A novel herpesvirus amplicon system for in vivo gene delivery - Nature

Gene Therapy (1997) 4, 1132–1141  1997 Stockton Press All rights reserved 0969-7128/97 $12.00

A novel herpesvirus amplicon system for in vivo gene delivery S Wang1, S Di1, W-B Young1, C Jacobson2 and CJ Link Jr1 1

Human Gene Therapy Research Institute, Iowa Health System, Des Moines; and 2School of Veterinary Medicine, Iowa State University, Ames, IA, USA

For gene therapy approaches to succeed, improved vector systems are needed that combine a large carrying capacity with high transduction efficiency in vivo. Towards this goal, we have developed a novel herpes simplex virus (HSV) amplicon vector, pHE, which contains an HSV-1 replication origin (ori S) and packaging sequence that permit vector replication and packaging into HSV-1 capsids. The vector also contains the Epstein–Barr virus (EBV) unique latent replication origin (ori P) sequence and a modified EBNA-1 gene to allow the vector to be maintained as an episome in

transfected E5 helper cells. This system allows for efficient packaging of high-titer vector since the E5 cells are first selected for the presence of the pHE vector before helper virus infection. The infectious pHE vector has efficient transgene expression in a variety of human cell lines in vitro. Stereotactic injection of pHE vector supernatant into the rat brain resulted in high, localized reporter gene expression. Finally, the pHE vector could carry a stable 21 kb DNA payload into HSV virions. This pHE vector system should have a broad range of gene transfer applications.

Keywords: viral vectors; gene therapy; herpes simplex virus; Epstein–Barr virus; amplicon

Introduction One central requirement of gene therapy is the need to develop improved gene transfer methods into somatic cells. Disabled viral vectors have shown promising results both in vitro and in vivo1–5 and several viral vectors are now in widespread use in phase I and II clinical trials. Viral vectors currently in trials have various limitations such as an inability to infect post-mitotic cells, low insert carrying capacity, inefficient transfer or immune destruction. Continued improvements in these standard vector classes are occurring, such as the recently described Lenti retroviral vector that can infect post-mitotic cells.6 None the less, these limitations have led investigators to explore alternative viral vector systems such as those based on the herpes simplex virus. Modified herpes simplex virus type-1 (HSV-1) gene transfer vectors are capable of infecting a variety of mammalian cells and can persist in a latent state in neural cells.7,8 HSV-1 is a large, enveloped, double-stranded DNA virus that is composed of approximately 152 kb encoding 81 genes.9 The HSV-1 genome contains three replication origins and encodes several proteins that modulate viral and cellular gene expression in a temporal cascade from three different transcription units, immediate–early (IE), early or late.10,11 Helper viruses have been generated by deleting IE genes essential for virus replication (eg IE3). These helper viruses can only replicate in helper cells that complement the missing IE gene product (ICP4 protein for IE3-deleted helper

Correspondence: CJ Link, Human Gene Therapy Research Institute, Iowa Health System, 1415 Woodland Avenue, Des Moines, Iowa 50309, USA Received 21 August 1996; accepted 3 July 1997

virus).12,13 The HSV vectors are engineered to carry transgenes in deleted portions of their genomes to allow expression in suitable host cells, without virus replication. These vectors have high titer (up to 1 × 109 p.f.u./ml), but often have significant host cell toxicity which is predominantly seen in dividing cells in culture and depends on which viral genes have been deleted.8,14,15 These vectors are the first of two types of HSV-based vectors under development. The second broad category of HSV-based vectors is amplicons.16 Plasmids containing a HSV-1 lytic replication origin (ori S) and the ‘a’ HSV-1 terminal packaging signal sequence, can be amplified and packaged into infectious HSV-1 virions in the presence of transacting helper virus. 16–19 This plasmid-based vector permits facile cloning and carries genomic information between prokaryotic cells and eukaryotic cells as a shuttle vector. The amplicon systems retain the merits of wide tropism of standard HSV-1 vectors, but the viral stocks tend to have lower titers. The titers of amplicon vectors can be increased by serial passaging of the amplicon viral stocks on the complementing cell line. 16 However, this process also increases the probability of generating wild-type virus within the amplicon stocks.8 In addition, the ratio of amplicon to helper virus changes with each passage. This ratio must be monitored at each round of production to identify optimal ratios and titers of amplicon. Theoretically, an amplicon vector could carry DNA inserts up to 140 kb in size, but a previously reported amplicon system was unable to package and maintain large-size DNA fragments (.15 kb).17 This vector instability may result from increased recombination during serial passages. Both HSV vectors and HSV amplicon vectors have demonstrated efficient gene transfer in vivo.8,20–24 Our novel amplicon vector, pHE, contains HSV-1 ori-

