Tamplicon-7, a Novel T-Lymphotropic Vector Derived from Human Herpesvirus 7

Abstract

We describe the derivation of a novel T-cell-defective virus vector employing the human herpesvirus 7 (HHV-7). The new vector, designated Tamplicon-7, replicates in CD4⁺ T cells. The system is composed of a helper virus and defective virus genomes derived by the replication of the input Tamplicon vector. There are two cis-acting functions required for the replication and packaging of the defective virus genomes in the presence of the helper virus: the viral DNA replication origin and the composite cleavage and packaging signal, which directs the cleavage and packaging of defective virus genomes. Viral DNA replication is compatible with the rolling circle mechanism, producing large head-to-tail concatemers of the Tamplicon vector. Thus, in the presence of the helper virus, the replicated vectors are packaged and secreted into the medium. Furthermore, we have shown that the vector can be employed to express a foreign gene, encoding the green fluorescent protein, in the T cells infected with the HHV-7 helper virus. We predict that the Tamplicon-7 vector might be potentially useful for gene therapy of diseases affecting the human CD4⁺ T cells, including autoimmune diseases, T-cell lymphomas, and AIDS.

Human herpesvirus 7 (HHV-7) was initially isolated in our laboratory from activated CD4⁺ T cells purified from the peripheral blood of a healthy 26-year-old individual (strain RK) (10). It has been found to be a ubiquitous virus which infects the majority of children when they are 3 to 6 years old (17, 34). The virus then becomes latent in peripheral blood mononuclear cells. It has been suggested that HHV-7 may cause roseola infantum (or exanthem subitum). Although HHV-7 was isolated from a roseola infantum patient (30), the vast majority of exanthem subitum isolates correspond to HHV-6 (36, 37), and it is possible that the virus was indirectly associated with the disease. Specifically, primary HHV-7 infections were found to result in a simultaneous increase of HHV-6 antibody titers, most likely reflecting the reactivation of HHV-6 from latency in peripheral blood mononuclear cells (16). It has been suggested that the reactivated HHV-6B could cause roseola infantum (16). Thus far, there is no disease known to be directly associated with HHV-7. Additionally, the virus can be found in the saliva of more than 80% of healthy individuals (3, 15, 35).

Latent HHV-7 can be induced to begin replication in vitro by T-cell activation (2, 10, 11, 16). It has been shown that the CD4 marker is a critical component of the HHV-7 receptor (20). This, most likely, is the basis for targeting CD4⁺ cells, although further analyses are required to determine that no other receptor(s) can mediate viral adsorption and entry. Also, the virus has been found to interfere with human immunodeficiency virus (20). Furthermore, it has recently been shown that HHV-7 could compete with human immunodeficiency virus type 1 in macrophages also carrying the CD4 surface moiety (4).

The entire HHV-7 DNA has been sequenced, including the JI and, recently, RK strains (22, 24). Reassessment of the genetic content of HHV-7 has indicated that the HHV-7 genome contains 84 different genes. The organization of the viral genes appears to be similar to the corresponding arrangements of HHV-6 genes. Fragments of HHV-7 were determined to have 50 to 60% nucleotide sequence similarity to those of HHV-6 DNA (22, 24, 27).

The construction of the Tamplicon-7 vector was based on aspects of viral DNA replication and packaging. Specifically, the HHV-7 genome is a linear molecule of 150 kb composed of a long unique DNA sequence (U) of 133 kb flanked by direct repeats (DRs) DR_L and DR_R and arranged as DR_L-U-DR_R (25, 27). This resembles the DNA structures of HHV-6 (14, 19, 21), the channel catfish virus (5), and equine herpesvirus 2 (31). The DRs range in size from 6 to 12 kb in different HHV-7 strains (27a). HHV-7 DNA replication is compatible with the rolling circle mechanism, producing large concatemeric genomes which are cleaved approximately a genome length apart, as determined by the locations of the DR elements. The exact cleavage sites are strictly determined by the nucleotide-measuring functions of the pac-1 and pac-2 signals (6; for a review, see reference 13). The DR elements within the HHV-7 genome are bound by the pac-1 and pac-2 signals, which determine the sites for the cleavage of replicated viral DNA. Cleavage occurs 44 and 33 bp from the pac-1 and pac-2 signals of HHV-7, respectively. The resulting cleaved DNA is packaged as has been shown for herpes simplex virus (HSV) and other herpesviruses (6, for a review, see reference 13). Additional units within the DR elements of HHV-7 and HHV-6 are telomeric-type reiterations featuring GGGTTA or other variations, which are reiterated various numbers of times in different viral strains (26, 27a, 32).

