Use of Amplicon-6 Vectors Derived from Human Herpesvirus 6 for Efficient Expression of Membrane-Associated and -Secreted Proteins in T Cells

Abstract

The composite amplicon-6 vectors, which are derived from human herpesvirus 6 (HHV-6), can target hematopoietic cells. In the presence of the respective helper viruses, the amplicons are replicated by the rolling circle mechanism, yielding defective genomes of overall size 135 to 150 kb, composed of multiple repeats of units, containing the viral DNA replication origin, packaging signals, and the selected transgene(s). We report the use of amplicon-6 vectors designed for transgene expression in T cells. The selected transgenes included the green fluorescent protein marker, the herpes simplex virus type 1 glycoprotein D (gD), and the gD gene deleted in the transmembrane region (gDsec). The vectors were tested after electroporation and passage in T cells with or without helper HHV-6A superinfections. The results were as follows. (i)The vectors could be passaged both as cell-associated and as cell-free secreted virions infectious to new cells. (ii)The intact gD accumulated at the cell surface, whereas the gDsec was dispersed at internal locations of the cells or was secreted into the medium. (iii)Analyses of amplicon-6-gD expression by flow cytometry have shown significant expression in cultures with reiterated amplicons and helper viruses. The vector has spread to >60% of the cells, and the efficiency of expression per cell increased 15-fold, most likely due to the presence of concatemeric amplicon repeats. Current studies are designed to test whether amplicon-6 vectors can be used for gene therapy in lymphocytes and whether amplicon-6 vectors expressed in T cells and dendritic cells can induce strong cellular and humoral immune responses.

Human herpesvirus 6 (HHV-6) and HHV-7 genomes are each composed of a long stretch of unique DNA sequences, flanked by right and left direct repeats (DR_R and DR_L, respectively) in the arrangement DR_L-U-DR_R (33, 34, 61, 76). The pac-1 and pac-2 signals identified previously as directing the cleavage and packaging of various herpesvirus DNAs (20, 21, 33) are placed within the DRs, at the genome termini. HHV-6 and HHV-7 DNA replication is thought to involve genome circularization, placing the pac-1 and pac-2 signals in adjacent configurations. This is followed by rolling-circle replication and packaging of the concatemeric DNAs by cleavage at the pac-1 and pac-2 junctions located approximately a full-size genome (headfull) away (32, 33, 64, 76, 81, 82).

The composite amplicon vectors (Fig. 1) consist of (i) defective genomes with multiple reiterations of amplicon units, each containing the DNA replication origin, the packaging signals, and the selected transgene(s), and (ii) an adequate helper virus that provides the DNA replication and packaging functions, as well as the structural virions. We have derived amplicon vectors from herpes simplex virus type 1 (HSV-1), HHV-6, and HHV-7 and found them to be efficient tools for studying viral DNA replication and packaging (29, 45, 64, 71). Cleavage during packaging occurs 29 to 35 bp away from the pac-1 signal and 40 to 45 bp away from the pac-2 signals, located at approximately a “headfull” or full-length distance, resulting in defective genomes of overall size similar to that of nondefective viral helper DNA close to 150 kb, containing multiple reiterations of amplicon units (21, 45, 48, 49, 64, 82).

FIG. 1.

Layout of the amplicon type vectors. The amplicon type vectors derived from HSV-1, HHV-6, and HHV-7 contain a DNA replication origin, the pac-1 and pac-2 signals, and the transgene(s). A rolling-circle replication of the amplicon plasmid with enzymes and functions contributed by the helper virus yielded defective virus genomes with multiple reiterations of the input amplicon plasmids. The concatemeric genomes are packaged in virions contributed by the helper virus.

Extended analyses of HHV-6 packaging signals were reported by Deng and Dewhurst (22), who examined the necessity of the pac-1 and pac-2 sequences in concatemeric junctions during packaging of plasmids containing HHV-6 oriLyt and pac sequences. These authors have also described an apparent lack of influence of the human telomere repeat sequences (TRs) on the packaging process. In addition, Turner et al. (77) recently characterized the domains and structure of the HHV-6 DNA replication origin within amplicon-6 constructs and described inhibitory effect of the U94 rep protein on vector replication. We describe here the use of the amplicon-6 vector for efficient gene expression of cell surface and secreted proteins in human T lymphocytes. Such vectors could potentially be used in gene therapy and vaccination.