HSV/EBV amplicon vector for gene delivery S Wang et al

gin and packaging sequences that permit vector replication and packaging into HSV-1 virions. We, and another group, have constructed HSV amplicons that also contain Epstein–Barr virus (EBV) sequences that maintain the plasmid as an episome in the transfected cell nucleus.25 EBV has been demonstrated to contain a unique latent replication origin (ori P) which directs viral self-replication and maintenance in cells without entering the lytic cycle.26,27 The EBV nuclear antigen 1 (EBNA-1) encodes a DNA binding transactivator for ori P.26–29 Investigators previously have demonstrated that plasmid vectors containing the EBV ori P that also expressed the EBNA-1 gene were more effective eukaryotic expression vectors.26 Various groups have used such EBNA-1-based vectors for expression in human tumors with therapeutic intent.30 The combination of the HSV amplicon with the EBV sequences improves the ease of use of the HSV amplicon system. The new system overcomes the inconvenience of serial passaging to produce high-titer amplicon vector. Since helper cells can be transfected and selected for the pHE plasmid, a single round of helper virus infection produces relatively higher titers (up to 2 × 106 b.f.u./ml). Our replication incompetent pHE vectors maintained wide tropism for delivering transgene(s) into both dividing and quiescent cells with high efficiency both in vitro and in vivo. Furthermore, our improved vector could carry a stable 21 kb DNA insert.

Results The pHE vector The pHE700, herpes amplicon vector is illustrated in Figure 1. The vector contains the HSV-1 ‘a’ sequence for the package/cleavage signal and an ori S replication origin to permit plasmid replication and packaging in the presence of helper virus (d120) in E5 helper cells. The vector also contains an ori P and a DEBNA-1 gene which is con-

trolled by the Rous sarcoma virus (RSV) promoter. The DEBNA-1 gene sequence is a modified version of the native EBNA-1 sequence which has less cytotoxicity compared with the original EBNA-1 gene (personal communication with Dr B Sugden, University of Wisconsin). This modified gene permitted less cytotoxicity in E5 helper cells and in target cells. Hygromycin resistant gene (hyg+) was driven by HSV-1 thymidine kinase (HSVtk) promoter and terminated by HSVtk poly A signal. The pHE700-lac was generated by inserting a lacZ gene into the multiple cloning site in pHE700 driven by the human cytomegalovirus (CMV) immediate–early gene enhancer– promoter. The pHE700lB-GFP contains both a 16.8 kb BamHI DNA fragment of l phage and a 4.2 kb DNA fragment containing a humanized, red shifted green fluorescent protein (GFP).31

Episomal maintenance and viral packaging The maintenance of the pHE vector as an episome was demonstrated by transfection of pHE700-lac into E5 cells and selection with hygromycin (Figure 2). The initial transfection efficiency was generally ,10% as detected by 5-bromo-4-chloro-3-indolyl-b-d-galactopyranoside (Xgal) staining 24 h after transfection (Figure 2a). By 16 days after hygromycin selection, resistant colonies developed and most cells expressed b-galactosidase (Figure 2d). Selected E5 cells containing pHE700-lac plasmid were then infected with d120 helper virus. The resulting supernatants contain both the pHE700-lac vector and helper virus. The multiplicity of infection (MOI) of the helper virus added was between 0.01 and 0.1 to induce viral vector production within 24–36 h. The titer of d120 helper virus was defined as plaque forming units (p.f.u.) and the titer of pHE700-lac was defined as blue cell forming units (b.f.u.). The average titer obtained was 2 × 106 b.f.u./ml with a ratio of pHE700-lac vector (b.f.u.) to d120 helper virus (p.f.u.) that ranged from 1:1 to 1:20.

Figure 1 Plasmid map of pHE700 vector showing all structural elements. AmpR, ampicillin resistance gene; ‘a’, HSV-1 packaging/cleavage signal; HSV-tk promoter, HSV-1 thymidine kinase promoter gene; hyg+, hygromycin resistance gene; HSV-tk poly A, HSVtk polyadenylation signal; CMV, cytomegalovirus immediate–early promoter; MCS, multi-cloning site; SV40 poly A, simian virus 40 polyadenylation signal; RSV promoter, Rous sarcoma virus promoter; DEBNA-1, modified EBV nuclear antigen gene; ori P, EBV unique latent replication origin; ori S, HSV-1 replication origin; and Col E1, E. coli replication origin.

1133


1134

Figure 2 Transfection and selection of E5 cells in culture after transfection with pHE700-lac. (a) The b-galactosidase gene expression was detected 24 h after transfection by X-gal staining. E5 cells were transfected with pHE700-lac and 48 h later were placed into selection with 150 mg/ml of hygromycin. (b) Six days or (c) 10 days or (d) 16 days later, cells were fixed and stained with X-gal.

After centrifugation by sucrose gradient ultra-centrifugation, we obtained a pHE700-lac titer of 1 × 108 b.f.u./ml. Without hygromycin selection, however, the titers of pHE700-lac vector can only be generated at the range of 104 to 105 b.f.u./ml from the transfected E5 cells depending on transfection efficiency.

