Establishment of an oriP/EBNA1-based episomal vector transcribing human genomic β-globin in cultured murine fibroblasts

Article metrics


A novel oriP/EBNA1-based episomal vector has been constructed that persists episomally in cultured murine fibroblasts. The vector, pBH148, is equipped with the entire 185-kb human β-globin gene locus. After amplification in bacteria, column-purified episomal pBH148 was transfected into both cultured EBNA1-expressing human D98/Raji positive control fusion cells (DRpBH148) and cultured EBNA1-negative murine fibroblast cells (A9pBH148). Cell cultures were maintained concurrently with and without hygromycin selection for a period of 3 months. We show long-term stable episome maintenance of the full-size 200-kb circular double-stranded pBH148 in both the DRpBH148 cultures and the A9pBH148 cultures, regardless of selective pressure by agarose gel electrophoresis and Southern blot. EBNA1 transgene was detected by PCR in all transfected cultures. In addition, we were able to detect correctly spliced human β-globin mRNA by RT-PCR in all transfected late-passage DRpBH148 and A9pBH148 cell cultures. These findings illustrate that this oriP/EBNA1-based episomal vector is stable in a previously nonpermissive murine cell line and is a potential vector for human gene therapy.


There are a variety of non-replicating and self-replicating viral-based vector delivery systems under consideration for long-term gene replacement therapy. Non-replicating vectors require chromosomal integration, while self-replicating vectors are maintained extrachromosomally. Most of the vectors used currently in gene therapy clinical trials are non-replicating types generated from the essential components of retrovirus, adenovirus, or adeno-associated virus.1 Although these vectors show great promise, they have limitations.2345

Self-replicating viral-based vectors are capable of long-term episomal persistence in mammalian cells.67 These vectors contain components derived from (1) Epstein–Barr virus (EBV; oriP/EBNA1 viral elements);8910 (2) bovine papilloma virus (BPV-1; ori elements);1112 (3) genomic components from either yeast (YACs) or human chromosomes (HACs);13141516 and (4) chimeric viral/genomic systems either engineered from the oriP/EBNA1-based pLIB or from mammalian artificial episomal chromosomes (MAECs).171819 The advantages and limitations of these various viral vector-based systems in biological research are discussed elsewhere.7 However, episomal vectors are being considered for use in human gene therapy.2021222324 Of these, the EBV-based vectors may have several distinct advantages.

EBV is a human gammaherpesvirus with a 172-kb double-stranded DNA genome that preferentially infects B-lymphocytes and some epithelial cells, although it has been detected in other tissues.7 Establishment of EBV requires just two viral encoded components, the latent origin of replication (oriP) and its trans-activating protein (EBNA1).2526 About 90% of the world's adult population is estimated to be infected with non-symptomatic EBV infection. Engineered EBV-based oriP/EBNA1 vectors persist in the nucleus of target cells as multi-copy episomes. They are capable of inserting a gene of interest that cannot be interrupted or subjected to regulatory constraints arising from chromosomal integration. They also have a low mutation rate, ie less than 105, and tend not to rearrange.2728 Use of EBV-based vectors have been shown to result in a higher transfection efficiency than the use of chromosome-integrating plasmids.2930 Finally, EBV-based vectors can be designed to shuttle between eukaryotic and prokaryotic cells allowing for the transfer of very large genomic fragments of DNA.931 Historically, EBV infection appeared to be limited to only primate cells having failed to replicate in hamster and mouse cells.1832 However, shortly thereafter, small EBV shuttle vectors were reported to replicate in rat PC12 cells and mouse fibroblastoid cell line.1933 More recently, EBV-based vectors of >30-kb have been reported to replicate in mice.3435 Although these reports are encouraging, these EBV-based vectors were not tested at delivering large (>100 kb) intact human gene loci.

As discussed in this paper, we developed a novel ‘second-generation’ non-infectious and non-transforming EBV-based vector capable of shuttling between prokaryotic, human and mouse cells. This vector carries the entire 185-kb human β-globin gene loci and transcribes correctly spliced human β-globin transcript. We were able to consistently isolate episomes from all transfected cell cultures, regardless of selective pressure for 3 months (the duration of the experiment). We calculated a very high episome retention rate of 93% in transfected EBNA1-negative murine cells, regardless of selective pressure as well. These results illustrate the potential of an EBV-based vector to deliver and maintain the large and intact human gene loci necessary for successful human gene therapy.


Stable transfection of human DR and mouse A9 cells

The episomal vector (pBH148, Figure 1) used in this study was initially generated as previously described and is equipped with the entire 185-kb human β-globin gene locus.10 In addition, the plasmid carries the two essential EBV elements oriP and EBNA1 which are required for latent episome replication and replication initiation, respectively.

Figure 1

Map of pBH148. pBH148 was generated by combining the essential components of EBV (oriP and EBNA1) required for episomal replication and maintenance in mammalian cells with specific BAC elements for propagation in bacteria. In addition, selectable markers for hygromycin, chloramphenicol and thymidine kinase have been incorporated. pBH148 is equipped with the entire 185-kb human β-globin gene locus complete with its well-defined origin.