The roseoloviruses HHV-6A, HHV-6B, and HHV-7 are members of the Betaherpesvirinae subfamily that contain distinct genes, as well as homologous genes, arranged similarly across the viral genomes (24, 55, 61, 69). The HHV-6A and HHV-6B are closely related variants, with DNA sequence homology ranging from 75 to 97%, depending on the gene(s) examined (24, 38). These variants differ in their growth, antigenicity, and restriction enzyme patterns, as well as in their epidemiology and disease association (1, 61, 68, 86, 87, 89). Whereas the HHV-6B variants are associated with diseases, symptomatic infections with HHV-6A variants are rather rare. HHV-6B infects the majority of children during the first 2 years of life (10, 37, 59, 80, 87, 89, 90). The virus causes roseola infantum or exanthem subitum (ES), which is usually a mild disease characterized by several days of spiky fever and skin rash (90). In some ES patients, the infection spreads to the central nervous system (CNS), causing grave complications (88, 89). In addition, the HHV-6B variants enter into a latency phase from which they can be reactivated in patients with impaired immune capabilities, including impairment due to AIDS and kidney and bone marrow transplantation, causing complications as severe as lethal encephalitis (16, 18, 19, 25, 30, 62, 63, 91). In contrast to HHV-6B disease association, it is thus far uncertain whether HHV-6A causes disease, although recent studies have suggested potential involvement with chronic fatigue syndrome (CFS) and with multiple sclerosis (MS). The association with CFS rests on serology, virus cultivation (2), and finding viral DNA in peripheral blood lymphocytes (PBL) of CFS patients by PCR at a higher prevalence than in control PBL (4, 23). However, questions were raised as to whether a CFS imbalanced immune response resulted in generalized reactivation of several latent herpesviruses, including Epstein-Barr virus, HSV-1, HSV-2, varicella-zoster virus, HHV-6, and HHV-7, as well as other viruses (13, 36, 65). Furthermore, additional studies were unable to find evidence for HHV-6A infection in CFS patients (6, 17, 43, 56, 70, 83). An association of the virus with MS has also been proposed and questioned (79). A positive association involved finding HHV-6A antibodies (60), viral DNA in serum and urine (3), and in peripheral blood mononuclear cells (PBMC) by nested PCR (42). Furthermore, 14.6% of MS patients were found to have active HHV-6A infection in PBMC and sera by quantitative real-time PCR. Plaques obtained from autopsy material of MS patients were found to have HHV-6 DNA at levels higher than those in control samples from healthy individuals. In contrast, Beck et al. (6) reported their failure to detect DNA in serum and spinal fluid of 27 MS patients by using nested PCR. The lack of association of HHV-6A with acute disease favors the potential use of the virus as a carrier vector for gene therapy and vaccination, as proposed here.

The entry of HSV into cells involves the interaction of several HSV glycoproteins, including glycoprotein B (gB), gC, gD, gH, and gL, with arrays of alternate cell receptors (15), including (i) heparan sulfate glycosaminoglycans, which mediate virus attachment; (ii) herpesvirus entry mediator A, a member of the tumor necrosis factor receptor family (57); (iii) several members of the nectin family, which belong to the immunoglobulin superfamily; and (iv) 3-O-sulfated heparan sulfate. The multiplicities of receptors enable modulations of viral entry into different types of cells. The virion envelope gD plays a cardinal role in virion structure and the interactions with viral entry receptors (15). Because of its strong immunogenic properties, gD has served as a vaccination target (7, 9, 35, 46, 53, 72-74, 85). In the present study we have placed HSV gD in amplicon-6 vectors to test gD expression in T cells.

MATERIALS AND METHODS

Cells and viruses.

The J-JHAN cells, derived from Jurkat T cells (75), were propagated in RPMI 1640 medium supplemented with 10% heat-inactivated fetal calf serum (FCS) and 50 μg of gentamicin/ml. The J-JHAN cells and the HHV-6A (U1102) were obtained from the late Robert Honess. The infected cells were cocultivated with fresh uninfected cells, yielding new infections several days later with a pronounced cytopathic effect, characterized by cell enlargement, ballooning, and the formation of syncytia.

Cloning of the pac signals for amplicon-6 vectors.

We derived recombinant amplicon vectors with the pac-1 and pac-2 signals adjacent to each other, similar to their arrangement in concatemeric viral genomes (20, 22, 33, 64). The construction of the recombinant HHV-6A pac signals (Fig. 2A) involved the following. (i) pPac-2 (pNF1154) was derived by cloning an MluI subfragment of the SalI L fragment of HHV-6A DNA (52). (ii) pPac1 was derived (pNF1155) by subcloning the BamHI G fragment of HHV-6A (U1102), followed by transfer of the HindIII to DraI subfragment into a HindIII-SmaI-cut pBluescript (Stratagene). (iii)During this cloning several nucleotides of the pac signals were deleted. To compensate for these losses, as well as to insert a new BamHI site between pac-1 and pac-2, the sequence GGGCGGATCCCCC was inserted into the clones. (iv) The recombinant pac construct (pNF1156) was made by cloning the ApaI-BamHI fragment from pPac1 into an ApaI-BamHI-cut pPac-2. This resulted in the arrangement of the pac-1 and pac-2 signals as in the natural pac signal, with cleavage predicted 29 bp away from pac-2 and 43 bp away from pac-1 (33), according to previously described cleavage and packaging “rules” (20, 21, 33) and as reported by Thomson et al. (76) for HHV-6. (v) Altogether, the pac-1 sequences in amplicon-6 constructs described here correspond to a 267-bp segment placed at map coordinates 151,509 to 151,242 of the HHV-6A (U1102), whereas the pac-2 signal corresponds to a 556-bp segment at map coordinates 7,525 to 8,081 of HHV-6A (U1102) genome (accession number X83413; gi853961).

