Gene Therapy
SEARCH     advanced search my account e-alerts subscribe register
Journal home
Advance online publication
Current issue
Press releases
For authors
For referees
Contact editorial office
About the journal
For librarians
Contact NPG
Customer services
Site features
NPG Subject areas
Access material from all our publications in your subject area:
Biotechnology Biotechnology
Cancer Cancer
Chemistry Chemistry
Dentistry Dentistry
Development Development
Drug Discovery Drug Discovery
Earth Sciences Earth Sciences
Evolution & Ecology Evolution & Ecology
Genetics Genetics
Immunology Immunology
Materials Materials Science
Medical Research Medical Research
Microbiology Microbiology
Molecular Cell Biology Molecular Cell Biology
Neuroscience Neuroscience
Pharmacology Pharmacology
Physics Physics
Browse all publications
June 2001, Volume 8, Number 11, Pages 846-854
Table of contents    Previous  Article  Next   [PDF]
Research Article
A novel system for the production of fully deleted adenovirus vectors that does not require helper adenovirus
N Cheshenko1, N Krougliak1, R C Eisensmith1,2 and V A Krougliak1

1Institute for Gene Therapy and Molecular Medicine, Mount Sinai School of Medicine, New York, NY, USA

2Department of Human Genetics, Mount Sinai School of Medicine, New York, NY, USA

Correspondence to: V A Krougliak, Institute for Gene Therapy and Molecular Medicine, Box 1496, Mount Sinai School of Medicine, One Gustave L Levy Place, New York, NY 10029, USA


Fully deleted adenovirus vectors (FD-AdVs) would appear to be promising tools for gene therapy. Since these vectors are deleted of all adenoviral genes, they require a helper adenovirus for their propagation. The contamination of the vector preparation by the helper limits the utility of currently existing FD-AdVs in gene therapy applications. We have developed an alternative system for the propagation of FD-AdVs, in which the adenoviral genes essential for replication and packaging of the vector are delivered into producer cells by a baculovirus-adenovirus hybrid. A hybrid baculovirus Bac-B4 was constructed to carry a Cre recombinase-excisable copy of the packaging-deficient adenovirus genome. Although the total size of the DNA insert in Bac-B4 was 38 kb, the genetic structure of this recombinant baculovirus was stable. Bac-B4 gave high yields in Sf9 insect cells, with titers of 5 ´ 108p.f.u./ml before concentration. Transfection of 293-Cre cells with lacZ-expressing FD-AdV plasmid DNA followed by infection by Bac-B4 at a MOI of 2000 p.f.u./ml resulted in rescue of the helper-free vector. Subsequent passaging of the obtained FD-AdV using Bac-B4 as a helper resulted in ~100-fold increases of the vector titer at each passage. This resulting vector was completely free of helper virus and was able to transduce cultured 293 cells. However, scaling-up of FD-AdV production was prevented by the eventual emergence of replication-competent adenovirus (RCA). Experiments are underway to optimize this system for the large-scale production of helper virus-free FD-AdVs and to minimize the possibility of generation of replication-competent adenovirus (RCA) during vector production. This baculovirus-based system will be a very useful alternative to current methods for the production of FD-AdVs. Gene Therapy (2001) 8, 846-854.


fully deleted adenovirus vector; baculovirus; gene therapy; baculovirus-adenovirus hybrid; production


Vectors derived from human adenoviruses exhibit a number of features that have made them extremely useful gene transfer reagents. Most notably, these vectors can transduce a broad range of both dividing and post-mitotic cells of different origin. In addition, first-generation adenovirus vectors are easy to manipulate and can be propagated to high titers. These latter properties permit the generation of sufficient amounts of vector for experimental, preclinical and clinical applications. Unfortunately, as is now well established, first-generation adenovirus vectors elicit strong cellular immune responses against the cells that they have transduced.1,2,3,4

Because this immunogenicity is largely a consequence of the low-level expression of virus genes that remain in the vector backbone, one approach to overcoming this immune response has been to engineer these vectors further to remove additional virus genes, especially other early regions such as E2 and E4.5,6,7,8,9,10,11,12,13,14 Although some studies have indicated that these 'second-generation' adenovirus vectors are less toxic and less immunogenic,9,11 there is no conclusive evidence that they are capable of significantly prolonged persistence. Moreover, in nearly every case, the introduction of additional deletions has significantly decreased the resulting titers, making the vectors more difficult to produce in clinically useful amounts.