To assess the transfection efficiency of pBH148 in the EBNA1-negative murine A9 cells (A9pBH148), we also co-transfected EBNA1-expressing human DR cells (DRpBH148). This EBV-permissive positive cell line was chosen because it has been used previously by our laboratory to establish stable clones carrying large pBR322-based vectors (up to 350-kb) of anonymous human genomic inserts.710 Approximately 40 A9pBH148 clones/106 transfected cells and 50 DRpBH148 clones/106 transfected cells were obtained yielding similar transfection efficiencies. All experiments described below were performed on the pooled clones. The clones were maintained under hygromycin selection for 1 month before experimentation. After 1 month under selection, 6 × 108 cells were lysed and the episomal pBH148 DNA was isolated. The episome integrity was assessed by agarose gel electrophoresis and subjected to PCR using hygromycin specific primers (Figure 2). For this experiment, we used bacteria propagated pBH148 as a positive control. Lanes 2, 3 and 4 were all positive for the hygromycin resistance gene on pBH148. Lanes 6 and 7 (background negative control lanes) were included to rule out any nonspecific primer amplification in either A9 or DR genomic DNA. These results demonstrated that the hygromycin sequence on pBH148 was only amplified from the transfected cell lines as expected.

Figure 2

Hygromycin detection by PCR in 1-month-old cell cultures under selection. Specific hygromycin primers were used as described in the text. The 431-bp hygromycin product is evident in the transfected 1-month-old A9pBH148 culture (lane 2), the transfected DRpBH148 culture (lane 3), and in the transformed DH5α bacteria bacteria culture (lane 4). No hygromycin was detected in either the untransfected A9 cells (lane 6) or the untransfected DR cells (lane 7) as expected.

Long-term stability of pBH148 in A9 cells

To test the stability of pBH148 episomes in the murine A9 cells over time, the A9 cell transformants were split into two separate cultures, one maintained with hygromycin selection and one without. These cultures were cultivated concurrently for a period of 3 months (length of the experiment). For comparison, the stably transfected DRpBH148 cell culture was maintained under selection for the same period of time and analyzed in parallel.

Pure unadulterated stable circular double-stranded episomal pBH148 (totally free of genomic DNA) was guaranteed by the episomal preparation. To ensure episome purity, we loaded the DNA on a 0.8% agarose 35-cm gel (Figure 3). The 200-kb circular double-stranded pBH148 episome migrated much slower than any other DNA fragments (<25-kb), including any linear fragments of pBH148 accidentally generated by mechanical shearing during the preparation. Although the episomal preparation developed by our laboratory consistently isolates large circular double-stranded plasmids, we have reported shearing and linear fragments before.102636 However, on three separate occasions, our laboratory demonstrated by FISH analysis that the large EBV-based episomal vectors did not integrate.101936 In support, this study shows conclusively that circular double-stranded pBH148 episomes were extracted from the late-passage A9pBH148 cells in a similar manner as those isolated from the late-passage DRpBH148 positive control culture by agarose gel electrophoresis and Southern blot hybridization Figure 3. The blot was probed with an AlfIII-digested pBH140 vector that was used to generate the 200-kb pBH148.10 All the A9pBH148 episomes isolated from any culture to week 12 (lanes 5–14) migrated at approximately the same position as the episome from the positive control DRpBH148 (lane 1). A DNA ladder, from the ethidium bromide-stained gel, is superimposed on the Southern blot to illustrate the distances between the intact 200-kb pBH148 circular episome, any sheared linear pBH148, and genomic fragments. Surprisingly, the intensity of the episome signal is very consistent between weekly episomal preparations, regardless of selective pressure, and intensity is still strong at week 12, even without selection. These results are indicative of stable pBH148 episomal maintenance. In addition, the genomic region on the blot (~25-kb) did not hybridize at all. This supports our assertions that (1) a carefully performed episomal preparation results in mostly large circular episomes with very little DNA shearing; and (2) the very presence of 200-kb circular DNA and the lack of any signal elsewhere in the blot guarantees that pBH148 did not integrate into the host chromosomes.

Figure 3

Episomal analysis of the oriP/EBNA1-based vector in EBNA1-negative A9 cell transformants. The episomes isolated from the A9pBH148 cultures with (+) and without (−) selection were subjected to agarose gel electrophoresis, blotted on to nylon, and probed as described in text. The ~200-kb pBH148 episome migrated much slower than either the genomic DNA fragments or the linear pBH148 fragments generated by mechanical shearing during the episomal preparation. Circular double-stranded pBH148 was isolated from all the A9pBH148, regardless of selection over the 12 weeks of the experiment (lanes 5–14), as it was from the DRpBH148 positive control culture (lane 1). Signal is seen in the wells of the blot because not all episomal DNA enters the gel because of RNA and protein contamination. A DNA ladder was superimposed on the blot for comparison between 200-kb episomes and genomic/pBH148 linear fragments (lane 4).