Transduction and expression in vitro of the pHE700-lac vector The pHE700-lac containing supernatants were used to transduce human target cells in vitro. All in vitro experiments were carried out by employing amplicon vector stocks with a 1:1 ratio of pHE700-lac vector to d120 helper virus. The b-galactosidase gene expression was evaluated after infection with pHE700-lac vector (3–10 MOI b.f.u.) in various cultured human cells including VA13 human fibroblasts (Figure 3a), T98G human glioblastoma cells (Figure 3b), IGROV human ovarian carcinoma cells (Figure 3c), and XP2OS xeroderma pigmentosum fibroblasts (Figure 3d). All cells were fixed and stained with X-gal 2 days after infection. When using a lower MOI (1 MOI of b.f.u.), the lacZ gene expression was detected for approximately 2 weeks, with a peak expression occurring 48–72 h after transduction. To demonstrate the cytotoxicity caused by helper virus, a low ratio of pHE700-lac vector to d120 helper virus (1:10) was

used to infect VA13 and T98G cells which delivered about 30 MOI of d120 helper virus when 3 MOI of pHE700-lac amplicon vector was applied. The cytotoxicity was evident 2 days after transduction. The transduced cells demonstrated altered morphology secondary to cytotoxicity (Figure 3e) and the number of the stained cells was greatly reduced due to cell loss (Figure 3f). To eliminate the possibility that the lacZ activity observed was from ‘pseudo-transduction’ secondary to b-galactosidase protein contamination of the packaged vector, a time course of lacZ gene expression was tested. VA13 cells were transduced by pHE700-lac vector, d120 helper virus or mock-transduced with culture medium. The cells were lysed 2 or 24 h later and the b-galactosidase activity was determined by O-nitrophenyl-b-d-galactopyranoside (ONPG) assay.32 The lacZ gene expression was only noticed 24 h after the pHE700-lac vector transduction. The 2 h time-point demonstrated a similar ONPG reading as compared with the VA 13 cells that were not transduced or transduced with d120 helper virus (Figure 4). Therefore, the b-galactosidase activity visible at the 24 h time-point reflects the transduced gene expression from the pHE700-lac vector.

The pHE vector insert capacity A 16.8 kb fragment obtained by BamHI restriction bacteria l phage genomic DNA and a 4.2 kb plasmid 31 that


1135

Figure 3 In vitro expression of the pHE700-lac vector. Photographs of various human cell lines are shown 2 days after infection with pHE700-lac virus stock. The cells transduced with 3 MOI (a, e and f) or 10 MOI (b–d) of pHE700-lac vector at the ratio 1:1 (a–d) or 1:10 (e and f) of the vector to helper virus. (a) VA13 normal fibroblasts, (b) T98G human glioblastoma cells, (c) IGROV human ovarian carcinoma cells, and (d) XP2OS xeroderma pigmentosum fibroblasts, (e) VA13, (f) T98G.

carries a humanized green fluorescent protein (GFP) gene were inserted into the pHE700 vector. The resulting plasmid, pHE700lB-GFP, was transfected into E5 cells and packaged by infection with CgalD3 helper virus. The packaged pHE700lB-GFP vector was then used to infect VA13 cells and green fluorescence was observed from the transduced cells (data not shown). To demonstrate further the stability of the amplicon packaged DNA, the

episomal vector DNA and the helper viral DNA were isolated from either the transfected E5 cells or the packaged virus stocks. The isolated DNA was digested with BamHI or XhoI and underwent pulsed field gel electrophoresis. The membrane was probed using a 32P-labeled 16.8 kb l BamHI fragment. An autoradiogram of this Southern analysis is shown in Figure 5a. Transfected pHE700lBGFP vector DNA is present as an episome in DNA


1136

maximum distance of 1 mm surrounding the site. No secondary sites of X-gal staining were detected. The majority of cells expressing b-galactosidase were neurons defined by the presence of typical neuronal processes in the caudate (Figure 6a) or by their location in the dentate gyrus and morphology characteristic of granular cells (Figure 6b and c). LacZ gene expression was observed for at least 2 weeks after injection with greatly decreased b-galactosidase activity (Figure 6d). Control rats injected with media alone showed essentially no X-gal-positive staining in these areas at 2 or 14 days after injection.

Discussion

Figure 4 ONPG test of b-galactosidase activities after pHE700-lac and d120 transduction. Packaged pHE700-lac and d120 were applied to VA13 cells and the ONPG test was performed 2 or 24 h after infection, respectively. The b-galactosidase activities were represented by optical density reading at 420 nm. Only cells transduced with pHE700-lac vector demonstrated b-galactosidase activity 24 h later.

harvested from E5 cells (Figure 5a, lane 1). Restriction analysis with XhoI results in a linearized intact vector of 31.6 kb (Figure 5a, lane 2), while restriction with BamHI results in a 16.8 kb band representing the l DNA insert (Figure 5a, lane 3). Figure 5a, lanes 4–6, are corresponding restriction analyses of DNA obtained from packaged pHE700lB-GFP vector. Lane 4 shows the unrestricted full-length vector with multiple copies of pHE700lB-GFP vector genome (152 kb) and lanes 5 and 6 represent XhoI and BamHI restriction digest, respectively. A schematic diagram of the pHE700lB-GFP vector is shown in Figure 5b. These results demonstrate that the pHE700 vector has the capability to carry large DNA inserts and transfer functional gene (GFP, in this case) into host cells through infectious packaged vector.