The Tamplicon-7 vector system.

The defective HHV-7 vector system was designated Tamplicon-7 to delineate the capability of the vector to replicate in T cells. The system has features similar to those of the HSV vector amplicon (12, 18, 29). Specifically, the Tamplicon-7 system consists of a mixture of helper virus and defective virus genomes generated from engineered Tamplicons. Two cis-acting functions are required for the propagation of the defective virus genomes in the presence of the helper virus: the viral DNA replication origin (oriLyt) and the composite pac signal required for the cleavage and packaging of the viral genome. The helper virus contributes, in trans, the replication and packaging machinery, such as DNA replication enzymes, packaging functions, and the proteins and glycoproteins participating in the buildup of the structural virions. The studies concerning the cis-acting signals required for propagation of the defective genomes are briefly summarized below.

HHV-7 oriLyt.

The lytic replication origin, oriLyt, of HHV-6 DNA was identified in HHV-6B(Z29) by Dewhurst and coworkers (7). Figure 1 displays the localization of the DNA replication origin of HHV-7(RK). Based on the alignment of the HHV-6A(U1102) and the HHV-7(JI) nucleotide sequences, we estimated that the oriLyt of HHV-7(RK) is located in the region between the major DNA binding protein (U41) and the conserved herpesvirus transactivator (U42). To obtain a functional replication origin, cytoplasmic DNA was prepared from Sup-T1 cells (1, 28) which were infected with HHV-7(RK), and a cloned library of BsrFI-cleaved DNA was derived. From the generated clones, a 1.9-kb clone (pNF1166) was found to contain the putative viral DNA replication origin. Furthermore, as shown below, a 1-kb subclone, pNF1168, was shown to replicate efficiently in helper virus-infected cells.

FIG. 1.

Cloning of oriLyt of HHV-7(RK). Virion DNA was extracted from the cytoplasmic fraction of HHV-7(RK)-infected Sup-T1 cells and was used to generate a cloned library. BsrFI-cleaved viral DNA fragments were cloned into a dephosphorylated Bluescript plasmid previously cleaved with XmaI. (a) In order to identify colonies containing the expected 1.9-kb BsrfI insert, the library was screened with the two PCR primers RH1 (5′-CCGAAACAACAGTTTCATTATC-3′) and RH3 (5′-AAAGAAGTTGATTCTATAGATTTTGAA-3′). The primers were synthesized from the two open reading frames flanking the putative location of HHV-7 oriLyt (orf U41, a major DNA binding protein, and orf U42, a herpesvirus-transactivating protein). Taq DNA polymerase (AB) and Taq extender (Stratagene) were used, and 30 cycles of amplification were performed under the following conditions: 94°C for 1 min, 59°C for 1 min, and 72°C for 1 min. PCR was performed on all the colonies, and positive colonies showed a specific 700-bp PCR product. (b) One of the positive clones (pNF1166) was cleaved with HpaI and BpmI, and after blunting of the BpmI site, a 1-kb fragment was cloned into the EcoRV site of a Bluescript plasmid, yielding pNF1168. Shown are map coordinates, in kilobases, for the positions of clones and primers that were used.

HHV-7 oriLyt replication operates by rolling circle mechanism, generating large concatemers of replicated DNA.