pBH148 long-term persistence over a period of time

In vitro functional and therapeutic studies with transfecting DNA are usually performed using cell cultures maintained under marker selection. Therefore, A9pBH148 cultures were grown continuously for 3 months with and without hygromycin selection in the medium and passaged twice a week at 1/10 dilution. This culture regimen corresponds to approximately 15 cell doublings every 2 weeks, for a total of 90 doublings. Episomes extracted from these late-passage cultures were analyzed by Southern blot hybridization as mentioned above. Although the persistence of pBH148 episomes was evident Figure 3, we were interested in evaluating the retention rate of the pBH148 episome in the EBNA1-negative A9 cells with and without hygromycin selection as a function of time. Therefore, episomal DNA was isolated at the indicated time points (Figure 4) and qualified by agarose gel electrophoresis followed by Southern blot using the AlfIII digested pBH140 as described.19 The values from the blot was then assessed by phosphoimaging Figure 4. Strikingly, pBH148 is very stable in the EBNA1-negative A9 cells even without selection. The plot highlights an episome retention rate greater than 93%. The plot also indirectly confirmed that pBH148 was maintained as an episome in the A9 cells and has not integrated because some signal loss occurred. Specifically, the episome signal decreased from the actively dividing A9 cells with and approximate half-life of 15 days. Given a 24-h doubling time for A9 cells, this rate corresponds to a 93% efficiency of episome retention per cell division.1931

Figure 4

Long-term persistence of the 200-kb oriP/EBNA1-based vector grown in EBNA1-negative murine A9 fibroblasts with () and without (▪) hygromycin selection. Episomes extracted from these late-passage cultures were analyzed to determine the episome retention rate of the pBH148 in the EBNA1-negative A9 cells with and without hygromycin selection as a function of time. Episomal DNA was isolated at the indicated time-points, qualified by agarose gel electrophoresis, and Southern blotted as described in the text. The values from the blot were then assessed by phosphoimaging. Strikingly, pBH148 is very stable in the EBNA1-negative A9 cells even without selection and the plot highlights an episome retention rate greater than 93%.

We know that the viral encoded trans-activator protein, EBNA1, is absolutely required for replication of an oriP-based plasmid in human cells and may be required for episomal maintenance in nonpermissive cells (unpublished results). DR fusion cells endogenously express EBNA1 and maintain an oriP-based plasmid in the absence of selection. It was important to prove that the A9 cells did not express any endogenous EBNA1 protein. Indeed, EBNA1 was detected by PCR in all the untransfected DR cell preparations, regardless of selection (Figure 5). More importantly, EBNA1 was not detected in the untransfected A9 cell preparations and was detected in all the A9pBH148 transfected cell cultures, regardless of selection.

Figure 5

EBNA1 detection by PCR in late-passage cell cultures. Specific EBNA1 primers were used as described in text. The 227-bp amplified gene product, signifying the presence of the EBNA1 transgene, is present in all transfected A9pBH148 cultures either with (+) or without (−) selection (lanes 2–11), the tranfected DRpBH148 control (lane 12), and in the untransfected EBNA1-positive DR cells (lane 15). Significantly, there is no detection of EBNA1 in the untransfected A9 cells (lane 14) as expected.

EBNA1 binds as a homodimer to two locations within the oriP: (1) the family of repeats (FR) is necessary for nuclear retention and episome segregation; and (2) the dyad of symmetry (DS) is needed for replication initiation. Interestingly, previous studies demonstrated that oriP does not function in rodent cells.1825 In contrast, it was shown that a 170-kb full-size EBV genome can stably replicate as a circular double-stranded episome in murine cells.19 Therefore, EBV may have evolved an alternative origin of replication, distinct from oriP. In support, one such site was recently mapped.37 However, pBH148 has only two EBV-derived genes (oriP and EBNA1) and does not have this presumed origin. We believe that replication initiation of pBH148 may be occurring at the origin site recently mapped within the inserted human β-globin locus.38 Admittedly, this is difficult to prove under the conditions of this study. However, even though A9 cells are incapable of maintaining an oriP/EBNA1-based episome without selection, our study suggests (1) the stability of a large 200-kb EBV/BAC-based episomal vector in mouse A9 cells; and (2) the potential utility of the human β-globin locus origin in maintaining this type of episome in nonpermissive EBNA1-negative A9 cells.

Persistent human β-globin transgene expression in nonpermissive A9 cells

RT-PCR using specific primers flanking intron 2 was applied to test potential transgene expression from the human β-globin gene cluster on the pBH148 vector in the stably transfected murine A9 cells (Figure 6). The untransfected A9 negative control cells did not transcribe any human β-globin gene product (lane 15). However, RT-PCR on the total RNA prepared from the A9pBH148 cells generated the unique 231-bp β-globin gene product regardless of selective pressure (lanes 2–12) and is identical in size to that obtained from the permissive DRpBH148 positive control total RNA prep (lane 13). In addition, no other product was detectable (data not shown) including the approximately 1-kb fragment that would be generated from unspliced DNA contamination (data not shown). Thus, all primary transcripts made on the human β-globin genomic template were correctly spliced, at least across this intron 2.39 This observation verified sustained human β-globin transcription from a stably maintained extrachromosomal episome during prolonged in vitro cell culture in a murine A9 cell line without selective pressure. Hence, this study shows similar persistent transcriptional activity of the human β-globin gene locus in murine A9 cells as was previously demonstrated in permissive human DR cells.10

Figure 6

Human β-globin transgene transcription in EBNA1-negative mouse A9 fibroblasts (weeks 1, 2, 4, 8 and 12) and in EBNA1-expressing DR cells (week 12) by RT-PCR. Specific primers sequences flanking intron 2 in the human β-globin gene were synthesized based upon published sequences as described in the text. The correctly spliced transcript of 231-bp is evident in all transfected A9pBH148 cultures with (+) or without (−) selection (lanes 2–11) as it was in the transfected DRpBH148 cultures with (+) and without (−) selection (lanes 12–13). No primary transcript was detected in either the negative control (lane 14) or the A9 cells alone (lane 15).


