SHORT COMMUNICATION SIV MAC Vpx improves the transduction of dendritic cells with nonintegrative HIV-1-derived vectors

Transcription

1 (2009) 16, & 2009 Macmillan Publishers Limited All rights reserved /09 $ SHORT COMMUNICATION SIV MAC Vpx improves the transduction of dendritic cells with nonintegrative HIV-1-derived vectors G Berger 1,2,3, C Goujon 1,2,3, J-L Darlix 1,2,3 and A Cimarelli 1,2,3 1 LaboRetro, Department of Human Virology, Ecole Normale Supérieure de Lyon, Lyon, France; 2 INSERM, U758, Lyon, France and 3 University of Lyon, Lyon1, IFR128 BioSciences Lyon-Gerland, Lyon-Biopole, Lyon, France Lentiviral vector (LV)-mediated gene therapy bears an intrinsic risk of insertional mutagenesis following integration into the host genome. Nonintegrative LVs may offer an alternative avenue at least in nondividing cells where episomal viral DNA persists stably. Owing to their central role in immune system functions, differentiated dendritic cells (DCs) offer an interesting cell target for these vectors. We have previously described that the transduction of DCs with wild-type HIV-1-derived vectors can be considerably improved by providing DCs with noninfectious virion-like particles (VLPs) carrying Vpx (Vpx-VLPs), a nonstructural protein coded by members of the SIV SM /HIV-2 lineage that removes a specific restriction to lentiviral infection in these cells. Here, we describe that the transduction efficiency of DCs with nonintegrative HIV-1 vectors can also be improved via Vpx-VLPs that promote the accumulation of complete and episomal viral DNA. In this setting, Vpx increases both the number of transduced cells and the levels of transgene expression. Thus, these results describe a simple procedure by which transduction of differentiated DCs can be achieved at low viral inputs with safer LVs to improve both the number of transduced cells and the levels of transgene expression. (2009) 16, ; doi: /gt ; published online 31 July 2008 Keywords: nonintegrative; HIV-1 lentiviral vectors; DCs; Vpx One of the major interests of lentiviral vectors (LVs) in gene therapy is their ability to provide long-lasting transgene expression following integration into the host genome. This very property may lead to severe consequences on the cell s physiology owing to viral integration into genes that deregulate the normal cellular transcriptional program. 1 Given the propensity of lentiviral integrases (IN) for active transcriptional units, 2 this represents a non-negligible issue. To circumvent this problem, several groups have evaluated the use of integration-incompetent LVs. 3 8 Gene expression from these vectors is driven from viral DNA molecules that remain in an episomal form after their entry into the nucleus as 1 or 2 long terminal repeat circles (1- or 2-LTRs, respectively). These episomes are considered as a deadend product in the normal retroviral infection process, but they are a functional substrate for the transcriptional machinery. 9 As they lack an origin of replication, these forms are transient in dividing cells, but remain stably present as episomes in the nucleus of nondividing cells. Dendritic cells (DCs) are an interesting cell type for gene therapy purposes in light of their master role in the regulation of immune system responses. 10 Although viral infection of DCs can be obtained with lentiviral-derived vectors, the efficacy of this modification is generally low unless high viral inputs are used, because DCs are poorly Correspondence: Dr A Cimarelli, LaboRetro, Department of Human Virology, ENS-Lyon INSERM, U758, 46 Allée d Italie, Lyon, France. acimarel@ens-lyon.fr Received 6 May 2008; revised 23 June 2008; accepted 7 July 2008; published online 31 July 2008 susceptible to lentiviral infection. 11,12 We have previously reported that members of the SIV SM /HIV-2 lineage code for a protein, Vpx, are capable of relieving the restriction that is responsible for the poor susceptibility of DCs to lentiviral-mediated modification. 