The ultimate form of adenovirus vector modification is the creation of fully deleted adenovirus vectors (FD-AdVs), also called 'gutless', 'gutted', 'amplicon', 'high-capacity' or 'helper-dependent' adenovirus vectors. These are adenovirus-based vectors that contain only the cis elements necessary for replication and packaging, but lack all adenovirus genes. All of the FD-AdVs created thus far15,16,17,18,19,20,21,22,23,24,25,26 share a number of disadvantages. Foremost among these is the use of helper viruses or plasmid cotransfection to provide the necessary virus proteins in trans. The use of helper viruses almost always results in some contamination of the vector by the helper virus. The use of plasmids usually results in much lower titers of vector. Some additional disadvantages of this system include the need for highly time-consuming, multiple intermittent rounds of vector expansion in 293 and/or 293-Cre cell lines in order to make large-scale vector preparations and to limit contamination by the helper virus. Furthermore, helper-dependent FD-AdVs cannot be easily titered, since they do not form plaques.

In the present study, we have examined the potential utility of a novel system for the production of FD-AdVs that are free of contaminating helper virus. In this system, a packaging-deficient, replication-competent adenovirus helper genome is delivered to the FD-AdV-producing cells by a recombinant baculovirus/adenovirus (Bac/Ad) hybrid. This helper genome provides all of the trans functions necessary for the propagation of FD-AdVs. The absence of the adenovirus packaging signal from this helper genome completely prevents its packaging into infectious virus particles. The conception of this system was derived from three observations: (1) recombinant baculovirus based on the genome of Autographa californica multiple nuclear polyhedrosis virus (AcMNPV) can efficiently transduce certain mammalian cells in tissue culture;27,28,29 (2) when properly joined, adenovirus ITRs can be efficiently utilized as an adenoviral origin of replication;30,31 and (3) Cre recombinase can efficiently excise and circularize a DNA fragment flanked by two loxP sites.32,33 Previous studies demonstrated that AcMNPV is capable of transducing mammalian cells without virus gene expression or replication.34,35,36,37,38 Thus, this hybrid Bac/Ad helper virus is not likely to multiply during the course of FD-AdV production, so that only the genome of the FD-AdV will be packaged into the resulting Ad particles. Additionally, since the baculovirus capsids are enveloped in lipid membrane, they can be easily removed from the final vector stocks by chloroform extraction, which does not affect the infectivity of adenovirus virions. The complete elimination of a helper virus from the vector preparations may further improve the safety of these vectors, decrease the toxicity associated with their administration, and prolong the duration of transgene expression. In addition, by obviating the need for physical separation of the vector from the helper virus, the system described in this study should greatly facilitate large-scale production of FD-AdVs for preclinical and clinical studies.