FIG. 2.

Cloning of OriLyt and recombinant pac signals into the amplicon-6 vector. (A) Location of the HHV-6A clones used for the construction of the recombinant pac plasmid (pNF1156). The arrows denote the pac-1 and pac-2 location and orientation. The solid line in clone pNF1156 denotes the vector sequences surrounding the new BamHI site. The human telomeric sequences are represented by the letter “t”. The BamHI (B), MluI (M), DraI (D), PstI (P), and ApaI (A) sites are shown. (B) Structures of pOriLyt (pNF1164), the recombinant pac plasmid (pNF1156), amplicon-6 (pNF1158), and the clones with pac-1 and pac-2 deleted. pNF1158 and pNF1159, which contain the recombinant pac in reverse orientations, were digested with BamHI and self-ligated. The resultant clones had deletions of the pac-1 or pac-2 signals. The BamHI (B), SmaI (Sm), XhoI (X), HindIII (H), EcoRV (EV), MluI (M), DraI (D), PstI (P), ApaI (A), ScaI (Sc), NotI (N), and SnaBI (SN) sites are shown.

Derivation of the amplicon-6 vector.

The oriLyt of HHV-6B (Z29) located within the 10.6-kb BamHI F fragment was subcloned, generating the pNF1164 (Fig. 2B). The HindIII-to-EcoRV segment of this clone contains the oriLyt sequences from map coordinates 69,561to 68,356 of HHV-6B (Z29) DNA (accession number NC_000898; gi9633069). The recombinant pac signal in the NotI-ScaI segment of pNF1156 was inserted into the EcoRV-cut pOriLyt (pNF1164). This generated pNF1158 and pNF1159, differing in the orientations of the pac inserts (Fig. 2B). Deletions of the BamHI segments of the two clones generated the constructs pNF1184 (delpac-1) and pNF1183 (delpac-2).

Amplicon-6 replication and propagation.

To verify the functionality of the vectors, J-JHAN cells were transfected with equal amounts of the constructed amplicon-6 vector (pNF1158) and the delpac-1 (pNF1184) and the delpac-2 (pNF1183) constructs, as well as the plasmid containing solely oriLyt (pNF1164). The cultures were super infected with HHV-6A (U1102) helper virus. DNA was analyzed at 7 to 10 days postelectroporation in the nuclear and cytoplasmic fractions. To eliminate traces of nonpackaged cellular and viral DNA, the cytoplasmic fractions were treated first with DNase I. DNA was also prepared from cell free virions, which were recovered from the medium. The DNAs, which were cleaved with DpnI (to prove replication in animal cells) and HindIII (which cleaves once per amplicon unit), were analyzed by Southern blotting. The results (Fig. 3) can be summarized as follows. (i) All constructs replicated in the nuclei. (ii) The amplicon-6 construct was transmitted to the cytoplasm. Trace amounts of the delpac-1 and delpac-2 defective constructs, but not the oriLyt construct, were recovered in the cytoplasm. (iii) Only the intact amplicon-6 (pNF1158) could be recovered as secreted particles in the medium (Fig. 3B). (iv) To test whether amplicon-6 replication generated concatemeric DNAs, cytoplasmic DNA of cells transfected with amplicon-6 (Fig. 3C, lane 1) or the amplicon-6 delpac-1 (Fig. 3C, lane 2) were fully digested with DpnI and partially digested with the ScaI. The analyses revealed the presence of DpnI-resistant monomer, dimer, trimer, and higher-order packaged concatemers, which were not fully resolved in this type of gels. Almost no high-molecular-weight DNA was found in the amplicon-6 delpac-1 sample. These results verify that the constructed HHV-6 pac signal was functional in packaging the defective genome vectors in the cells, similar to previously constructed HSV-1 and HHV-7 amplicons (29, 32, 33, 64).

FIG. 3.

Transient packaging assays of amplicon-6 and deletion constructs. (A) Nuclear and cytoplasmic DNAs from transfected cells. (B) Cell-free virus DNA collected from the medium of the transfected cells. The samples were digested with the DpnI and HindIII enzymes. Plasmids included pNF1158 (lane 1), pNF1184 (lane 2), pNF1183 (lane 3), and pNF1164 (lane 4). The sizes of the replicating plasmids were as follows: pNF1158, 5.4 kb; pNF1184, 5.1 kb; pNF1183, 4.8 kb; pNF1164, 4.6 kb. (C) Cytoplasmic DNA of cells transfected with amplicon-6A and with amplicon-6A that had a deletion of pac-1. The intact amplicon-6 (pNF1158, sample 1) and delpac-1 (pNF1184, sample 2) DNAs were digested with DpnI and partially digested with ScaI.