13,14 Indeed, Vpx augments the efficacy of infection of DCs with integrationcompetent HIV-1 vectors when it is added in trans onto DCs via noninfectious virion-like particles (VLPs) derived from SIV MAC (Vpx-VLPs). In this setting, Vpx promotes the accumulation of viral DNA. Thus, we studied whether similar effects could be observed following modification of DCs with nonintegrative HIV-1-based LV (as depicted in Figure 1a). Viral integrases (IN) exert several functions during the viral life cycle, ranging from viral particle formation to reverse transcription, and possibly nuclear import. Thus, IN mutants have been divided into two classes according to the specific or pleiotropic defects they impose on mutant viruses (class I or II, respectively 18 ). Mutants in the catalytic site of HIV-1 IN belong to the former. We first constructed an integration defective HIV-1 LV by introducing a single point mutation in the catalytic site of the IN (IN-D116A). This mutation did not impair viral production and yielded similar infectious titers as compared to wild-type (WT) on HeLa cells, if transduced cells were examined 3 days after infection. However, a sharp loss of green fluorescent protein (GFP)-expressing cells was specifically observed in cells transduced with the IN-D116A mutant when cells were analyzed 11 days after infection (Figure 1b). This result is not surprising for dividing cells transduced with a nonintegrative LV, as episomal DNA is expected to dilute over cell divisions. Indeed, when the amount of integrated viral DNA was

2 160 Figure 1 Schematic representation of the lentiviral vectors (LVs) used in the study and characterization of a nonintegrative LV. (a) Integration-deficient LVs were engineered by introducing a single point mutation in the catalytic site of integrase (IN-D116A) in the context of an HIV-1-based packaging construct (8.2). 15 LVs were obtained after transfection of 293 T cells with the packaging construct coding gag-pro-pol and all the accessory proteins (Tat, Rev, Vif, Vpr and Nef; not marked for clarity), a miniviral genome bearing a CMV-GFP expression cassette and a construct coding the vesicular stomatitis virus G envelope protein, as described. 15 Noninfectious virion-like particles (referred to as Vpx-VLPs) were similarly produced by transfection of an SIV MAC -based construct coding gag-pro-pol and viral accessory proteins (SIV3 +, Tat, Rev, Vif, Vpx, Vpr and Nef 16 ). No viral genome was transfected in this case. We have previously documented that Vpx is the protein responsible for the positive effect of VLPs on the infectivity of incoming HIV-1 LVs. 13 Virions were purified by ultracentrifugation on a double step sucrose cushion (45 25%) and normalized by protein content against standards of known infectivity, or by their infectious titers on HeLa cells, as previously described. 17 Infections were carried out on 10 5 cells for 2 h prior to extensive cell washing. The percentage of transduced GFP-positive cells was determined 3 days after infection by flow cytometry. When indicated, Vpx- VLPs were provided together with GFP-coding vectors at a multiplicity of infection (MOI) of 2. (b) HeLa cells were transduced with normalized amounts of wild-type (WT) and IN-D116A LVs and analyzed by flow cytometry 3 and 11 days after infection. Control infections were performed in the presence of the reverse-transcriptase inhibitor azidothymidine (AZT) at 10 mg/ml. (c) The presence of integrated provirus was analyzed on transduced cells infected at MOI 10 by Alu-PCR, as described previously. 9 Briefly, the first round of PCR was carried out for 20 cycles with a primer specific for the proviral DNA and a second round specific for Alu sequences in the genome. Fivefold dilutions of the first PCR were then used as the template for a second semiquantitative PCR using two internal primers specific for viral DNA. Amplified products were transferred onto a nylon membrane and hybridized with 32 P-labeled specific probes prior to Phosphorimager quantification. A representative gel is shown here together with a graph presenting data obtained from three independent experiments. CMV, cppt, central polypurine tract; GFP, green fluorescent protein; p.i., post-infection; RRE, Rev-responsive element; *, self-inactivating. determined by Alu-PCR, none was observed in cells transduced with the IN-D116A mutant (Figure 1c). Next, the ability of normalized WT and IN-D116A mutant LVs was analyzed on primary DCs, macrophages and phytohemagglutinin-stimulated primary blood lymphocytes (PBLs) (Figure 2a). As we had previously reported, Vpx-VLPs increased by over 10-fold the percentage of GFP-positive DCs following infection with WT HIV-1 vectors. 