Design and construction of the baculovirus/adenovirus hybrid helper

The principal design of the helper-free system is outlined in Figure 1. The baculovirus/adenovirus hybrid helper virus Bac-B4, containing an excisable packaging-deficient and replication-competent adenovirus genome, was constructed as follows. The circular adenoviral helper genome was derived from the plasmid pBHG10,39 which carried a functional copy of the adenovirus genome deleted in the packaging signal and the E1 and E3 regions, and which contained fused ITRs that could act as an origin of adenovirus replication. Adenovirus sequences interrupted in the E1 region were excised from this plasmid, and inserted in the recombinant baculovirus genome together with two loxP sites in such a way that loxP sites flanked all inserted sequences. Inclusion of two loxP sites allowed excision of the circular adenovirus helper genome from the baculovirus chromosome upon infection of Cre-expressing 293B5B1 cells. This excised, packaging-deficient and replication-competent genome should be able to replicate to a high copy number and to provide all of the helper functions required for production of FD-AdVs (Figure 1). Additionally, the vesicular stomatitis virus glycoprotein G (VSVG) envelope protein expression cassette driven by the AcMNPV polyhedrin promoter was inserted into the region adjacent to one of the loxP sites. The use of the baculovirus polyhedrin promoter maximized VSVG expression in insect cells while minimizing or preventing expression in mammalian cells. The VSVG gene was included to permit more efficient transduction of the mammalian cells by the pseudotyped baculovirus.40 This recombinant baculovirus genome, Bac-B4, was generated through site-specific recombination in E. coli, and rescued and propagated in insect cells Sf9.

Genetic stability of recombinant baculovirus Bac-B4

The resulting hybrid helper baculovirus Bac-B4 contained an insert of 38 kb of foreign DNA. In order to test whether this recombinant baculovirus was genetically stable, Southern analyses were performed on Bac-B4 DNA using 32P-labeled wild-type Ad5 DNA as a probe (Figure 2). The structure of the Ad insert in Bac-B4 did not change during passaging, suggesting that the presence of adenovirus sequences in the recombinant genome did not interfere with baculovirus growth. This finding, however, did not prove that the presence of adenovirus sequences was not able to induce rearrangements in the baculovirus sequences. Such rearrangements, if they occurred, would generate defective baculovirus genomes. To test for this possibility, we transformed competent E. coli DH10B cells with the recombinant baculovirus DNA isolated from two large-scale preparations of Bac-B4. The bacmid DNA from the resulting colonies was isolated and its infectivity was tested by transfection of Sf9 cells. Bacmids isolated from all tested clones were able to generate infectious baculovirus upon transfection of Sf9 cells. This result indicated that the baculovirus backbone of all of the analyzed bacmid clones was stable despite the insertion of ~38 kb of adenovirus DNA. In addition, Southern analysis of the baculovirus/adenovirus hybrid virus DNA confirmed that the structure of adenovirus sequences remained indistinguishable from those of the parental bacmid (data not shown). These data suggest that, despite the presence of the large insert of adenovirus sequences, the genetic structure of Bac-B4 remained stable in Sf9 cells.

Excision of the adenoviral sequences from Bac-B4

Adenovirus sequences in Bac-B4 genome were flanked by two loxP sites, which should allow the Cre-mediated excision of these sequences to generate a circular adenovirus helper genome similar to pBHG10. We tested whether the Cre recombinase-mediated recombination of these sites could occur in Cre-expressing cells following infection by Bac-B4. Cre recombinase-expressing 293B5B1 cells were infected by concentrated Bac-B4 at a MOI of 100. Forty-eight hours after infection, low-molecular weight DNA was extracted and subjected to PCR using primers specific to the excised Ad genome (Figure 3). The appearance of the 1735 bp PCR product, which confirmed that excision of the adenovirus genome fragment from the baculovirus backbone had occurred, was observed only in 293B5B1 cells infected by the indicated baculovirus/adenovirus hybrid. This product was not observed in 293 cells infected by the baculovirus/ adenovirus hybrid or in mock-infected 293B5B1 cells. These results indicated that in Cre-expressing cells the adenovirus helper genome could be efficiently excised from the Bac/Ad hybrid.