In vivo expression of the pHE700-lac vector in the rat central nervous system The efficiency of pHE700-lac vector for gene delivery into the central nervous system (CNS) was evaluated next. The caudate nucleus or hippocampus of rats were injected with pHE700-lac viral vector supernatant (2 × 105 b.f.u.) stereotactically. High b-galactosidase expression was found around the injection site within the caudate nucleus and the dentate gyrus (hippocampus) as determined by X-gal staining 48 h after injection (Figure 6). Gene expression in these brain areas was detected between 2 and 14 days after injection and was most prominent between 2 and 7 days. Intense X-gal staining was observed in cells at the injection site and to a

Our results demonstrate the utility of a novel herpesvirus-based amplicon vector system, pHE, for generating high-titer amplicon stocks, efficient transgene transfer into mammalian cells in vitro and in vivo and an improved carrying capacity. The vector is efficiently packaged in the presence of IE3 defective helper virus because the vector contains the HSV-1 package/cleavage elements and replication origin. Our vector also contains the DEBNA-1 gene and the EBV ori P to permit the vector to be maintained as an episome within the nucleus and to replicate with cell division. Our vector, in comparison with the other recently reported system,25 is relatively smaller in size and contains a modified EBNA-1 to enhance plasmid stability while causing less toxicity than the native EBNA-1 gene. This plasmid form allows facile cloning of transgene(s) into the vector. Only a single packaging cycle is needed to generate high titer of amplicon stock after selection for the episomally maintained amplicon in helper cells. The advantage of only one cycle amplicon vector production is a possible decrease in the probability of generating wild-type virus as observed with amplicon stock derived by the serial passaging process.8,14,16 We have not detected any wildtype virus in our pHE700-lac viral stocks. The self-replication permitted by the ori P and DEBNA-1 genes permit high titers of the viral vector production (approximately 1 × 108 b.f.u./ml after centrifugation). Previous reports in amplicon systems, found that three or four repeat cycles of helper virus superinfection were needed to achieve high vector titers.16,18,23 The hygromycin selection in our system results in almost all of the E5 cells containing pHE plasmid (Figure 2d) before the addition of d120 helper virus and results in a viral titer of 2 × 106 b.f.u./ml. In comparison, without hygromycin selection, the transfected E5 cells only generated titers of 104 to 105 b.f.u./ml after infection with helper virus. However, the ratio of amplicon vector to helper virus in both systems varied widely. We obtained ratios from 1:1 to 1:20 of amplicon to helper virus. Helper virus may replicate and package preferentially compared with amplicon vectors. The titer of helper virus used to package amplicon vectors was a critical factor, but we have not yet determined how to produce lower ratios of helper virus consistently. Using a 0.1 to 0.01 multiplicity of helper virus generally gave favorable results. The replication defective HSV helper virus (d120) in amplicon viral stocks causes substantial cytotoxicity to infected cells.8,14,15 This cytotoxicity correlates with a low ratio of amplicon vector to helper virus (Figure 3e and f).25 However, injection into rat brain with a total of 2 × 105 b.f.u. pHE700-lac vector demonstrated excellent in


1137

a

b

Figure 5 Autoradiogram of pHE700-lB-GFP vector and schematic. (a) Lanes 1–3 show analyses of DNA isolated from pHE700-lB-GFP plasmidtransfected E5 cells. Lane 1 was nonrestricted and represents episomal plasmid. Lane 2 was treated with XhoI that linearized the plasmid. Lane 3 was restriction digested with BamHI to drop out the 16.8 kb l phage insert. Lanes 4–6 show analyses of DNA isolated from packaged pHE700-lB-GFP vector supernatant. Lane 4 was nonrestricted and represents the full-length 152 kb vector genome. Lanes 5 and 6 show corresponding restriction digest analyses of the packaged pHE700lB-GFP DNA after XhoI and BamHI restriction digest, respectively. The marker is l phage DNA with HindIII digestion and MegaBase II (GibcoBRL). (b) Schematic representation of tandem copies of packaged pHE700-lB-GFP DNA with the restriction sites and expected fragments sizes shown. ‘a’: HSV-1 packaging/cleavage signal; lB: BamHI fragment of l phage.

vivo transduction and the transgene expression can last for at least 2 weeks (Figure 6). The higher ratio of amplicon to helper virus is a critical factor to be considered when using this system for gene delivery. Our pHE vector provides the ability to generate higher vector to helper virus ratio stocks compared with other amplicon systems. Developing even more favorable ratios of amplicon and helper virus, generating less cytotoxic helper viruses by deletion of more IE genes8 or using the newly developed helper virus-free packaging systems should further reduce or eliminate the cytotoxicity produced by amplicon vectors.33 Another major advantage of this novel amplicon system is the large DNA insert capacity compared with other viral vector systems. The pHE700lB-GFP contains both a 16.8 kb l phage DNA fragment and a 4.2 kb plasmid fragment containing GFP sequence. This vector was transfected in E5 cells and effectively packaged into HSV-