13 When compared to WT, similar amounts of GFP-expressing cells were also obtained following transduction of DCs with the IN-D116A mutant. Similar to WT, the addition of Vpx-VLPs during IN-D116A LV infection dramatically increased the percentage of transduced cells at all viral inputs tested (by 10- to 20-fold). Accordingly, the percentage of macrophages transduced with the IN-D116A mutant in the presence of Vpx-VLPs increased, although to a lower extent than with DCs (three- to fivefold). 13 As expected in light of the cell-type-specific function of Vpx-VLPs, no effect was observed on stimulated PBLs. Of note, transduction with the IN-D116A mutant virus yielded a substantially lower percentage of transduced cells than WT, even if analysis was carried out 3 days postinfection, that is, prior to substantial dilution of episomal DNA following cell division. We believe that this result can be due to lower transcriptional activity from the episomal DNA in PBLs as opposed to the integrated forms. In addition to their positive effect on the percentage of HIV-1-LVs transduced cells, Vpx-VLPs also increased the median fluorescence intensity of GFPexpressing cells (MFI; Figure 2b), suggesting that in cells in which they exert a positive effect on viral infection, Vpx-VLPs also act positively on the levels of transgene expressed intracellularly. Modified DCs could be kept in culture for over a month, retaining stable expression of the transgene (Figure 2c). The advantage of cells transduced in the presence of Vpx-VLPs appeared evident at all time points.

3 161 Figure 2 Virion-like particles (VLPs) carrying Vpx (Vpx-VLPs) increase the efficiency of transduction of dendritic cells (DCs) with an integration-incompetent lentivirus (LV). Primary monocytes were obtained from the blood of healthy donors after ficoll/percoll purification of leukocytes and negative depletion of unwanted cell contaminants (Miltenyi Biotech), as described. 14 Monocytes were differentiated into macrophages or DCs for 4 6 days following incubation with granulocyte macrophage colony-stimulating factor (GM-CSF; AbCys) or GM- CSF plus interleukin 4 (IL-4; AbCys), respectively. Primary blood lymphocytes (PBLs) were activated 24 h before transduction with phytohemagglutinin (PHA) and IL-2. (a) Transductions were carried out with normalized amount of virions as described in the legend to Figure 1 in the presence or absence of Vpx-VLPs provided at multiplicity of infection (MOI) 2. The bar graph presents the typical effect observed on the infectivity of wild-type (WT) HIV-1 vectors (MOI 1) when DCs are pre-incubated with Vpx-VLPs, as previously shown. 14 To exclude the possibility that green fluorescent protein (GFP)-positive cells arouse from the uptake of contaminant GFP protein present in viral preparations (pseudotransduction), infections were carried out in the presence of the reverse-transcriptase inhibitor azidothymidine (AZT) at 10 mg/ml (obtained through the AIDS reagents and reference program of the NIH). No GFP-positive cells were observed in this case. (b) The median fluorescence intensity (MFI) of GFP-expressing cells was determined in cells transduced at MOI 10, 3 days post-infection (p.i.). (c) DCs transduced with the different LVs were analyzed at the indicated time post-infection by flow cytometry to analyze the stability of reporter gene expression over time. Data obtained from four to five independent experiments are shown here. ND, not determined. Finally, the effect of Vpx-VLPs was evaluated by PCR by examining the amount of viral DNA synthesized at 24 h post-infection in a comparison between PBLs, where Vpx-VLPs display no effect, and DCs (Figure 3). No obvious effect was observed in the amount of full-length and episomal viral DNAs synthesized following infection of PBLs with WT, IN-D116A and IN-D116A supplemented with Vpx-VLPs. In contrast, Vpx-VLPs increased the amount of full-length and episomal viral DNA produced at 24 h post-infection in DCs, confirming the positive effect of Vpx on viral DNA accumulation. 