Design and construction of a FD-AdV

The plasmid pAVec9-2 was used for rescue of the FD-AdV AdAVec 9-2. pAVec9-2 (Figure4) contained two inverted copies of the Ad5 left ITR with the packaging signal. A stuffer sequence, originating from the human phenylalanine hydroxylase gene,41 was inserted between these ITRs. A lacZ reporter gene driven by the human EF1alpha promoter was placed inside the stuffer sequence. Two SwaI sites flanked the ITRs, allowing precise excision of the total FD-AdV genome from the plasmid backbone. Following construction, the functionality of this vector was tested using a helper-dependent system similar to that developed by Parks et al.19 The AdAVec9-2 rescued in 293B5B1 cells using an adenovirus helper virus with a Cre recombinase-excisable packaging signal was propagated to a titer of 5 ´ 109 b.f.u./ml (data not shown). Although all vector stocks also contained 5-10% of the helper virus, as estimated by plaque assay, this experiment proved that, in the presence of helper genome, the adenovirus ITRs and packaging signal in pAVec9-2 were fully functional, efficiently mediating vector replication and encapsidation.

Rescue of a FD-AdV using Bac-B4 as a helper

To test whether the baculovirus/adenovirus hybrid virus was able to permit rescue and propagation of helper virus-free FD-AdVs, pAVec9-2 DNA was digested with SwaI to liberate the adenovirus termini and transfected into 293B5B1 cells, which then were transduced by Bac-B4 at MOIs of 100 or 2000 p.f.u. of Bac-B4 per cell. The transduced cells were harvested 4 days after infection and crude lysates were prepared. These crude lysates were titered for the presence of the vector by infection of naïve 293 cells followed by X-gal staining. The same lysates were tested for the presence of the helper adenovirus by plaque assay in 293 cells. The obtained preps of helper-free vector were passaged in 293B5B1 cells in the presence of Bac-B4. The results of these experiments are shown in Table 1. When Bac-B4 was used at a MOI of 100 p.f.u. per cell, the rescue of an FD-AdV was not efficient. However, increasing the MOI of Bac-B4 to 2000 p.f.u. per cell greatly increased the efficiency of rescue of the AdAVec9-2 and allowed its propagation to relatively high titers (~108 b.f.u./ml), with an approximately 100-fold increase of vector titer with each passage. This multiplication rate was similar to that obtained in a helper-dependent system using an adenovirus helper, suggesting that the Bac-B4 was capable of complementing FD-AdV replication and packaging almost as efficiently as an adenovirus helper. Unfortunately, we were unable to evaluate the full potential of this system because further passaging of the AdAVec9-2 in the presence of 2000 p.f.u. per cell of Bac-B4 resulted in the appearance of RCA. Thorough testing confirmed that the original rescue and first three passages of FD-AdV were completely free of RCA, indicating that Bac-B4 sufficiently complemented the replication of FD-AdV. Furthermore, although the emerging RCA was able temporarily to boost the titer of AdAVec9-2 (Table 1), the RCA quickly overgrew the FD-AdV during further passaging driving it out of the stock. This study was repeated four times with quantitatively similar results (data not shown).

The AdAVec9-2 DNA obtained from these preparations, shown to be negative for RCA, was analyzed by Southern analysis (Figure 5). The restriction pattern of the rescued vector was almost identical to a restriction pattern of pAVec9-2. The only discernable difference was the absence of the 1.8 kb band in the vector DNA preparation. This fragment, containing the plasmid origin of replication and antibiotic-resistance gene corresponds to a SwaI fragment excised from the pAVec9-2 before the FD-AdV rescue and is not supposed to be present in the vector genome. Thus, this observation clearly indicates that the vector genome was stable during propagation when Bac-B4 was used as a helper.

Transduction in cell culture

The ability of the helper-free AdAVec9-2 preparation to transduce naïve cells was tested in tissue culture. 293 Cells were infected by 0.1 ml of crude lysates generated in the second passage of AdAVec9-2, which was propagated in 293B5B1 cells using Bac-B4 as a helper. Twenty-four hours after infection, cells were stained for beta-gal expression (Figure 6). The absence of helper adenovirus in the examined lysates was confirmed by analyzing the lysates by plaque assay in 293 cells. Since the sensitivity of plaque assay is sufficient to reveal the presence of helper virus at a concentration as low as 1-5 p.f.u./ml, the inability of the obtained vector stocks to generate plaques in 293 cells clearly indicates that these stocks are free of helper virus, or at least that the ratio between the vector and the helper is greater than 108:1.