The present study demonstrates the relevance of an oriP/EBNA1-based episomal vector to deliver and transcribe human therapeutic genes from a stably maintained episome in EBNA1-negative murine cells. The episomal vector, pBH148, used in this research project was generated by combining the bare essential components of the latently episomal human herpes virus type 4 Epstein–Barr virus (EBV), some specific BAC elements, and an entire 185-kb human β-globin gene locus, including the locus control regions. Propagation in bacteria did not result in any unwanted methylations and stable transfection occurred in both late-passage cultured human EBNA1-expressing DR fusion cells and EBNA1-negative murine A9 fibroblast cells.

Stable episomal persistence of pBH148

The criteria used to demonstrate episomal maintenance in this study is based on episomal preparation of stably transfected cultured eukaryotic cells and the relative migration rate of the recovered circular double-stranded DNA as analyzed by agarose gel electrophoresis. This method was employed previously for the rapid and accurate identification of plasmids up to 350-kb.36 Indeed, we were able to show conclusively that the 200-kb circular double-stranded pBH148 episome was isolated from both the permissive DR human cells and the nonpermissive A9 murine cells, regardless of selective pressure for the length of the experiment. However, this is not the first time human DNA has been maintained extrachromosomally in rodent cells. YACs, with up to 1500-kb of human DNA, replicated as episomes in mouse cells following yeast spheroplast fusion, but were unstable in the absence of selective pressure.13144041 Although these YACs contained large human DNA fragments (complete with replication origins) they were devoid of any segregation apparatus. In other experiments, use of microcell fusion demonstrated that as little as 100-kb of human DNA could be maintained episomally in EBNA1-expressing mouse cells in low copy number, presumably by EBNA1 protein binding to the oriP on these vectors.41942 In addition, oriP-based plasmids with up to 30-kb of random DNA insert were maintained extrachromosomally in EBNA1-expressing hamster cells, but the episomes were rapidly lost from these rodent cells when selection was removed.1843 We believe that these experiments demonstrate the necessity of both endogenous expressed EBNA1 and selective pressure for an oriP/EBNA1-based plasmid to be maintained episomally in rodents. Ideally, oriP is bound by EBNA1 at the family of repeats (FR) and the dyad of symmetry (DS). EBNA1 binding to FR is absolutely required for episome segregation, while EBNA1 binding to DS initiates episome replication. We postulate that the FR segregation function of oriP is ubiquitous in nature, while the DS initiation of replication function of oriP may be species-specific or is inactive in the mouse. We base this assumption on the extensive research conducted in our laboratory demonstrating that oriP-based plasmids devoid of DS are lost more quickly than oriP-based plasmids devoid of FR in rodent cells (unpublished results). Therefore, we feel a ubiquitous mammalian origin may be required if oriP-based plasmids are to be used in rodent disease models. Consequently, pBH148 was engineered with an entire human β-globin locus, including a well-studied and fully mapped mammalian origin.44 We believe our results suggest that the human β-globin origin may be complementing an inactive oriP. Admittedly, we could strengthen this argument by testing a pBH148 episome devoid of a β-globin gene origin, but (due to our laboratory closure) we were unable to continue this work. However, in this study large circular double-stranded episomes were clearly isolated from EBNA1-negative nonpermissive murine A9 cells, regardless of selective pressure. Furthermore, since pBH148 constitutively expresses EBNA1, it is not restricted to cells expressing a pre-established EBNA1 gene as needed in previous studies.454647

Human β-globin transgene expression

The procedure used to verify large circular double-stranded episomes from the pBH148 transfected cell cultures as described above will not necessarily detect small sequence alterations or minor rearrangements within the vector. It has been shown that rearrangements of any kind in the β-globin gene locus, especially in the DNase hypersensitive sites within the LCR, usually silences gene expression.41 In fact, recent approaches to overcome gene silencing include either lentivirus-encoded ‘shaved-down’ β-globin gene locus transcription units or circumvention of the β-globin gene locus entirely by erythropoietin delivery in myoblasts, by electrotransfer, and by adeno-associated virus gene transfer.4849505152 Although these approaches have had limited success, there remain unresolved safety issues using lentivirus and adeno-associated virus in human applications. In addition, there are concerns about encapsulated myoblast technology and longevity, as well as the cost and efficacy of repeated electrotransfers of erythropoietin in diseased humans. All previous cases of de novo activation of globin gene expression in vivo or in vitro involved an erythroid genetic background.53 Intriguingly, transcription of the non-integrated β-globin gene occurred in non-erythroid human cells. The authors proposed that transfection of large genomic fragments as self-replicating naked DNA in somatic cells may allow basal transcription before nucleosome repression.10 Once a transcriptionally active chromatin has been established, autonomous replication may preserve its stable inheritance in dividing cells.5455 Clearly, the generality of this phenomenon must be tested, particularly in vivo on animal disease models.

In the present study, we demonstrate stable transgene transcription of the human β-globin gene by RT-PCR in both late-passage permissive DR and nonpermissive A9 cell cultures. Furthermore, no human β-globin transgene transcription of any type was detected in untransfected cells, indicating that all primary transcripts originated from the pBH148 episomal vector.

