the different movements exhibited by HIV-1 complexes within the cytoplasm and nucleus of infected cells. It has also provided an invaluable tool for the study of virus/host cell interactions during infection.

Vectorology and Gene Transfer

An increasing interest is given to lentiviral gene transfer vectors in all areas of gene therapy. We have shown that incorporation of the DNA flap structure within lentiviral vector constructs strongly stimulates gene transfer in all cells and tissues examined ex vivo or in vivo (haematopoietic stem cells (HSCs), brain, liver etc.). "Flap lentiviral vectors" have the capacity to transduce non-dividing cells, thus overcoming one of the main hurdles for efficient gene transfer using "classical" retroviral vectors derived from Moloney virus (MoMLV). Moreover, many cell types that are poorly permissive or even refractory to transfection are now stably transduced with close to 100% efficiency. This is the case, for example, with primary hepatocytes, astrocytes, T lymphocytes, macrophages, and dendritic cells.

Our aim has been, and continues to be, the establishment of optimal conditions for gene transfer ex vivo or in vivo in cells and tissues of major therapeutic interest, such as the brain, the liver, and HSCs, for which we have already shown 10-fold increases in transduction efficiencies following incorporation of the DNA flap within the lentiviral vector. Gene therapy projects for diseases complemented by gene transfer in HSCs, such as leukodystrophies (coll. N. Cartier), are currently under way. Moreover, following the recent serious side effects encountered with the use of retroviral vectors derived from MoMLV in the SCID-X1 trial, we currently participate in the setting up of a safer gene therapy protocol for "bubble children" diseases based on lentiviral vectors (coll. A. Fischer and M. Cavazzana).

Fundamental Vectorology

We are carrying out the optimisation of parameters for the construction and production of lentiviral vectors derived from HIV-1 and including the central DNA flap. Our work includes the elaboration of new generation lentiviral vectors including a symmetrical DNA flap and transcriptional Pol II or Pol III units in the U3 region; the choice of promoters adapted to the different cell types studied; production of vector particles using an optimised stable encapsidation cell line; and optimisation of techniques for vector titration using quantitative PCR. Finally, an important priority has been, and remains, the improvement of vector design in order to optimise gene transfer safety.

Moreover, we have set up a system for the expression of siRNAs from lentiviral vectors, for the induction of stable RNA interference in a reproducible and cost-effective manner. These constructs also pave the way for the generation of transgenic knockdown animals (coll. C. Gouget and P. Avner).


The capacity of lentiviral vectors to transduce with high efficiency cells of the immune system, and particularly antigen presenting cells (dendritic cells, DCs), make them an invaluable tool for the development of anti-viral and anti-tumoral vaccination protocols.

We are currently working on the development of an optimised protocol for therapeutic vaccination against AIDS using lentiviral vectors coding for HLA-A2 or -B7 restricted HIV-1 polyepitopes. These allow us to generate, after a single injection in "humanised" HLA-A2 or -B7 mice, a strong and broad primary T cell response, and to establish efficient memory immunity. Lentiviral vectors have the advantage of permitting the stable transduction of DCs, thus enabling antigen presentation during the entire lifetime of DCs in vivo. This presentation is carried out both on MHC class I for the endogenous pathway, and by cross-presentation of the vector GAG/POL epitopes on MHC class II. This leads to the stimulation of both CD8+ T lymphocytes (CTL) and CD4+ lymphocytes. Moreover, lentiviral vectors have a natural tropism for DCs in vivo: a single sub-cutaneous injection of vector particles is equivalent, indeed even superior, in terms of T cell response efficiency compared with the far heavier protocol of re-infusion of ex vivo transduced DCs.

We also focus our studies on other viral models, in particular West Nile and Dengue viruses (coll. P. Desprès), SARS, influenza (coll. S. Van Der Werf), and Rift Valley virus (coll. M. Bouloy). In the case of the West Nile virus (WNV), we have shown that a single immunisation in a mouse model with a minute dose of lentiviral vector coding for the envelope protein induces a very quick, long-lasting protective immune response against WNV infection.

In the context of anti-tumoral vaccination, we have shown that a single injection in HLA-A2 mice of lentiviral vectors coding for a melanoma polyepitope induces a strong and broad CTL response against melanoma epitopes.


Activity Reports 2004 - Institut Pasteur

Page Top research Institut Pasteur homepage

If you have problems with this Web page, please write to rescom@pasteur.fr