14 Overall, the data presented here indicate that transduction of DCs and, to a lower extent, of macrophages with nonintegrative HIV-1 vectors can be substantially improved using noninfectious SIV MAC VLPs carrying Vpx, thereby extending our previous observations on integrative HIV-1 vectors. 13,14 The positive effect of Vpx-VLPs on nonintegrative vectors was not totally unexpected in light of our previous findings. 13 Yet, the possibility existed that Vpx required a functional integration complex to exert its function, or that nonintegrative vectors were submitted to defects other than integration that were not sensitive to Vpx. Although at present we ignore the mechanism by which it exerts its functions, Vpx increases the amount of full-length viral DNA that accumulates during infection of DCs. Two recent reports indicated that the function of Vpx during SIV infection could be exerted through the recruitment of a Cul4-based E3-ubiquitin ligase complex. Indeed, Vpx has been reported to associate with this complex either via an adaptor protein, the DDB1- (damaged DNA-binding protein 1) and CUL4-associated factor 1, or via its structural component, the DDB1. 19,20 Removal of these factors by small interfering RNAs impaired the functionality of Vpx, suggesting that Vpx may target a cellular factor to this E3-ubiquitin ligase complex and induce its degradation. This function would be required to remove an antiviral restriction present in DCs, thereby exerting a protective effect on viral nucleoprotein complexes. For gene therapy applications, this effect translates in a higher proportion of infectious viral genomes reaching the nucleus and, thus, in a higher fraction of transduced cells in the absence of detectable changes in the physiology of DCs. 13 Safety from the deleterious effects of insertional mutagenesis has become a serious concern in retroviral

4 162 Figure 3 Virion-like particles (VLPs) carrying Vpx (Vpx-VLPs) promote complete and episomal viral DNA accumulation in transduced dendritic cells (DCs). Primary blood lymphocytes (PBLs) and DCs transduced with normalized amounts of wild-type (WT) or IN-D116A LVs in the presence or absence of Vpx-VLPs were harvested, lysed and the amount of synthesized viral DNA was determined by semiquantitative PCR using primers that recognize specifically full-length (FL) and episomal (2-LTRs) DNA forms, as described. 14 Briefly, PCRs were carried out on fivefold serial dilution of samples to ensure the linearity of the reaction, amplified products were transferred onto a nylon membrane and hybridized with 32 P-labeled specific probes prior to Phosphorimager quantification. The signals were normalized for those obtained after amplification of actin DNA and with respect to WT. Data obtained from four to five independent experiments are shown here. IN, integrase; LTR, long terminal repeat. vector-mediated gene therapy applications. Nonintegrative vectors have become an important alternative to their use, as they remove the cause of such effects, that is, integration. However, given that transgene expression from these vectors depends on the presence of episomal DNA, their use is limited to applications in which transgene expression is required either transiently in dividing cells or stably in differentiated nondividing cells. In this respect, DCs as well as macrophages are suitable targets for modification via nonintegrative vectors, as they are differentiated antigen-presenting cells, but display a partial resistance to lentiviral infection and elevated viral inputs are needed to achieve high levels of transduction that may modify the cells physiology. 