To test the replication ability of pNF1168, 10⁷ Sup-T1 cells were infected for 7 days with the helper virus HHV-7(RK). The infected cells were then exposed to electroporation with the pNF1168 clone by using a Bio-Rad gene pulser at 300 V and 960 μF in 0.8 ml of phosphate-buffered saline (PBS) without Ca²⁺ and Mg²⁺. One week later, the culture was harvested and the total cell DNA, prepared as previously described (8), was digested with DpnI in order to discriminate between replicated viral DNA and the unreplicated input plasmid DNA. DpnI cleaves methylated dam-type (GATC) DNA that has replicated in bacteria but not unmethylated DNA that has replicated in animal cells. As seen in Fig. 2, the cells contained large concatemeric DpnI-resistant DNA molecules, which represented the replication progeny of the input oriLyt plasmid. Specifically, the DpnI-resistant replication product consisted of high-molecular-weight concatemeric DNA (Fig. 2, lane 2) that, upon partial digestion with XhoI (lanes 3 to 8), could be converted into a ladder of DNA fragments with sizes representing multimers of the linear pNF1168 clone. Full XhoI digestion resulted in the complete conversion of such multimeric products into unit-length molecules (Fig. 2, lane 9). We have concluded that the 1-kb segment contained a functional replication origin that operates by the rolling circle mechanism. It is noteworthy that at the time we analyzed the oriLyt of HHV-7(RK), van Loon et al. (33) identified the location of oriLyt of HHV-7 (strain R-2) in a PCR product.

FIG. 2.

Transient-replication assay of oriLyt. The mechanism of replication of the 1-kb oriLyt construct (pNF1168) was shown to be compatible with the rolling circle mechanism. In lanes 2 through 9, DNAs were cleaved overnight with DpnI and partially digested with XhoI (7 U per sample). In lane 10, the DNA was cleaved overnight with DpnI and BglII to verify that the plasmid did not replicate by integrating into the viral genome. There are no BglII sites in the construct but there are several sites in the HHV-7 genome. As seen in lane 10, the high-molecular-weight DNA was not cleaved by BglII. As the negative control, 10⁷ uninfected Sup-T1 cells were transfected with pNF1168, and the extracted DNA was digested overnight with DpnI and XhoI (lane 11). Southern blot hybridizations were performed with a nonradioactive digoxigenin-dUTP-labeled Bluescript probe DNA (Boehringer Mannheim). Shown are the positions of the partially digested concatemeric DNA (arrowheads) and the times of XhoI digestion. A time of 0 min corresponds to a sample taken before the addition of XhoI. Numbers on the sides are molecular sizes, in kilobases.

The cleavage and packaging signal of HHV-7.

As outlined above, the pac-1 and pac-2 signals of HHV-7(RK) are located within the DR segments several kilobases away from each other. The circularization of the viral genome followed by DNA replication creates viral DNA concatemers with pac-1 and pac-2 properly oriented for measurement and cleavage. As shown in Fig. 3a, the PCR amplification of HHV-7(RK) employing pac-1 and pac-2 sequences as primers yielded a segment of 170 bp which was cloned and sequenced as displayed in Fig. 3b, which also shows its homology to HHV-7(JI). As shown in Fig. 3c, the arrangement of the conserved pac-1 and pac-2 motifs and their proximities to the genomic termini are consistent with those found in other herpesviruses (13). In consequence, cleavage was estimated to be 44 bp from the pac-1 element and 33 bp from the pac-2 element of HHV-7(RK), as is the case in HSV-1(F), HHV-6A(U1102), HHV-6B(Z29), and HHV-7(JI).

FIG. 3.

Cleavage and packaging signal of HHV-7. (a) Location of the primers (H7TER2 and H7TER1) used to amplify the DR_R-DR_L junction. The sequences of the primers were taken from the work of Secchiero et al. (26). (b) HHV-7(RK) total DNA was used as a template for PCR amplification with the H7TER2 and H7TER1 primers (26). The 170-bp PCR product was cloned and sequenced and compared to the complete sequence of HHV-7(JI). The conserved sequence elements of pac-1 and pac-2 are boxed. The sequences of the primers are underlined by arrows. (c) Pac-1 and pac-2 sequence homologues are shown for HSV-1(F), HHV-6A(U1102), HHV-6B(Z29), HHV-7(JI), and HHV-7(RK) (22–24, 32).