Potential of double-stranded circular oriP-based vectors

With the exception of phenotypic rescue of the beige mutation by circular episomal YACs in mouse cells, stable disease complementation by vectorology in human cells has yet to be accomplished.40 This and two previous studies demonstrated that large virus-based circular therapeutic episomes might soon become a reality.819 In addition, initial research using oriP/EBNA1-based ‘mini-EBV’ vectors led to the generation of an enhanced EBNA1 variant capable of improved episomal maintenance and transgene expression by a new improved second generation mini-EBV-based vector.91026 These vectors have been used for in vitro phenotypic correction of B cell lines by transduction of cDNAs coding for X-linked Lesch–Nyhan syndrome and the severe hereditary Fanconi's anemia complementation group C.3136 In other studies, EBV-based vectors have been used to express cytokines, growth factors, and human collagen V.555657

Use of oriP/EBNA1-based episomal vectors can potentially revolutionize basic chromosome research ranging from function and gene expression to transgene production of therapeutics to somatic/germline gene therapy. This study adds to the growing list of specialized gene transfer and expression systems derived from EBV-based plasmids. EBV-based episomal vectors allow for extrachromosomal maintenance of full-size human genes eliminating genomic position effects and reducing the potential long-term pathological consequences due to irreversible genetic modifications of the genome as seen when integrative vectors are used.7 In addition, episomes can be used for gene therapy that rely on ectopic genomic activation on episomes as an alternative strategy for syndromes recalcitrant to stable phenotypic correction by cDNA expression. We believe the research presented here illustrates a suitable gene transfer system that transfects mammalian cells ubiquitously, displays stable episomal replication, and fully expresses a transgene over extended periods of time.

Methods and materials

Preparation of the oriP/EBNA1-based pBH148

The episomal vector used in this study Figure 1was generated as previously described.10 Briefly, a 185-kb human β-globin genomic clone was obtained from a human genomic library and ligated to the alkaline phosphatase-treated oriP/EBNA1-based plasmid, pBH140 (not shown), engineered in our laboratory. The only difference between pBH140 and pBH148 is the presence of the 185-kb human gene insert.

Circular double-stranded episomal pBH148 was purified from a (shaken) 2-liter overnight culture of transformed Escherichia coli (DH5α) cells using Qiagen columns (Chatsworth, CA, USA). Eluted episomal DNA was quantified by fluorescent spectroscopy (DynaQuant, Amersham Sciences, NJ, USA) and adjusted to 10 μg/μl in sterile Tris-HCl (pH 8.0). Episomal integrity was assessed by agarose gel electrophoresis.

Cell transfections

For transfections, column purified episomal pBH148 was diluted as recommended (Promega Transfection Guide, WI, USA). Human DR positive control cells and mouse A9 cells (ATCC) were plated in a treated 12-well culture plate (Corning, Costar, MA, USA). Dulbecco's modified Eagle's medium (DMEM) (GIBCO) supplemented with 10% fetal calf serum (FCS) was added and the cells incubated in a humidified 5% CO2 chamber at 37°C until they were approximately 60% confluent (2 days). Lipofection was carried out according to the manufacturer's protocol (GIBCO, BRL, NY, USA) with a few modifications. Briefly, concentrated episomal pBH148 was added to a 1.5 ml tube containing 150 μl DMEM (with neither serum nor antibiotics) and 20 μl Lipofectamine Plus reagent to obtain DNA μg concentrations of 0.0, 0.1, 0.25, 0.5 and 1.0. The mixture was incubated for 15 min at room temperature. 8 μl of Lipofectamine reagent was added and the tube incubated as before. Cell cultures were rinsed with sterile phosphate buffered saline (PBS) before transfection. After transfection, 0.75 ml of DMEM supplemented with 10% FCS was added and the plate incubated as before for 5 h. Following incubation, 1 ml of DMEM (supplemented with 10% FCS, 1 × penicillin/streptomycin, and L-glutamine; Gibco) was added and the cells incubated for 2 days as before. Medium was changed on day 3 and every 2 days thereafter for 1 week with complete DMEM supplemented with 200 μg/ml hygromycin (DR) and 700 μg/ml (A9) for selection of pBH148. After 1 week, the surviving DR and A9 cells (from the 1.0 μg and 0.5 μg wells) were trypsinized, plated separately into 25-cm2 flasks (Costar), and grown to confluency under selection. Cells were split as needed into 75-cm2 flasks (Costar) under selection for 1 additional month. After verification of pBH148 DNA by PCR, the A9pBH148 culture was split into two 225-cm2 flasks (Costar). The cultures were grown to confluency and split as needed. In this experiment, less than 10% of the confluent culture seeded the new culture guaranteeing that episomal pBH148 correctly segregated to daughter cells during replication. The cultures were grown with and without hygromycin selection for a period of 3 months changing the medium as needed (about every 3rd day). The DRpBH148 was split similarly to the A9pBH148 culture, but it was maintained under hygromycin selection only. In addition, untransfected DR cells and untransfected A9 cells, under hygromycin selection, expired within 9 days, while the same cells without selection survived for the length of the experiment.