11,12 The protocol described here allows an efficient modification of DCs with nonintegrative and, thus, safer vectors, in the presence of low viral inputs. This property may be important to avoid viral overload of DCs and possible deleterious consequences on the physiology of DCs both ex vivo and in vivo. Acknowledgements We are indebted to Jeanine Bernaud and Dominique Rigal for help with blood sample collection. AC received support of Sidaction and the ANRS. References 1 Hacein-Bey-Abina S, Von Kalle C, Schmidt M, McCormack MP, Wulffraat N, Leboulch P et al. LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1. Science 2003; 302: Bushman FD. Integration site selection by lentiviruses: biology and possible control. Curr Top Microbiol Immunol 2002; 261: Yanez-Munoz RJ, Balaggan K, MacNeil A, Howe SJ, Schmidt M, Smith AJ et al. Effective gene therapy with nonintegrating lentiviral vectors. Nat Med 2006; 12: Saenz DT, Loewen N, Peretz M, Whitwam T, Barraza R, Howell KG et al. Unintegrated lentivirus DNA persistence and accessibility to expression in nondividing cells: analysis with class I integrase mutants. J Virol 2004; 78: Philippe S, Sarkis C, Barkats M, Mammeri H, Ladroue C, Petit C et al. Lentiviral vectors with a defective integrase allow efficient and sustained transgene expression in vitro and in vivo. Proc Natl Acad Sci USA 2006; 103: Nightingale SJ, Hollis RP, Pepper KA, Petersen D, Yu XJ, Yang C et al. Transient gene expression by nonintegrating lentiviral vectors. Mol Ther 2006; 13: Loewen N, Leske DA, Chen Y, Teo WL, Saenz DT, Peretz M et al. Comparison of wild-type and class I integrase mutant-fiv vectors in retina demonstrates sustained expression of integrated transgenes in retinal pigment epithelium. J Gene Med 2003; 5: Fang JY, Mikovits JA, Bagni R, Petrow-Sadowski CL, Ruscetti FW. Infection of lymphoid cells by integration-defective human immunodeficiency virus type 1 increases de novo methylation. J Virol 2001; 75: Wu Y, Marsh JW. Selective transcription and modulation of resting T cell activity by preintegrated HIV DNA. Science 2001; 293: Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, Liu YJ et al. Immunobiology of dendritic cells. Annu Rev Immunol 2000; 18: Gruber A, Kan-Mitchell J, Kuhen KL, Mukai T, Wong-Staal F. Dendritic cells transduced by multiply deleted HIV-1 vectors exhibit normal phenotypes and functions and elicit an HIVspecific cytotoxic T-lymphocyte response in vitro. Blood 2000; 96: Tan PH, Beutelspacher SC, Xue SA, Wang YH, Mitchell P, McAlister JC et al. Modulation of human dendritic-cell function following transduction with viral vectors: implications for gene therapy. Blood 2005; 105: Goujon C, Jarrosson-Wuilleme L, Bernaud J, Rigal D, Darlix JL, Cimarelli A. With a little help from a friend: increasing HIV transduction of monocyte-derived dendritic cells with virionlike particles of SIV(MAC). 2006; 13: Goujon C, Riviere L, Jarrosson-Wuilleme L, Bernaud J, Rigal D, Darlix JL et al. SIVSM/HIV-2 Vpx proteins promote retroviral escape from a proteasome-dependent restriction pathway present in human dendritic cells. Retrovirology 2007; 4: 2.

5 15 Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, Gage FH et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996; 272: Mangeot PE, Negre D, Dubois B, Winter AJ, Leissner P, Mehtali M et al. Development of minimal lentivirus vectors derived from simian immunodeficiency virus (SIVmac251) and their use for gene transfer into human dendritic cells. JVirol2000; 74: Goujon C, Jarrosson-Wuilleme L, Bernaud J, Rigal D, Darlix JL, Cimarelli A. Heterologous human immunodeficiency virus type 1 lentiviral vectors packaging a simian immunodeficiency virus-derived genome display a specific post-entry transduction defect in dendritic cells. J Virol 2003; 77: Lu R, Limon A, Devroe E, Silver PA, Cherepanov P, Engelman A. Class II integrase mutants with changes in putative nuclear localization signals are primarily blocked at a postnuclear entry step of human immunodeficiency virus type 1 replication. J Virol 2004; 78: Sharova N, Wu Y, Zhu X, Stranska R, Kaushik R, Sharkey M et al. Primate lentiviral Vpx commandeers DDB1 to counteract a macrophage restriction. PLoS Pathog 2008; 4: e Srivastava S, Swanson SK, Manel N, Florens L, Washburn MP, Skowronski J. Lentiviral Vpx accessory factor targets VprBP/ DCAF1 substrate adaptor for cullin 4 E3 ubiquitin ligase to enable macrophage infection. PLoS Pathog 2008; 4: e