Plasmid construction and purification.

All of the amplicon-6 final transgenes have the human cytomegalovirus (HCMV) promoter and the simian virus 40 polyadenylation signal and were prepared in Escherichia coli DH10B or E. coli K-12 GM2163 (DAM⁻ DCM⁻) bacteria by using a Nucleobond AX Plasmid Maxi-Prep kit (Macherey-Nagel). The green fluorescent protein (GFP) gene was removed from pEGFP-C3 (Clontech) and cloned into pBluescript II SK (Stratagene). The cleavage and packaging signals and the origin of replication (r-pac/orilyt fragment) were added later, generating amplicon-6-GFP, designated pNF1194 (see Fig. 5F). The gD gene was derived by PCR of the BamHI-J fragment of HSV-1 (F) (pNF417). Two PCR primers containing oligonucleotide tails with the AgeI and BclI restriction enzyme sites were used: sense (including the AgeI site), 5′-CAG CTT CAC G acc ggt AG GTC TCT TTT GTG TGG TGC-3′; and antisense (including the BclI site), 5′-GAT ACT AGC C tga tca GG GGT ATC TAG TAA ACA AGG-3′. These sites match the AgeI and BclI (shown in lowercase letters) bounding the CMV promoter and the simian virus 40 poly(A) signal of pNF1194. The amplicon-6-gD construct (pNF1215) was produced in E. coli K-12 GM2163 (DAM⁻ DCM⁻) competent bacteria. The gD fragment, digested with AgeI and BclI restriction enzymes, was ligated into the parallel sites of the pNF1194 fragment without the GFP gene. The resultant colonies were screened by PCR picking. A number of the positive colonies were sequenced and compared to the original sequence by using NCBI/BLAST. The matching plasmid amplicon-6-gD (pNF1215) contains the intact gD gene. To construct a secreted form of the gD gene, the transmembrane region of the gD gene was deleted by PCR, resulting in a protein of 327 amino acids instead of the original 394 amino acids of the intact gD gene. The gDsec antisense primer sequence, including the BclI site (lowercase letters) and stop codon (underlined), was 5′-ACT AGC C tga tca CT AGG CGT CCT GGA TCG ACG G-3′. The gDsec fragment was digested with the AgeI and BclI restriction enzymes and ligated into the parallel sites on the pNF1194 vector, resulting in the amplicon-6-gDsec (pNF1219).

FIG. 5.

Fluorescence microscope photographs of the J-JHAN T cells transfected with amplicon-6-GFP with or without helper virus at P0 and P1. (A.P0) Electroporated culture viewed 7 days p.t. In the left panel is a phase-contrast exposure combined with fluorescence; in the right panel is a fluorescence exposure. (B.P0 inf) Cells were transfected and, after 48 h, superinfected with the helper HHV-6A (U1102). They were viewed 7 days p.t. In the left panel is a phase-contrast exposure combined with fluorescence; in the right panel is a fluorescence exposure. (C.P1) Attempts to “passage” the transfected cultures, which did not receive the HHV-6A helper virus, showed no GFP expression. (D.P1 inf) The P0 transfection-superinfected vectors were passaged to new uninfected cells (shown 1 week later). (E.P1 cf) The P1 medium was filtered through 0.45-μm-pore-size filters, allowing passage of the virus but preventing cell passage. The filtered medium was used to infect new cells, which were found to express the GFP 1 week later. (F) Structure of the ampllicon-6-GFP (pNF1194).

Transfection and superinfection.

J-JHAN cells (400 μl at concentrations of 10⁷ cells/ml) in RPMI 1640 medium were electroporated with 50 μg of purified plasmid DNA in 4-mm-gap disposable cuvettes (BTX P/N 640) by one pulse at 250 V for 24 ms by using the electrocell manipulator ECM 395. The electroporated cells were incubated for 10 min on ice and then transferred to 5 ml of RPMI 1640 with 10% FCS and 50 μg of gentamicin/ml at a final concentration of 8 × 10⁵ cells/ml. At 24 to 48 h after electroporation, the cells were mixed with equal numbers of HHV-6A (U1102) fully infected cells or with concentrated virus obtained by ultracentrifugation of the HHV-6A (U1102)-infected cell medium. The cultures were harvested 5 to 6 days later for further passaging and protein extraction.

Gel electrophoresis and Western blots.

The electroporated and/or infected cells were harvested and lysed in 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.5% NP-40, and protease inhibitors (Complete Protease Inhibitor; Roche). Electrophoresis was done in sodium dodecyl sulfate-10% polyacrylamide gels. The proteins were transferred to nitrocellulose membrane (Schleicher & Schuell) and immunoblotted with mouse H-170 gD anti-gD immunoglobulin G (IgG; 1:500). The antibody was provided by Lenore Pereira, Department of Stomatology, School of Dentistry, University of California, San Francisco. The secondary antibody was peroxidase-conjugated goat anti-mouse IgG (Jackson). The membranes were reacted by using enhanced chemiluminescence (Pierce).