Episomal preparation of transfected pBH148 in A9 and DR cells

Episomal gels

Circular double-stranded episomal pBH148 DNA was isolated from 6 × 108 stably transformed cells as previously described with some modifications.7 Briefly, cells were lysed in a solution of 50 mM NaCl, 8 mM EDTA, and 1% sodium dodecyl sulfate (SDS) (pH 12.42) by vigorous vortexing for 30 s. The lysate rocked very gently end-to-end on a platform at 22°C for approximately 1 h. DNA was precipitated as described in the original technique. Approximately 25 μg of episomally prepared DNA was loaded on a 15 cm long 0.8% agarose gel in 1 × TAE (Tris Acetate EDTA) buffer and the gel ran at 85 V or 0.12 mV per cm2 for 12 h in a cold recirculating buffer chamber. DNA was visualized after ethidium bromide staining (0.65 mg per ml) for 2 h.


A hygromycin fragment of 431-bp was detected in the 1-month-old DRpBH148 and A9pBH148 cell cultures, maintained under selection, by the polymerase chain reaction (PCR). PCR conditions were for 30 cycles: 93°C, 65°C, and 72°C for 30 s each without either an initial hot start or a final extension, using forward 5′-CGTCTGCTGCTCCATACAAGCC-3′ and reverse 5′-TGCCTGAAACCGAACTGCCC-3′ primers specific for the hygromycin selection gene on pBH148. Each reaction was in a volume of 25 μl with 10 μg of episomally prepared DNA following standard protocol (Promega).

An EBNA1 fragment of 227-bp was detected in all the late-passage DRpBH148, A9pBH148, and in the DR background positive control. However, EBNA1 was not detected in the A9 cell cultures by PCR verifying their EBNA1-negative condition. PCR was carried out using the conditions described above using forward 5′-GGAAGCTTGGAAAGCATCGTGGTC-3′ and reverse 5′-ATGGATCCAAAGGGGAGACGACTCA-3′ EBNA1 specific primers.

Southern blot and episomal persistence

In preparation for alkaline transfer, the ethidium bromide-stained episomal gel was immersed in 0.25 M HCl for 30 min to fractionate the large circular double-stranded pBH148 episomal DNA. The gel was rinsed briefly in sterile water and placed in a 0.5 M NaOH/1.5 M NaCl alkaline solution for 2 h. DNA was transferred to a prewetted Magnagraph nylon membrane (MSI, MI, USA) by overnight capillary action in new alkaline solution. The membrane was rinsed briefly in sterile water for 30 s and DNA was UV crosslinked to the nylon using the automatic crosslink cycle on a UV Stratalinker 1800 (Stratagene, La Jolla, CA, USA) at 1200 μjoules × 100. During the transfer, an AflIII digested pBH140 probe was labeled with 50 μCIF α32P duct according to Prime-A-Gene kit protocol (Promega). The membrane was prehybridized in 10 ml preheated UltraHyb (Ambion, TX, USA) for 30 min at 42°C. Labeled probe was well mixed into the prehybridization solution and incubation continued overnight at 42°C. After the hybridization solution was removed, the blot was washed twice at room temperature for 5 min each in 2 × SSC/0.1% SDS then twice for 15 min each in 0.1 × SSC/0.1% SDS. The rinsed blot was wrapped in cellophane, placed in a cassette equipped with an intensifying screen, and exposed to X-OMAT AR film (Eastman Kodak, NY, USA) at −80°C for 60 min. The film was processed in a Konica SRX 201A automatic developer (Tokyo, Japan) and photographed.


Transgene expression of the human β-globin gene was analyzed by reverse transcription-based polymerase chain reaction (RT-PCR). Total cellular RNA was isolated following the manufacturer's protocol (RNAqueous, Ambion) from all stably transfected DRpBH148 cells, A9pBH148 cells, DR cells and A9 cells. The RNA was resuspended in 250 μg/μl aliquots and stored at −80°C until needed. RT-PCR was carried out using Superscript One-Step with platinum Taq (Gibco) following this thermo cycling profile: 50°C for 30 min (cDNA synthesis), 94°C for 2 min (reverse transcriptase denaturation), and 40 cycles of: 94°C, 65°C, 72°C for 30 s (PCR). Forward 5′-GGACCCAGAGGTTCTTTGAGTCC-3′ and reverse 5′-GCACACAGACCAGCACGTTGCCC-3′ primers specific for the human β-globin gene region located in exons 2 and 3 were used. A product of 231-bp and 1-kb is generated from correctly spliced human β-globin RNA transcript and unspliced DNA, respectively.

In memoriam: Jean-Michel Vos

We are saddened by the death of Jean-Michel Vos on 31 November 2000, who succumbed to cancer at the early age of 44 years. Jean-Michel came to the University of North Carolina in 1987, where he was an active member of the Department of Biochemistry and Biophysics, the Lineberger Comprehensive Cancer Center, the Program in Molecular Biology and Biotechnology, and the Curriculum in Genetics. He was internationally recognized as an expert in the field of artificial chromosome technology and gene therapy in which he was a pioneer. Jean-Michel will not only be missed as a scientist, but also as a fun-loving and good human being. He was committed to his family – his wife Eliane and his two daughters Natasha and Olivia. He left us too soon and I am honored to have known him. He will be missed by all of us in the Vos laboratory.