1 virions (Figure 5). The pHE700lB-GFP vector successfully transduced VA13 human fibroblasts and expressed GFP activity as noted by green fluorescence. Southern analysis of the harvested pHE700lB-GFP viral DNA demonstrated a stable insert carried by the vector and packaged into HSV virions. One 16.8 kb BamHI band was demonstrated after restriction of DNA from pHE700lBGFP virions (Figure 5). This is the first demonstration that amplicon vector could carry larger inserts without significant DNA loss compared with previously reported HSV-1 based amplicon vectors.17 Southern analyses on pHE700lB-GFP vector stocks after serial passage showed that the insert was stable for at least six passages (data not shown). One possibility is that our vector containing the ori P and DEBNA-1 elements may aid not only plasmid maintenance but also plasmid genome stability after transfection or during packaging. The ability to carry larger DNA inserts offers a potential advantage over size


1138

Figure 6 In vivo expression of pHE700-lac viral vector in rat brain. A total of 7 ml of viral stock solution containing 2 × 105 b.f.u. pHE700-lac amplicon vectors was injected. The brain was sectioned into 20 mm slices on a cryostat and stained with 0.1% X-gal solution. (a) The caudate putamen (× 100), 2 days after injection. (b) The hippocampus (× 100), 2 days after injection. (c) The hippocampus (× 40), 2 days after injection. (d) The hippocampus (× 40), 14 days after injection.

limitations of other currently used vectors. Larger cDNAs for complementing genetic diseases, such as the Duchenne muscular dystrophy gene (14 kb cDNA), might be transferred. This was recently demonstrated for a novel adenoviral vector propagated with helper virus,34 although the upper limit of such an adenoviral vector will be about 30 kb, while the theoretical capacity of the pHE700 amplicon vector may be about 140 kb. If realized, this would provide the opportunity for transfer of a gene in the genomic DNA form to human cells. The packaged pHE virions contain multimeric forms of the original monomeric vector after rolling-circle DNA replication (Figure 5b). The transgene copy number therefore depends on the insert size. The multiple copies of the transgene in the packaged vector provides severalfold more transcription units than vectors derived from the entire HSV viral genome. Accordingly, transgene expression was observed in tumor cells in as short a time as 5 h after pHE transduction of target cells (data not shown). The rapid, high-level transgene expression will provide an ideal system for studies of gene expression in vitro or in vivo in situations where short-term, rapid onset gene expression is desired. For efficient treatment of genetic or neurological diseases, long-term persistence of amplicon vector genome and constant transgene expression is needed. Our data and others indicate that

the vector carried lacZ reporter gene can be expressed for as long as 2–5 weeks25 in vitro and 2 weeks in vivo (Figure 6d). The transient gene expression observed may be due to decreased CMV promoter activity because the amplicon can persist at least for 2–3 months after in vivo transduction, as demonstrated in a recent study.35 Longterm transgene expression may be improved by replacing the CMV promoter with the transcriptional control mechanisms of the natural viral latency active promoter or HSV latency associated promoter.8 The vector might also be able to transfer a genomic gene construct containing an intact promoter, introns and other cis-acting elements to provide normal gene regulation by the cell. Our HSV amplicon vector may have utility for in vivo gene therapy, particularly into the CNS due to neurotropism.8,18,23,24 For many neurodegenerative diseases where therapies do exist, such as Parkinson’s disease, drugs which control symptoms often fail. This has led investigators to evaluate genetic therapy to treat these diseases. Current approaches to neurologic diseases with genetic therapies have been mostly restricted to ex vivo strategies with retroviral vectors; although, newly developed Lenti retroviral vectors can transduce nondividing neurons in vivo.6 HSV-based vectors are still attractive for gene transfer into the CNS. Long-term behavioral recovery in Parkinsonian rats has been reported after gene transfer with


a HSV vector.24 Several groups are also developing HSVbased vectors for antitumor therapy against gliomas.36,37 Like these HSV systems, our vector efficiently transduced a variety of human cell lines, and in addition, is easy to use and has an improved carrying capacity. These properties may allow advantages in human in vivo gene transfer experiments. For example, the pHE vector capacity would permit simultaneous expression of a number of different anticancer genes. These modified vectors may also be excellent for use in DNA vaccine strategies, allowing an immune response to the highly expressed transgene proteins.