TCA protein precipitation.

At 24 to 48 h before the precipitation of secreted proteins, the RPMI 1640 medium was replaced by Bio-Ram-1 protein free medium. The cells were removed by centrifugation, and the medium was filtered through 0.45-μm-pore-size filters. Proteins were precipitated with 10% trichloroacetic acid (TCA) by using 2 μg of bovine serum albumin carrier per ml. The pellet was resuspended in 12 to 20 μl of loading buffer containing β-mercaptoethanol, and 0.5 to 7 μl of 1 M Tris (pH 8.0) was added until the sample turned blue.

Analyses of GFP expression in lymphocytes.

Cell samples were rinsed with phosphate-buffered saline (PBS) and placed on glass slides coated with poly-l-lysine (1 mg/ml). The cells were fixed for 15 to 20 min with 4% paraformaldehyde. After a rinse in PBS, Galvanol mounting reagent was added, and the slides were viewed with a Zeiss Axioscope fluorescence microscope. Camera photographs were obtained by using MC-100 camera.

Confocal microscope analyses.

To determine the location of expressed gD and gDsec proteins in the cells, cell samples were concentrated, rinsed with PBS, and placed on glass slides coated with poly-l-lysine (1 mg/ml). After fixation with 4% paraformaldehyde, the cells were perforated by treatment with 0.1% Triton X-100 and rinsed with PBS. The slides were blocked with 20% FCS in PBS to reduce background and then incubated for 30 min with the gD H170 antibody, followed by the addition of Cy3-conjugated Goat anti-mouse IgG. After a rinse in PBS, Galvanol mounting reagent was added, and the slides were covered with a coverslip prior to viewing them in an Axiovert 135M confocal microscope (Carl Zeiss) equipped with an argon-krypton laser and a ×100 objective lens. Excitation was at 488 and 568 nm. The contrast and intensity for each image were manipulated uniformly by using Adobe Photoshop software.

FACS analysis.

Aliquots of 10⁶ J-JHAN cells were rinsed twice with PBS containing 2% FCS. The cells were stained with the H-170 mouse anti-gD IgG (1:200), followed by the goat anti-mouse R-phycoerythrin-conjugated secondary antibody (Jackson ImmunoResearch Laboratories). After a rinse with PBS containing 2% FCS, the cells were fixed with 2% paraformaldehyde in PBS and examined in a fluorescence-activated cell sorting (FACS) analyzer (Becton Dickinson).

RESULTS

Construction of amplicon-6-GFP (pNF1194) as a model for amplicon-6 vector propagation in T cells with or without HHV-6A helper virus.

To monitor the propagation of the amplicon-6 vector in CD4⁺ T cells and to test gene expression capability with or without helper virus, the amplicon-6-GFP construct (pNF1194) containing the HHV-6 DNA replication origin and cleavage and packaging signals was introduced into J-JHAN cells by electroporation (Fig. 4 and 5F). As schematically diagrammed in Fig. 4, a portion of the culture was removed 2 days posttransfection (p.t.) and mixed with cells infected with the U1102 helper virus at the peak of infection. The electroporated and electroporated-superinfected cultures, termed passage 0 (P0), were incubated for five additional days, after which the cultures were viewed for GFP expression and used for further propagation by adding new cells, producing P1. Analyses of the P0 cultures have demonstrated GFP expression, which was enhanced about four- to fivefold in superinfected cells (Fig. 5A.P0 and B.P0 Inf). Furthermore, the HHV-6A (U1102) superinfected P 1 cells also showed significant GFP production (Fig. 5D.P1 Inf). In contrast, when the electroporated cultures receiving amplicon plasmid only were “passaged” without added helper virus, there was no detectable GFP made (Fig. 5C.P1). We conclude that in the presence of the helper virus, the cell-associated vectors could be continuously propagated. In addition, the GFP amplicons were secreted into the medium as cell-free infectious virions; when the medium was filtered through 0.45-μm-pore-size membranes, followed by ultracentrifugation to concentrate the virus, the resultant virions could be used to infect new J-JHAN cells, which showed GFP expression 7 days later (Fig. 5E.P1 cf). In contrast, the medium of the P0 cultures receiving amplicon plasmid only did not contain any filterable materials transmitted to new cells (data not shown).

FIG. 4.

Experimental protocols of cell-associated and cell-free amplicon-6-GFP. J-JHAN T cells were electroporated with the amplicon-6-GFP plasmid. Two days later, some of the cells were superinfected with the helper HHV-6A (U1102). The transfected-superinfected cells (P0) were examined for GFP expression and passaged by the addition of uninfected cells, producing P1 vectors. Vectors secreted into the medium at P0 were collected by filtration at 0.45-μm-pore-size membranes to produce cell free virions, which were further passaged in uninfected cells, producing cell-free P1 vectors. The analyses and passaging were repeated by adding new uninfected cells, producing P2 virus. The cultures that were electroporated with the amplicon vector without added helper virus were similarly treated.

