Authors: Frédéric Coutant, Marie-Pascale Frenkiel, Philippe Despres, Pierre Charneau
Lentiviral vectors are under intense scrutiny as unique candidate viral vector vaccines against tumor and aggressive pathogens because of their ability to initiate potent and durable specific immune responses. Strategies that alleviate safety concerns will facilitate the clinical developments involving lentiviral vectors. In this respect, the development of integration deficient lentiviral vectors circumvents the safety concerns relative to insertional mutagenesis and might pave the way for clinical applications in which gene transfer is targeted to non-dividing cells. We thus evaluated the potential use of nonintegrative lentiviral vectors as vaccination tools since the main targeted cell in vaccination procedures is the non-dividing dendritic cell (DC). In this study, we demonstrated that a single administration of nonintegrative vectors encoding a secreted form of the envelope of a virulent strain of West Nile Virus (WNV) induces a robust B cell response. Remarkably, nonintegrative lentiviral vectors fully protected mice from a challenge with a lethal dose of WNV and a single immunization was sufficient to induce early and long-lasting protective immunity. Thus, nonintegrative lentiviral vectors might represent a safe and efficacious vaccination platform for the development of prophylactic vaccines against infectious agents.
Prevention of infectious diseases through vaccine development is one of the greatest achievements of modern medicine. Nonetheless, considerable challenges remain for the development of new vaccines combining efficiency with enhanced safety profiles. Recently, human immunodeficiency virus (HIV)-1–derived lentiviral vectors have emerged as very promising vaccination tools. These vectors elicit both specific cytotoxic and strong humoral immune responses in several animal models , , , , . Immune responses elicited by lentiviral vectors are more efficient than those induced by conventional viral vector vaccines , , . These properties rely in part on the ability of these vectors to mediate more efficient gene transfer into dendritic cells (DC) than other vaccinal vectors such as for instance the widely used adenoviral vectors . This major advantage allows a sustained expression and endogenous presentation of tumoral or viral Ags by transduced DC and subsequent activation of more potent and persistent specific immune responses , , , , . Moreover, problems of vector-specific immunity are largely reduced with the use of LV because of the absence of pre-existing immunity in humans.
Although lentiviral vector-based vaccines have been shown to elicit unparallel specific and protective immunity against tumor and infectious agents, their transition as vaccinal vectors from preclinical evaluation to clinical development is hindered by their status of integration-competent viral vectors and potential problems linked to the non-specific integration of the transgene in the genome of the transduced cells. However, several arguments point out that lentiviral vectors possess a much safer profile than other retroviral vectors, especially when they are applied to the field of vaccination. Firstly, in a vaccination scenario based on direct injection of Ag-encoding lentiviral vectors, transduced cells that express the relevant Ag become targets of the elicited immune response and are eliminated within a few weeks from the vaccinated organism . Secondly, the deletion in the 3′ LTR of the viral promoter and enhancer sequences in self-inactivating lentiviral vectors limits the likelihood of endogenous promoter activation . Thirdly, although many lymphocytes of HIV-infected patients carry a large number of transcriptionally active HIV integrated proviruses, there is no evidence to suggest that integration of lentiviruses would cause oncogenesis. Consistently, a recent study has provided evidences that lentiviral vector transduction, even at high integration loads do not accelerate tumorigenesis in a tumor-prone murine model . Nevertheless, investigations to confirm the safety of lentiviral vectors or elaboration of novel strategies bypassing these safety considerations need to be actively pursued. In this context, the development of lentiviral vectors that allows a targeted integration or does not integrate at all would provide an important step toward the development of fully safe vector-based vaccines. The last strategy can be easily achieved with the use of integration-deficient lentiviral vectors carrying a defective integrase mutated in residues critical for the integration of viral DNA into the host genome. One of the most well-studied mutation is the D64V substitution in the catalytic domain of the enzyme, which has been shown to block the DNA cleaving and joining reactions of the integration step . The incorporation of this mutation in pseudotyped HIV-1 particles or in lentiviral vector particles leads to only barely detectable integration events and accumulation of extrachromosomal non-replicating circular genomes in the nucleus of the transduced cells . These circular genomes are diluted through cell division but support transcription by the cellular machinery, allowing stable transgene expression in non-dividing cells. As such, lentiviral vectors defective for integration could supersede their integrating counterparts in clinical purposes that target postmitotic cells. This notion was supported by recent studies demonstrating the efficiency of nonintegrative lentiviral vectors to transduce primary cells such as neuronal cells , muscle cells  or retinal cells and to rescue representative models of retinal degeneration .
In this study, we asked whether integration-deficient lentiviral vectors could initiate a specific and protective immune response. We hypothesized that the induction of Ag specific immune responses in vaccine strategies with lentiviral vectors should not require vector integration, since targeted cells are non-dividing DC. To test this hypothesis, we firstly evaluated the ability of these vectors to transduce in vitro different subpopulations of DC. We next investigated whether immunization with nonintegrative lentiviral vector particles was efficient to stimulate a specific and protective humoral immune response against West Nile Virus (WNV), a flavivirus responsible for the largest recognized epidemic of neuroinvasive human disease in North America.
We show that integration-deficient lentiviral particles are not only efficient in transducing DC but also that a single immunization with these vectors coding for the secreted form of the WNV envelope can prime humoral Ag-specific responses and confers complete protection against lethal challenge.