Generation of the constructed Tamplicon-7 vector.

Employing the 1-kb oriLyt segment in pNF1168 and the 170-bp pac segment in pNF1181, we produced the Tamplicon-7 vector in pNF1182, a vector capable of replicating in T cells. The three clones are schematically displayed in Fig. 4a.

FIG. 4.

The Tamplicon-7 vector system. (a) Derivation of the HHV-7 Tamplicon. The 1-kb oriLyt fragment was excised from pOrilyt (pNF1168) by digestion with SmaI and HindIII. The fragment was ligated to the HindIII and HincII sites of pNF1181 that contained the 170-bp pac fragment. The resulting plasmid that contained both elements was designated Tamplicon-7 (pNF1182). The 1.6-kb GFP gene was excised from the pEGFP plasmid (Clontech) and ligated to pNF1182 between the BamHI and PstI sites. The resulting plasmid was designated Tamplicon-7.GFP (pNF1196). Shown are restriction enzyme sites for EcoRV (EV), SmaI (Sma), HindIII (H), HincII (HcII), BamHI (B), and PstI (P). (b) Schematic structure of the packaged HHV-7 helper virus and the packaged defective Tamplicon-7. (c) Packaging assay of Tamplicon-7. Two independent infected cultures were electroporated with Tamplicon-7 and a third culture was electroporated with pOrilyt-7. Nuclear (nuc) and cytoplasmic (cyto) DNA preparations and DNA from purified virions prepared from the medium (med.) were extracted from all three cultures, digested with XhoI and DpnI, and hybridized as previously described (lanes 2 to 9). Lanes 1 and 10 contained a 1-kb DNA marker (M).

To test the constructed Tamplicon-7 vector system, Sup-T1 cells were infected for 7 days with HHV-7(RK) helper virus until infection was clearly apparent. The infected cells were then exposed to electroporation with plasmids containing oriLyt or the Tamplicon-7 construct. Seven days later, the cells were harvested and fractionated into the cytoplasmic and nuclear fractions as previously described (27). Briefly, the infected cells were treated with 0.6% Nonidet P-40 and Dounce homogenized, and the nuclei and cytoplasmic fractions were separated by centrifugation. To prepare the packaged DNA, 0.5% sodium deoxycholate, DNase I (50 μg/ml), and RNase (10 μg/ml) (all from Sigma) were added to the cytoplasmic fraction. In addition, purified virions were obtained from the medium of the infected cultures. The medium of infected cells was cleared of cells and cell debris by low-speed centrifugation (10 min, 3,000 rpm) followed by filtration through 0.4-μm-pore-size sterile disposable syringe filters (Corning). The filtered cell-free medium was centrifuged in a Beckman ultracentrifuge with a SW28 rotor (25,000 rpm, 3 h, and 4°C). The concentrated virions were digested with DNase I, and then the packaged DNA was extracted. All the DNA samples were digested overnight with DpnI and XhoI and were hybridized as previously described.

The results shown in Fig. 4c can be summarized as follows. (i) The Tamplicon-7 and oriLyt-7 plasmid DNAs were replicated in the Sup-T1 nuclear fraction (Fig. 4, lanes 2 to 4). Both types of plasmids yielded replicated DNA that was DpnI resistant. (ii) Analyses of the cytoplasmic DNA fractions revealed that Tamplicon-7 DNA was protected from DNase digestion (lanes 5 and 6). No DNase-protected DNA of the oriLyt-7 construct was recovered from the cytoplasm (lane 7). Thus, as expected, the cleavage and packaging signal present in the Tamplicon-7 was required for the transport of the replicated and encapsidated DNA into the cytoplasmic fraction. (iii) Tamplicon-7 DNA molecules were also recovered from the purified virion samples obtained from the medium (lane 9). Thus, in the presence of helper virus, the Tamplicon-7 vector was secreted into the medium as packaged in DNase-protected particles, enabling its further utilization as infectious virus. (iv) In addition, we prepared a chimera construct containing the HHV-6B(Z29) oriLyt and the HHV-7(RK) 170-bp pac segment. In the presence of HHV-6A(U1102) helper virus, this construct was found to be packaged (data not shown). The reciprocal chimera construct containing the HHV-7(RK) oriLyt and a cloned pac segment of HHV-6A(U1102) (27a), was also packaged in the presence of HHV-7(RK) helper virus (data not shown).