Materials and methods Construction of pHE vectors The pHE700 is a 10 617 bp plasmid generated from the combination of components from several other plasmids. Plasmid pTO11 (the source for the ‘a’ packaging signal and the ori S of HSV-1, kindly provided by Dr N Stow, MRC, Glasgow, UK) was restricted with NspI and ligated to generate PTO11-NspI. Plasmid p500 (kindly provided by Dr Hayakawa, Kyushu University, Fukuoka, Japan) was then restricted with XmnI and HindIII, and the fragment containing the EBNA-1 sequence was ligated to a XmnI and HindIII fragment of PTO11-NspI. The resulting plasmid p501 was restricted with XhoI and XmnI and ligated to a XmnI and ClaI restricted fragment of p205 (kindly provided by Dr B Sugden, University of Wisconsin, USA). The resultant plasmid p206 was restricted with NarI and HindIII and ligated to pTO11-NspI restricted with HindIII and BamHI. The resultant plasmid was termed pHE100. The pHE100 was then restricted with HindIII followed by ligation to a XbaI and NruI fragment from pREP10 (Invitrogen, Carlsbad, CA, USA) containing the hygromycin resistance gene and the multi-cloning site (MCS) to create plasmid pHE600. The pHE600 was then restricted with SalI and the large fragment was isolated and ligated to the SalI fragment of pCEP4 (Invitrogen) containing an expression cassette which has a human cytomegalovirus (CMV) immediate–early gene enhancer–promoter and a simian virus 40 polyadenylation signal (SV40 poly A) to generate pHE700 vector. pHE700-lac was constructed by inserting a lacZ gene into HindIII and NotI sites of multiple cloning sites of pHE700. The lacZ gene was from a HindIII–NotI fragment of pCDMV3-lac (kindly provided by Dr T Tsukada, Kyoto University, Japan). The 31.6 kb pHE700-lB-GFP was constructed by ligation of a 16.8 kb BamHI fragment of l phage DNA (GibcoBRL/Life Technologies, Gaithersburg, MD, USA) and a 4.2 kb NheI linearized phGFP-S65T plasmid (Clontech, Palo Alto, CA, USA) into the BamHI and NheI sites of the pHE700 vector. Cells and viruses All cells were grown and maintained in DMEM (GibcoBRL) containing 10% fetal bovine serum (FBS; GibcoBRL), glutamine and penicillin/streptomycin, and incubated at 37°C in a humidified, 5% CO2 incubator. VA13 cells are SV40-transformed human WI38 fibroblast cells (ATCC, Rockville, MD, USA). XP2OS xeroderma pigmentosum cells were kindly provided by Dr Takebe (Kyoto University, Japan). T98G is a human glioblastoma cell line (ATCC). IGROV is a human ovarian carcinoma

cell line. The d120 (IE3 −) HSV helper virus and E5 helper cells were kindly provided by Dr Neal A DeLuca (University of Pittsburgh, PA, USA). CgalD3 is an IE3 − helper virus containing a lacZ expression cassette (kindly provided by PA Johnson, University of California, San Diego, CA, USA).14 E5 cells are resistant to G418 and are efficient hosts for ICP4 (IE3)-deficient viruses. All viral stocks were grown and titrated in E5 cells.

Viral vector production pHE700-lac was transfected into E5 cells with Lipofectamine per the manufacturer’s protocol (GibcoBRL). Cells were trypsinized and plated at 6 × 105 cells/100 mm dish 2 days after transfection. Transfected cells were placed under selection with Hygromycin B (ICN Biomedical, Aurora, OH, USA) 1 day after plating. The selected hygromycin-resistant E5 cells were super-infected with helper virus (d120) when resistant colonies appeared, generally after 2 weeks of drug selection. The colonies were trypsinized and 3 × 106 cells were plated on a 100 mm dish. When the cells were nearly confluent, 0.01 to 0.1 MOI of d120 helper virus in 1 ml of Opti-MEM (GibcoBRL) was added. The viruses were allowed to adsorb to the cells for 2 h at 37°C in a humidified, 5% CO2 incubator. The virus solutions were removed and 10 ml of DMEM with 10% FBS was added and incubated for an additional 24 to 36 h. The medium containing cell debris was collected and centrifuged and the vector supernatants were used for virus titration and infection experiments. Vector concentration Viral supernatant and cell debris were collected from infected cell monolayers after complete cell lysis. The lysate was frozen and thawed three times at 37°C and then centrifuged at 2400 g to pellet the cell debris. The supernatant was filtered through a disposable Nalgene 0.8 mm nylon filter bottle (Nalge, Rochester, NY, USA) to remove any remaining cell debris. Thirty-four milliliters of viral vector supernatant were added to a Beckman polyallomer Quick-Seal centrifuged tube (40 ml volume; Beckman, Palo Alto, CA, USA) on top of a 6 ml cushion of 25% sucrose in 1 × PBS. The tubes were then ultracentrifuged at 75 000 g at 4°C in a Beckman 55.2Ti rotor for 16 h to pellet the virus particles. The pellet was resuspended with 0.2–0.5 ml of sterile Hank’s balanced salt solution (HBSS) buffer for further applications. Vector titration The viral stocks were serially diluted in Opti-MEM media and placed on to confluent E5 cells monolayers in sixwell plates. The viruses were allowed to adsorb to the cells for 2 h at 37°C. The virus solutions were aspirated and the cells were washed with HBSS and overlaid with 2 ml DMEM containing 5% FBS and 0.3% methylcellulose and then incubated for an additional 3 days. Cells were fixed and plaques visualized by staining with 0.5% crystal violet for 10 min. The titers were reported as p.f.u./ml or b.f.u./ml. pHE700-lac viral stocks were titrated 24 h after infection of VA13 cells in 24-well plates. Cells were rinsed with HBSS and fixed for 5 min at room temperature in 2% formaldehyde, 0.3% glutaraldehyde in HBSS. Following two additional rinses with HBSS, cells were stained with chromophore solution containing 0.1% X-gal (Promega, Madison, WI, USA), 5 mm K4 Fe(CN)6 3H2O,

1139


1140

5 mm K3 Fe(CN)6 and 2 mm MgCl2 in HBSS. The X-galpositive cells (blue cells) were visualized 1 h after incubation.