The major limitation of the currently existing helper-dependent systems for the production of FD-AdVs is the contamination of the final vector preparation by contaminating helper virus. The presence of this helper adenovirus can potentially increase the immunogenicity of FD-AdVs, thereby jeopardizing their safety and persistence. Additionally, the presence of the helper complicates the production of these vectors and impairs the scalability of the system. Despite extensive efforts, complete elimination of helper virus from the vector preparations has yet to be achieved.

The helper-free system developed in this study offers a means of propagating FD-AdVs so that the final preparations can be free of helper virus. The major difference between this system and the helper-dependent system is that, instead of deleting the packaging signal from the helper genome during the vector propagation, we have deleted it before it gets into cells. This modification precludes the packaging of this genome into adenovirus virions.

In the design of this system we took advantage of a unique feature of the baculovirus AcMNPV to infect human cells efficiently while remaining virtually non-toxic and transcriptionally and replicatively inactive. The genome of AcMNPV consists of a 131 kb double-stranded circular DNA molecule.42 Although it was thought that the extra packaging capacity of baculovirus might exceed 20 kb, the actual upper packaging limit of baculovirus capsids has not yet been established. Some investigators have suggested that baculoviruses can merely extend their corkscrew-like capsids to accommodate extra DNA sequences. Indeed, the recombinant baculovirus Bac-B4 created in this study contained an insert of 37 705 bp, which apparently did not affect virus viability. Moreover, we were also able to generate several additional recombinant baculoviruses carrying inserts of 40 and 47 kb (data not shown). Thus, the upper packaging limit of the AcMNPV capsid still has not been reached, but is likely to exceed 50 kb. Our data also suggest that, despite the presence of such a large insert, the genetic structure of Bac-B4 remained stable, allowing passaging and expansion of the virus. In addition, recombinant baculoviruses are easy to propagate to relatively high titers (108-109 p.f.u./ml) and can be even further concentrated by centrifugation27,28 or by cation-exchange chromatography.43

Since baculovirus genes are not expressed in mammalian cells, it is unlikely that the Bac/Ad hybrid genome will replicate in mammalian cells. However, the replication of adenovirus genome is essential to generate the large numbers of templates for the synthesis of the large amounts of viral proteins that are necessary to obtain high yields of vector. Thus, in order to complement the replication and packaging of FD-AdV in an efficient fashion, the adenovirus helper genome should be able to replicate to a high copy number following transduction of vector-producing cells by the Bac/Ad hybrid.

To permit replication directed from the adenoviral ITRs, the adenovirus sequences in the Bac-B4 were flanked by two loxP sites that allowed excision of the circular packaging-deficient adenovirus genome from the baculovirus chromosome in Cre-expressing 293B5B1 cell line. This excised circular adenovirus genome should retain the fused adenovirus inverted terminal repeats that have been proved functional in promoting adenovirus DNA replication in permissive cells.39 We expected that such an excision would permit subsequent replication of the helper genome using the adenovirus replication machinery, leading to higher levels of adenovirus gene expression and thus to higher titers of the FD-AdV. While we were able to demonstrate that the specific excision of the adenovirus helper genome occurred in Cre-expressing cells, we do not have direct evidence that replication of the helper genome does in fact occur during the course of FD-AdV propagation. Our inability to rescue the FD-AdV in 293 cells that do not express Cre recombinase indirectly indicates that Cre-mediated excision of adenovirus genome from the hybrid genome is essential for efficient complementation of FD-AdV. However, the fact that only high doses of the Bac-B4 (2000 p.f.u. per cell) allowed efficient rescue of the FD-AdV may suggest that the replication is not occurring or is inefficient. Alternatively, this result may reflect the possibility that only small proportions of input baculovirus/adenovirus hybrid genomes were involved in complementing the FD-AdV. The question of whether the baculovirus-delivered helper genome is capable of replicating is under active investigation in our laboratory.