Expression of the GFP in Sup-T1 T cells.

To test whether the Tamplicon-7 system can be employed for gene transfer in T cells, we inserted the 1.6-kb green fluorescent protein (GFP) gene (pEGFP; Clontech) into the Tamplicon-7 vector (pNF1182). The GFP gene was constructed with the cytomegalovirus promoter and the simian virus 40 polyadenylation signal. The generated construct, Tamplicon-7.GFP (pNF1196 [Fig. 4a]), was tested for transfer and gene expression. Specifically, Sup-T1 cells were exposed to electroporation with 15 pmol of pNF1196 plasmid DNA under the conditions described above. After 24 h, the electroporated lymphocytes were mixed with equal numbers of HHV-7(RK)-infected Sup-T1 cells and incubated for 7 days. To further propagate the virus, the cultures were then mixed with uninfected Sup-T1 cells (at a ratio of 1:4). After 7 days, 100-μl samples were taken to test for GFP expression. The cells were washed with PBS, resuspended in 20 μl of PBS, and dried on eight-well slides. Fixation was done with a 3.7% formaldehyde solution and 100% cold acetone. A slow-fade–antifade kit from Molecular Probes (Eugene, Oreg.) was employed for mounting. The preparation was photographed by a Zeiss Axioskop fluorescent microscope. Figure 5 shows the phase contrast of the infected Sup-T1 cells exposed with fluorescence and of fluorescence exposure only. As can be seen in Fig. 5, GFP was expressed in the infected cells, most likely the cells which received both helper virus and the Tamplicon-7 vector virus (Fig. 5C and D).

FIG. 5.

GFP expression of Tamplicon-7.GFP vector in Sup-T1 cells. Sup-T1 cells with the plasmid pNF1196 and the HHV-7(RK) helper virus were photographed in a fluorescent microscope. (A, C, and E) Phase-contrast exposure combined with fluorescence; (B, D, and F) fluorescence exposure. Magnifications, ×270 (A, B, C, and D) and ×135 (E and F).

Potential use of the Tamplicon-7 vector in gene therapy.

We describe in this communication the establishment of a novel vector system replicated in CD4⁺ T cells. Two attractive features of the system are noteworthy with regard to the potential use of the system for gene therapy. First, the host range of HHV-7 might be limited to lymphocytes carrying the CD4 moiety, as the receptor for the virus has been suggested to reside within this cell surface component (20). It remains to be seen whether the CD4 moiety is the sole virus-related receptor or whether viral adsorption and entry into the cells might occur through interaction with a secondary, as-yet-unidentified, type of receptor(s). It will be of special interest to determine whether the vector will also target CD4⁺ macrophages as does the parental HHV-7. Second, HHV-7 is probably not involved directly in any known disease, as is supported by the fact that the virus is persistently present in the salivary glands of more than 80% of healthy individuals (15, 34, 35). Hence, the newly designated Tamplicon-7 vector is potentially attractive for future use in gene therapy for diseases afflicting the CD4⁺ T cells, such as autoimmunity, T-cell lymphomas, and AIDS. Current studies in our laboratory are geared towards developing a Tamplicon-7 vector system devoid of packaged helper virus by constructing a helper virus which does not have the packaging signal, potentially removing the ability of the helper virus to become packaged. One possible solution would be to transfect Sup-T1 cells with HHV-7 DNA in combination with the Tamplicon-7 plasmid DNA. This would produce, in the transfected cells, all the proteins needed in trans for the production of infectious Tamplicon-7 particles. The cleavage and packaging sequences would be deleted from the helper virus DNA prior to the transfection. A similar approach has been performed by Fraefel et al. (9) for a defective virus amplicon-like system derived from HSV-1.