  1. 1

    Anderson WF . Human gene therapy Nature 1998 392: 25–30

  2. 2

    Naviaux RK, Verma IM . Retroviral vectors for persistent expression in vivo Curr Opin Biotechol 1992 3: 540–547

  3. 3

    Drittanti L et al. Optimised helper virus-free production of high-quality adeno-associated virus vectors J Gene Med 2001 3: 59–71

  4. 4

    Vos J-M . The simplicity of complex MACs Nature Biotechnol 1997 15: 1257–1259

  5. 5

    Hartigan-O'Connor D, Amalfitano A, Chamberlain JS . Improved production of gutted adenovirus in cells expressing adenovirus preterminal protein and DNA polmerase J Virol 1999 73: 7835–7841

  6. 6

    Vos J-M . Mammalian artificial chromosomes as tools for gene therapy Curr Opin Genet Develop 1998 8: 351–359

  7. 7

    Vos J-M . Therapeutic mammalian artificial episomal chromosomes Curr Opin Mol Ther 1999 1: 204–205

  8. 8

    Simpson K, McGuigan A, Huxley C . Stable episomal maintenance of yeast artificial chromosomes in human cells Mol Cell Biol 1996 16: 5117–5126

  9. 9

    Sun T-Q, Fenstermacher D, Vos J-M . Human artificial episomal chromosomes for cloning large DNA in human cells Nature Genet 1994 8: 33–41

  10. 10

    Westphal EM et al. A system for shuttling 200 kb BAC/PAC clones into human cells: stable extrachromosomal persistence and long-term ectopic gene activation Hum Gene Ther 1998 9: 1863–1873

  11. 11

    Baichwal VR, Sugden B . Vectors for gene transfer derived from animal DNA viruses: transient and stale expression of transfected genes Kucherlapati R (eds); Gene Transfer Plenum Press 1986 pp 148–177

  12. 12

    Ohe Y, Saijo N, Podack ER . Construction of a novel bovine papillomavirus vector without detectable transforming activity suitable for gene transfer Hum Gene Ther 1995 6: 325–333

  13. 13

    Featherstone T, Huxley C . Extrachromosomal maintenance and amplification of yeast artificial chromosome DNA in mouse cells Genomics 1993 17: 267–278

  14. 14

    Nonet GH, Wahl GM . Introduction of YACs containing a putative mammalian replication origin into mammalian cells can generate structures that replicate autonomously Somat Cell Mol Genet 1993 19: 171–192

  15. 15

    Harrington JJ et al. Formation of de novo centromeres and construction of first-generation human artificial microchromosomes Nature Genet 1997 15: 345–355

  16. 16

    Ikeno M et al. Construction of YAC-based mammalian artificial chromosomes Nature Biotechnol 1998 16: 431–439

  17. 17

    Krysan PJ, Haase SB, Calos MP . Isolation of human sequences that replicate autonomously in human cells Mol Cell Biol 1989 9: 1026–1033

  18. 18

    Wohlgemuth JG et al. Long-term gene expression from autonomously replicating vectors in mammalian cells Gene Therapy 1996 3: 503–512

  19. 19

    Kelleher ZT et al. Epstein-Barr-based episomal chromosomes shuttle 100 kb of self-replicating circular human DNA in mouse cells Nature Biotechnol 1998 16: 762–768

  20. 20

    Calos MP . The potential of extrachromosomal replicating vectors for gene therapy Trends Genet 1996 12: 463–466

  21. 21

    Cooper M et al. Safety-modified episomal vectors for human gene therapy Proc Natl Acad Sci USA 1997 94: 6450–6455

  22. 22

    Verma IM, Somia N . Gene therapy: promises. Problems Prospects Nature 1997 389: 239–242

  23. 23

    Lui D, Saltzman WM . Synthetic DNA delivery systems Nature Biotechnol 2000 18: 33–37

  24. 24

    Mountain A . Gene therapy: the first decade Trends Biotechnol 2000 18: 119–128

  25. 25

    Yates JL, Warren N, Reisman D, Sugden BA . A cis-acting element from the Epstein-Barr viral genome that permits stable replication of recombinant plasmids in latently infected cells Proc Natl Acad Sci USA 1984 81: 3806–3810

  26. 26

    Wendelburg BJ, Vos J-M . An enhanced EBNA1 variant with reduced GLY-ALA domain for long-term episomal maintenance and transgene expression of oriP-based plasmids in human cells Gene Therapy 1998 5: 1389–1399

  27. 27

    DuBridge RB et al. Analysis of mutation in human cells by using an Epstein-Barr virus shuttle system Mol Cell Biol 1987 7: 379–387

  28. 28

    DuBridge RB, Calos MP . Recombinant shuttle vectors for the study of mutation in mammalian cells Mutagenesis 1988 3: 1–9

  29. 29

    Belt PBGM et al. Efficient cDNA cloning by direct phenotypic correction of a mutant human cell line (HPRT2) using an Epstein-Barr virus-derived cDNA expression vector Nucleic Acids Res 1991 19: 4861–4866

  30. 30

    Mazda O, Satoh E, Yasutomi K, Imanishi J . Extremely efficient gene transfection into lympho-hematopoietic cell lines by Epstein-Barr virus-based vectors J Immunol Meth 1997 204: 143–151

  31. 31

    Banerjee S, Livanos E . Vos J.-M. Therapeutic gene delivery in human B-lymphoblastoid cells by engineered non-transforming infectious Epstein-Barr virus Nature Med 1995 1: 1303–1308