Expression of amplicon-6-gD mRNA in J-JHAN cells.

The expression of the gene in J-JHAN cells transfected with the amplicon-6-gD vector was tested by reverse transcriptase PCR (RT-PCR). The results showed that the electroporated cells expressed gD mRNA, similar to results obtained with Vero cells productively infected with HSV-1 (Fig. 6). In the absence of RT there was no PCR product.

FIG. 6.

Expression of amplicon-6-gD mRNA in J-JHAN cells. Expression of gD mRNA from cells electroporated with amplicon-6-gD at 24 and 48 h p.t. (lanes 1 and 2). Lanes 5 and 6 are identical to lanes 1 and 2 without RT. Lane 3, Vero cells infected with HSV-1; lane 8, RT was omitted; lane 4, plasmid DNA of the amplicon-6-gD vector.

Passaging and expression of the amplicon-6-gD vector.

The construct was electroporated into J-JHAN cells, and a fraction of the cells was superinfected with the helper virus. The cultures were tested for propagation and gD expression in Western blots and probed with gD H170 monoclonal antibody. A control of HSV-1-infected Vero cells was included in the test. The results (exemplified in Fig. 7) have demonstrated gD expression at 7 days postelectroporation with significant enhancement upon the HHV-6A (U1102) superinfection. In the P1 cultures tested for expression 1 week later, only the superinfected cultures showed gD expression, whereas the transfected cultures without helper virus did not contain detectable levels of gD. As described above for GFP, the filtered P0 medium receiving amplicon vector and helper virus contained cell-free virions, which were capable of infecting new cells. There was no detectable gD expression in the P1 cells of filtered medium in the absence of the helper virus. We conclude that the passaged virus retained both cell-associated and cell-free virions that are infectious to new cells, resulting in transgene expression.

FIG. 7.

Amplicon-6-gD expression. Control cultures of Vero cells infected with HSV-1 (lane 1), J-JHAN T cells infected with helper but no amplicon vector (lane 3), and amplicon-6-gD P0 transfection without or with helper virus (lanes 4 and 5) are shown. The medium of P0 amplicon-6-gD vector without or with helper virus was filtered, concentrated, and used to infect new cells generating cell free P1 (lanes 6 and 7). P1 of the cell-associated amplicon-6-gD vector without or with helper virus (lanes 8 and 9, respectively) are also shown.

Expression and secretion of amplicon-6-gDsec.

To construct a secreted form of the gD gene, the transmembrane region of gD was deleted by PCR, yielding a protein of 327 amino acids instead of the original 394 amino acids (Fig. 8A). The gD and gDsec encoded by the amplicon-6 vectors were electroporated into J-JHAN cells and produced proteins with estimated sizes of 60 and 45 kDa, respectively, as shown in a Western blot probed with the gD antibody (Fig. 8B). The expression of gDsec was tested in P0 up to P2 cultures with or without superinfecting helper virus (Fig. 9). Similar to the results with the intact gD, the P0 transfected-superinfected cultures showed significantly higher expression (Fig. 9, lanes 3 and 4), and the medium contained filterable amplicon-6-gDsec virions, which could be transmitted to new uninfected cells (Fig. 9, lane 5). In the presence of the helper virus, the gDsec vector could be further passaged, generating passages 1 and 2 (Fig. 9, lanes 7 to 8), whereas there was almost no gDsec expression in the cultures without helper virus (Fig. 9, lane 6).

FIG. 8.

Structure and expression of amplicon-6 vectors containing the gD or gDsec genes. (A) Vectors pNF1215 containing an intact gD gene driven by the HCMV promoter and pNF1219 containing a 201-bp deletion of the transmembrane signal. (B) Expression of amplicon-6-gD and gDsec in the electroporated J-JHAN T cells: Western blots containing proteins of control HSV-1-infected Vero cells (lane 1) and proteins from J-JHAN T cells infected with helper HHV-6A (lane 3) or cells electroporated with amplicon-6-gDsec and amplicon-6 intact gD (lanes 4 and 5), respectively. The blots were probed with H170 anti-gD monoclonal antibody.

FIG. 9.

Expression of amplicon-6-gDsec in J-JHAN T cells. Western blots containing (i) control Vero cells infected with HSV-1 (lane 1), (ii) J-JHAN cells electroporated with amplicon-6-gDsec without or with helper virus (lanes 3 and 4), (iii) filtered medium of the P0 amplicon-6-gDsec vector with helper virus, used to infect new cells, which were tested for gDsec expression 7 days later (lane 5), and (iv) P1 of the cell-associated amplicon-6-gDsec propagated without or with helper virus (lanes 6 and 7) are shown. Also shown is P2 of the amplicon-6-gDsec vector superinfected cells (lane 8). The blots were probed with the H170 anti-gD monoclonal antibody.