ONPG test VA13 cells (2 × 105) were transduced by pHE700-lac vector (0.03 MOI), d120 helper virus alone (0.03 MOI) or mock-transduced by culture medium. Two and 24 h after transduction, the cells were lysed in an ONPG lysis solution (0.45 mm O-nitrophenyl-b-d-galactopyranoside and 0.5% Nonidet P-40 in HBSS), incubated at 37°C in a 5% CO2 humidified incubator, and the optical density (OD) at 420 nm was taken 1 h later. Three duplicate samples were plated for each dose. Southern analysis Ten milliliters amplicon viral supernatant was collected from pHE700-lB/GFP plasmid transfected E5 cells, 3 days after infection with helper virus. Cell debris was eliminated from viral stock by centrifugation at 2000 g for 15 min and virions were pelleted by 20% sucrose gradient ultra-centrifugation at 125 000 g for 15 h. Virion pellet was resuspended in lysis buffer (2.5 mm Tris (pH 8.0), 5 mm EDTA and 5 mm NaCl) and digested with proteinase K (0.5 mg/ml) in 1% SDS condition for 2 h. Viral DNA was phenol–chloroform extracted and precipitated. For the isolation of cellular episomal vector DNA, the pHE700-lB-GFP transfected E5 cells were treated with Hirt’s DNA extraction method.38 Isolated cellular and viral DNA were digested with BamHI or XhoI and run on a FIGE Mapper pulsed field gel electrophoresis system (Bio-Rad, Hercules, CA, USA) on 1% agarose gel at 14°C in 0.5 × TBE at 180–120 V using 0.4–2.0 s ramping pulse time for 20 h then 0.4–3.5 s ramping pulse time for another 20 h. The DNA samples in the gel were nicked under 60 mJ of UV radiation and alkali blotted to Sureblot nylon membrane following the manufacturer’s protocol (Oncor, Gaithersburg, MD, USA). The 16.8 kb BamHI fragment of l phage DNA was used as a probe labeled with 32P-dCTP by using Random Primed DNA Labeling kit (Boehringer Mannheim, Indianapolis, IN, USA). In vitro and in vivo gene transfer with pHE700-Lac Cultured human cell lines were trypsinized, counted and seeded. When cells were nearly confluent, pHE700-lac viral stock was added into the cells at 3–10 MOI of b.f.u. for 2 h. The b-galactosidase activity was measured 24 h later to document the transduction efficiency. For in vivo experiments, 200–500 g male rats (Fischer 344; Harlan, Indianapolis, IN, USA) were anesthetized with a combination of ketamine (100 mg/ml) and xylazine (20 mg/ml) (2:1, v/v). Each rat was stereotactically injected at two separate sites using the following coordinates: the middle of left caudate–putaman (AP, −0.5; ML 2.5; DV, 5.5) and the right hippocampus (AP, −4.0; ML 2.5; DV, 3.0). Each site was injected with 7 ml of viral solution containing 2 × 105 b.f.u. of p700-lac amplicon vectors or media. The right side of the brain served as a negative control. At 2, 7 and 14 days after injection, animals were reanesthetized and perfused with heparinized saline and 4% paraformaldehyde with 5 mm EGTA and 2 mm MgCl2. The brains were removed, post-fixed for 2 h in the same fixative, then transferred to 25% sucrose with 2 mm MgCl2 overnight. Sections (20 mm) were cut on a cryostat and reacted

with 0.1% X-gal overnight followed by neutral red counter staining.

Acknowledgements We are grateful for generous gifts of materials from Drs N DeLuca, PA Johnson, B Sugden, T Tsukada, H Hayakawa, H Takebe and N Stow. We thank Ginny Austin and Jesse Lamson for excellent assistance with manuscript preparation, and Michael Smith for technical support.