  32. 32

    Yates JL, Warren N, Sugden B . Stable replication of plasmids derived from Epstein-Barr virus in various mammalian cells Nature 1985 313: 812–815

  33. 33

    Lindenboim L, Anderson D, Stein R . The use of Epstein-Barr virus-based shuttle vectors in rat PC12 cells Cell Mol Neurobiol 1997 17: 119–127

  34. 34

    Tsukamoto H et al. Enhanced expression of recombinant dystrophin following intramuscular injection of Epstein-Barr virus (EBV)-based mini-chromosome vectors in mdx mice Gene Therapy 1999 6: 1331–1335

  35. 35

    Stoll SM, Sclimenti CR, Baba EJ . Epstein-Barr virus/human vector provides high-level, long-term expression of α1-antitrypsin in mice Mol Therapy 2001 4: 122–129

  36. 36

    Sun TQ, Livanos E, Vos J-M . Engineering a mini-herpesvirus as a general strategy to transduce up to 180 kb of functional self-replicating human mini-chromosomes Gene Therapy 1996 3: 1081–1088

  37. 37

    Little RD, Schildkraut CL . Initiation of latent DNA replication in the Epstein-Barr virus genome can occur at sites other than the genetically defined origin Mol Cell Biol 1995 15: 2893–2903

  38. 38

    Aladjem MI et al. Participation of the human beta-globin locus control region in initiation of DNA replication Science 1995 270: 815–819

  39. 39

    Sierakowska H, Sambade MJ, Agrawal S, Kole R . Repair of thalassemic human beta-globin mRNA in mammalian cells by antisense oligonucleotides Proc Natl Acad Sci USA 1996 93: 12840–12844

  40. 40

    Perou CM, Justice MJ, Pryor RJ, Kaplan J . Complementation of the beige mutation in cultured cells by episomally replicating murine yeast artificial chromosomes Proc Natl Acad Sci USA 1996 93: 5905–5909

  41. 41

    Ellis J, Pannell D . The beta-globin locus control region versus gene therapy vectors: a struggle for expression Clin Genet 2001 59: 17–24

  42. 42

    Yates JL . Epstein-Barr virus DNA replication DePamphilis ML (eds); DNA Replication in Eukaryotic Cells Cold Spring Harbor Press 1996 pp 751–773

  43. 43

    Krysan PJ, Calos MP . Epstein-Barr virus-based vectors that replicate in rodent cells Gene 1993 136: 137–143

  44. 44

    DePamphilis ML . Origins of DNA replication DePamphilis ML (eds); DNA Replication in Eukaryotic Cells Cold Spring Harbor Press 1996 pp 45–87

  45. 45

    Lutfalla G, Armbruster L, Dequin S, Bertolotti R . Construction of an EBNA-producing line of well-differentiated human hepatoma cells and of appropriate Epstein-Barr virus-based shuttle vectors Gene 1989 76: 27–39

  46. 46

    Langle-Rouault F et al. Up to 100-fold increase of apparent gene expression in the presence of Epstein-Barr virus oriP sequences and EBNA1: implications of the nuclear import of plasmids J Virol 1998 72: 6181–6185

  47. 47

    Horlick RA et al. Combinatorial gene expression using multiple episomal vectors Gene 2000 243: 187–194

  48. 48

    May C et al. Therapeutic haemoglobin synthesis in beta-thalassaemic mice expressing lentivirus-encoded human beta-globin Nature 2000 406: 82–86

  49. 49

    Dalle B et al. Improvement of mouse beta-thalassemia upon erythropoietin delivery by encapsulated myoblasts Gene Therapy 1999 6: 157–161

  50. 50

    Payen E et al. Improvement of mouse beta-thalassemia by electrotransfer of erythropoietin cDNA Exp Hematol 2001 29: 295–300

  51. 51

    Bohl D et al. Improvement of erythropoiesis in beta-thalassemic mice by continuous erythropoietin delivery from muscle Blood 2000 95: 2793–2798

  52. 52

    Peterson KR . Production and analysis of transgenic mice containing yeast artificial chromosomes Genetic Engineer 1997 19: 235–255

  53. 53

    Dillon N, Grosveld F . Chromatin domains as potential units of eukaryotic gene function Curr Opin Genet Dev 1994 4: 260–264

  54. 54

    Wolffe AP . Chromatin structure and DNA replication: implications for transcriptional activity DePamphilis ML (eds); DNA Replication in Eukaryotic Cells Cold Spring Harbor Press 1996 pp 271–293

  55. 55

    Mucke S et al. Suitability of Epstein-Barr virus-based episomal vectors for expression of cytokine genes in human lymphoma cells Gene Therapy 1997 4: 82–92

  56. 56

    Garret M, McHendry RB, Spickofsky N, Margolskee RF . Isolation of a clone which induces expression of the gene encoding the human tumor necrosis factor receptor Gene 1992 111: 215–222

  57. 57

    Fichard A et al. Human recombinant alpha 1 (V) collagen chain. Homo-trimeric assembly and subsequent processing J Biol Chem 1997 272: 30083–30087

Download references

Author information

Correspondence to J Black.

Rights and permissions

Reprints and Permissions

About this article


  • gene therapy
  • β-globin
  • OriP
  • EBNA1
  • episomal maintenance

Further reading