Protein secretion into the medium.

The extracellular secretion of gD and gDsec proteins in transfected J-JHAN cells was tested by TCA precipitation of the culture media. The results (Fig. 10) revealed that the gDsec protein could be recovered by TCA precipitation of the culture medium, which was collected 2 and 7 days p.t. The electrophoretic mobility was similar but not identical to the non-TCA precipitated cultures, and the bands appeared higher in the blotted gel. Significantly more TCA-precipitable gDsec protein was recovered from the medium of cells, which were superinfected with the helper virus (Fig. 10, lane 6). Analyses of the cultures that received the amplicon-6 containing the intact gD, revealed also TCA-precipitable protein in the medium, which was smaller than the intact protein, prepared from the cells (Fig. 10, lanes 8 and 9 versus lane 7). It is possible that the membrane associated gD at the cell surface was fragmented and accumulated in the medium.

FIG. 10.

TCA precipitation of amplicon-6-gDsec and amplicon-6-gD from medium of J-JHAN cells with or without HHV-6 helper virus. Western blots contained control Vero cells infected with HSV-1 (lane 1) and J-JHAN cells transfected with amplicon-6-gDsec 2 days postelectroporation (lane 3). The medium of P0 transfected gDsec culture was TCA precipitated 2 and 7 days p.t., without helper virus (lanes 4 and 5) or 7 days postsuperinfection with helper virus (lane 6). The electrophoretic mobility was similar but not identical to the non-TCA-precipitated cultures. Lanes 7 to 10 show transfections with amplicon-6-gD, including electroporated cells (lane 7) and TCA-precipitated medium of the P0 electroporated amplicon-6-gD, 2 and 7 days p.t. without helper virus (lanes 8 and 9) or with helper virus (lane 10).

Confocal microscopic analyses.

To test the cellular distribution of the amplicon-6-gD and gDsec proteins, the transfected J-JHAN cells were processed for viewing in the confocal microscope by using the gD monoclonal antibody H170. For viewing the gDsec protein, the cells were perforated with Triton X-100 before staining. The results can be summarized as follows. A significant fraction of the cells, which were dually infected with the amplicon-6-gD vector and the HHV-6A (U1102) helper virus, expressed the protein at the cell surface (Fig. 11). When infected with amplicon-6-gDsec these cells showed accumulation of gDsec in the cytoplasm (Fig. 12), most likely in the endoplasmic reticulum and the Golgi apparatus, as known for secreted proteins. We conclude that the majority of the confocal microscope images had global appearances of the gD and gDsec proteins, as predicted from their structures. Some of the cells showed infected cell cytopathic effect, representing dual viral and amplicon infections.

FIG. 11.

Confocal microscopic images of U1102-infected J-JHAN cells transfected with amplicon-6-gD plasmids (pNF1215). Cells were transfected with the amplicon-6-gD and viewed with a confocal microscope using the H170 anti-HSV gD antibody. Each part of the figure is composed of a fluorescent photo (upper left), a differential interactions contrast (Nomarsky) photo (upper right), and a superposition of the fluorescent and Nomarsky photo (lower left). (A) HHV6A (U1102) mock-infected J-JHAN cells. (B to F) Representative images of J-JHAN cells transfected with amplicon-6-gD and superinfected with HHV6A (U1102).

FIG. 12.

Confocal microscopic images of cells transfected with the amplicon-6-gDsec (pNF1219). (A to F) Representative images of confocal microscope analyses of HHV6A (U1102)-infected J-JHAN cultures that are transfected with amplicon-6-gDsec plasmids (pNF1219), perforated, and viewed by using H170 anti-HSV gD antibody. Each part of the figure is composed of a fluorescent photo (upper left), a differential interactions contrast (Nomarsky) photo (upper right), and a superposition of the fluorescent and Nomarsky photo (lower left).

Flow cytometry of amplicon-6-gD with or without the helper virus.

To quantify gD expression, duplicate cultures of J-JHAN cells were electroporated with the amplicon-6-gD construct, with or without the superinfecting cell-free helper virus. The control cultures included uninfected cells and cells that were infected with the helper virus only, without the amplicon-6-gD vector. At 6 days p.t. the cultures were analyzed by flow cytometry. The results can be summarized as follows. (i) There was marginal fluorescence in the control cultures corresponding to a mean fluorescence intensity (MFI) of 3.8 in the uninfected cells (Fig. 13A) and an MFI of 6.2 in the HHV-6A (U1102)-infected culture (Fig. 13B). (ii)The amplicon-6-gD electroporated culture were estimated to contain two populations: one corresponding to the background fluorescence (MFI of 4.7) and a second population estimated to be 16.67% of the cells, expressing gD at MFI of 20.2 (Fig. 13C). (iii) In the amplicon-6-gD superinfected culture, the majority of the cells were found to be large, due to the infection, and to represent two populations. The first (labeled as R1) had marginal MFI values and most likely represented infected cells that did not receive the amplicon-6-gD vector. The second (labeled as R2) corresponded to 62% of the culture, with an MFI of 296.4 (Fig. 13D). Since only ca. 16% of the cells were transfected, it is apparent that the majority of J-JHAN cells expressing gD represented the spread of newly synthesized amplicon-6-gD defective viruses. Furthermore, the level of gD expression, in the superinfected culture, as measured by the MFIs was ∼15-fold higher than the culture that was transfected without the helper virus (Fig. 13E). We conclude on the basis of the FACS analyses that the addition of the helper virus resulted in virus spreading, as well as the amplification of the transgene expression per cell.