References 1 Anderson WF. Human gene therapy. Science 1992; 226: 401–403. 2 Miller AD. Retroviral vectors. Curr Top Microbiol Immunol 1992; 158: 1–24. 3 Mulligan RC. The basic science of human gene therapy. Science 1993; 260: 926–932. 4 Crystal RG. Transfer of genes to humans: early lessons and obstacles to success. Science 1995; 270: 404–410. 5 Wilson JM. Adenoviruses as gene delivery vehicles. New Engl J Med 1996; 334: 1185–1187. 6 Naldini L et al. In vivo gene delivery and stable transduction of nondividing cells by a Lentiviral vector. Science 1996; 272: 263–267. 7 Hill TJ. Herpes simplex virus latency. In: Roizman B (ed). The Herpesviruses. Plenum Press: New York, 1985, pp 175–240. 8 Glorioso JC, DeLuca NA, Fink DJ. Development and application of herpes simplex virus vectors for human gene therapy. Annu Rev Microbiol 1995; 49: 675–710. 9 Fields B, Knipe D, Howley P. Fields Virology: Herpes Simplex Viruses and Their Replication. Lippincott-Raven: Philadelphia, 1996, pp 2231–2296. 10 Honess RW, Roizman B. Regulation of herpes virus macromolecular synthesis. I. Cascade regulation of the synthesis of three groups of viral proteins. J Virol 1974; 14: 8–19. 11 DeLuca NA, Schaffer PA. Activation of immediate–early, early, and late promoters by temperature sensitive and wild-type forms of herpes simplex virus type 1 protein ICP4. Mol Cell Biol 1985; 5: 1997–2009. 12 DeLuca NA, McCarthy AM, Schaffer PA. Isolation and characterization of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate–early regulatory protein ICP4. J Virol 1985; 56: 558–570. 13 DeLuca NA, Schaffer PA. Activities of herpes simplex virus type 1 (HSV-1) ICP4 genes specifying nonsense peptides. Nucleic Acids Res 1987; 15: 4491–4511. 14 Johnson PA et al. Cytotoxicity of a replication-defective mutant of herpes simplex virus type 1. J Virol 1992; 66: 2952–2965. 15 Sabel BA, Vick A, Holt V. Neurotoxic reactions of CNS following gene transfer with defective herpes simplex virus (HSV-1) vector. NeuroReport 1995; 6: 2447–2449. 16 Spaete RR, Frenkel N. The herpes simplex virus amplicon: a new eucaryotic defective-virus cloning-amplifying vector. Cell 1982; 30: 295–304. 17 Kwong AD, Frenkel N. Herpes simplex virus amplicon: effect of size on replication of constructed defective genomes containing eukaryotic DNA sequences. J Virol 1984; 51: 595–603. 18 Geller AI, Breakefield XO. A defective HSV-1 vector expresses Escherichia coli b-galactosidase in cultured peripheral neurons. Science 1988; 241: 1667–1669. 19 Geller AI, Freese A. Infection of cultured central nervous system neurons with a defective herpes simplex virus 1 vector results in stable expression of Escherichia coli b-galactosidase. Proc Natl Acad Sci USA 1990; 87: 1149–1153. 20 Palella TD et al. Expression of human HPRT mRNA in brains of mice infected with a recombinant herpes simplex type 1 vector. Gene 1989; 80: 137–144. 21 Chiocca EA et al. Transfer and expression of the lacZ gene in rat


22

23

24

25

26

27

28

29

brain neurons by herpes simplex virus mutants. New Biol 1990; 2: 739–746. Anderson JK et al. Gene transfer into mammalian central nervous system using herpesvirus vectors: extended expression of bacterial lacZ gene in neurons using the neuron-specific enolase promoter. Hum Gene Ther 1992; 3: 487–499. Ho D, Mocarski E, Sapoloski R. Altering central nervous system physiology with a defective herpes simplex virus vector expressing the glucose transporter gene. Proc Natl Acad Sci USA 1993; 90: 3655–3659. During MJ et al. Long-term behavioral recovery in Parkinsonian rats by an HSV vector expressing tyrosine hydroxylase. Science 1994; 266: 1399–1403. Wang S, Vos J. A hybrid herpesvirus infectious vector based on Epstein–Barr virus and herpes simplex virus type 1 for gene transfer into human cells in vitro and in vivo. J Virol 1996; 70: 8422–8430. Yates JL, Warren N, Sugden B. Stable replication of plasmids derived from Epstein–Barr virus in various mammalian cells. Nature 1985; 313: 812–815. Reisman D, Yates J, Sugden B. A putative origin of replication of plasmids derived from Epstein–Barr virus is composed of two cis-acting components. Mol Cell Biol 1985; 5: 1822–1832. Rawlins DR et al. Sequence-specific interactions of cellular nuclear factor I and Epstein–Barr virus nuclear antigen with herpes virus DNAs. Cancer Cell 1986; 4: 525–542. Goldsmith K, Bendell L, Frappier L. Identification of EBNA1 amino acid sequences required for the interaction of the functional elements of the Epstein–Barr virus latent origin of DNA replication. J Virol 1993; 67: 3418–3426.

30 Judde JG et al. Use of Epstein–Barr virus nuclear antigen-1 in targeted therapy of EBV-associated neoplasia. Hum Gene Ther 1996; 7: 647–653. 31 Muldoon RR et al. Tracking and quantitation of retroviral mediated transfer using a humanized, red shifted green fluorescent protein gene. BioTechniques 1997; 22: 162–167. 32 Section 1: Escherichia coli, plasmids, and bacteriophages. In: Ausubel FM (ed). Current Protocols in Molecular Biology. John Wiley and Sons: New York, 1989, pp 1.4.2–1.4.3. 33 Fraefel C et al. Helper virus-free transfer of herpes simplex virus type 1 plasmid vectors into neural cells. J Virol 1996; 70: 7190– 7197. 34 Kochanek S et al. A new adenoviral vector: replacement of all viral coding sequences with 28 kb of DNA independently coexpressing both full-length dystrophin and b-galactosidase. Proc Natl Acad Sci USA 1996; 93: 5731–5736. 35 Starr P et al. Long-term persistence of defective HSV-1 vectors in the rat brain is demonstrated by reactivation of vector gene expression. Gene Therapy 1996; 3: 615–623. 36 Martuza RL et al. Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science 1991; 252: 854–856. 37 Boviatsis EJ et al. Long-term survival of rats harboring brain neoplasms treated with ganciclovir and a herpes simplex virus vector that retains an intact thymidine kinase gene. Cancer Res 1994; 54: 5745–5751. 38 Hirt B. Selective extraction of polyoma DNA from infected mouse cell cultures. J Mol Biol 1967; 26: 365–369.

1141