FIG. 13.

Dot plot of flow cytometry of amplicon-6-gD transfected J-JHAN cells with or without superinfecting helper virus. (A) Cultures of uninfected J-JHAN cells; (B) cultures infected with helper virus only; (C) cells electroporated with amplicon-6-gD; (D) cells receiving both the amplicon-6-gD and the helper HHV-6A (U1102). (E) MFI values for the different cultures.

DISCUSSION

Amplicon-6-GFP, gD, and gDsec.

We have described the use of the amplicon-6 vectors for the introduction of selected transgenes into human T cells. The electroporation and passaging efficiency of the vectors in J-JHAN T cells was measured with or without the helper HHV-6A. Transfection of T cells by electroporation with naked amplicon-6 DNA resulted in GFP, gD, and gDsec expression, as measured by microscopy and flow cytometry. Gene expression was documented by RT-PCR and by Western blot analyses. As clearly quantified by FACS analyses, the addition of the helper virus increased the spread of the transgene from 17 to 62% of the cells. Furthermore, the expression level per infected cell was enhanced more than 15-fold, most likely reflecting the formation of large concatemeric DNA molecules with reiterated oriLyt, packaging signals, and the gD transgene. The generated amplicon-6 particles could be transmitted as cell associated vectors, as well as cell free virions secreted into the medium for at least two passages, whereas no gene expression was noted in parallel cultures receiving no helper virus.

Expression of cell surface and secreted proteins.

The confocal microscopy and flow cytometry revealed that the amplicon-6-gD was expressed on the lymphocyte cell surface. In contrast, the gDsec protein accumulated at intracytoplasmic locations and was secreted out of the cells as documented by TCA precipitation of the medium. Interestingly, when the intact gD gene was used, gD was also recovered from the medium as a TCA-precipitable protein but at a lower molecular weight, as if it was fragmented.

Amplicon-6 host range.

HHV-6 uses CD46 as a cellular receptor (14, 67), enabling entry into a wide range of cells, including mature T lymphocytes (18, 31, 39, 62), lymph nodes (47), macrophages, monocytes (44), dendritic cells (5, 41), kidney tubule endothelial cells (58), and salivary glands (28, 39), as well as CNS (50, 62). The HHV-6-based vectors appear to be well suited for transfer of genes into lymphocytes, known to resist most common transfection methods, including calcium phosphate precipitation, electroporation, the use of DEAE-dextran, and lipofection (11, 12, 66). Several groups have reported that lymphocytes could not be efficiently transduced by using adenovirus vectors (11, 12, 54, 78). In their studies of 33 different lymphocytic cell lines, Meeker et al. (54) monitored adenovirus vectors carrying the β-galactosidase marker and showed that only a limited number of cell lines had significant staining, whereas the majority of tested cell lines had low expression efficiency. Five different T-cell lines tested showed almost no expression.

Retroviral vectors, such as Moloney murine leukemia virus, are commonly used to express genes in T lymphocytes (27). However, the expression levels are often unsatisfactory (8, 84). Significantly improved vectors have been recently described by Engels et al. (26). However, retrovirus vectors might have disadvantages due to their integration into the host chromosomes, which might cause hazardous disruption or activation of host gene expression (26, 40, 51).

Advantages and potential applications of amplicon-6 vectors.

There are several advantageous features of the amplicon-6 vectors, including (i) efficient replication due to reiterations of cis-acting replication signals; (ii) large gene capacity and transgene reiterations; (iii) stability of gene expression (under the HCMV promoter) for at least 7 days postsuperinfection and continuation of gene expression after one or two passages; (iv) no integration into the host genome, avoiding potential insertional mutagenesis; (v) the ability to target dividing and nondividing cells; and (vi) inhibition of HHV-6 by ganciclovir (62), which increases safety in potential gene therapy.

The efficient expression of transgenes in hematopoietic cells could be advantageous for numerous applications, including molecular biology studies of selected T-cell-related genes and functions gene therapy and/of T cells employing suicidal genes, cytokines, and costimulatory molecules. Finally, of major relevance to the studies presented here is the ability of the vector to infect hematopoietic cells, including B cells and dendritic cells, for potential use in vaccination. Further experiments are planned to test the ability of the amplicon-6-gD and gDsec to induce cellular Th1 and humoral Th2 immune responses in susceptible animals.