In our experiments, we were able to achieve approximately 100-fold increase of titers of the FD-AdV AVec9-2 during each propagation cycle. While this multiplication rate is not greater than that achieved using the standard helper-dependent vector production system, the complete absence of the helper virus in the vector stocks is a great advantage of our helper-free system over the helper-dependent system. Recent findings indicating that the adenovirus genome contains additional cis-acting elements involved in replication and packaging of the adenovirus genome26 suggest that the inclusion of these elements in the FD-AdV genome may further increase the production capacity of this baculovirus-based helper-free system.

Although we were able to demonstrate the potential utility of the baculovirus-based helper-free system for production of FD-AdV, the generation of RCA during the propagation greatly diminishes the utility of this system in its current configuration. The emergence of RCA was apparently the product of homologous recombination between the adenovirus sequences contained in 293 cells and the E1-deleted adenovirus helper genome. The adenovirus genome carried by Bac-B4 contained an E1 deletion corresponding to positions 190-1340 in wtAd5 genome. The remaining Ad sequences significantly overlapped with the adenovirus sequences harboring in 293 cells. Recombination in this region would result in rescue of both the E1 and the packaging signal into the adenovirus helper genome, rendering it packaging-competent. The rapid increase of the RCA titer from non-detectable level to 0.9 ´ 107 p.f.u./ml during one passage might be accounted for by the fact that the RCA, once emerged, is capable of undergoing multiple cycles of infection and multiplication. As a result, during 4 days of incubation a large proportion of cells was infected by RCA producing significant amounts of infectious helper-independent virus. Apparently, the emerging RCA has growth advantages as compared with the FD-AdV and rapidly overgrows the vector, also boosting the vector titers. It is obvious that this system needs to be further modified to prevent the generation of RCA. We are currently implementing several modifications of this system that theoretically should prevent RCA formation. We expect that re-design of the Bac/Ad hybrid and/or change of the cell line used for the FD-AdV production will obviate the generation of RCA. The successful completion of these studies should yield a readily scalable system for the production of FD-AdVs that are completely free of contaminating helper virus.

Materials and methods

Cell culture

Low-passage 293 cells were obtained from Microbix Biosystems, (Toronto, Ontario, Canada). The 293-derived cell line 293B5B1, which constitutively expresses Cre recombinase,18 was kindly provided by Dr Andre Lieber (University of Washington, Seattle, WA, USA). These cells were maintained in minimal essential medium (Gibco BRL, Life Technologies, Rockville, MD, USA) supplemented with 10% FBS, 100 units/ml penicillin, 100 mug/ml streptomycin and 0.25 mug/ml of amphotericin B. Spodoptera frugiperda (Sf9) cultured cells (ATCC No. CRL 1711) were grown in monolayers or in suspension in serum-free medium (Sf-900 II SFM), also containing 100 units/ml penicillin, 100 mug/ml streptomycin and 0.25 mug/ml of amphotericin B. All cell culture media were purchased from Gibco BRL.


Competent E. coli DH10Bac cells harboring both an episomally replicating baculovirus AcMNPV genome containing a Tn7 integration site (the so-called bacmid) and a helper plasmid expressing Tn7 transposase were obtained from Gibco BRL. The transfer vector backbone, pFastBac1, containing both the left and right arms of Tn7, was also obtained from Gibco BRL. The plasmid pBHG10, containing an E1- and E3-deleted packaging deficient adenovirus genome with two ITRs fused together, was purchased from Microbix Biosystems. The cloning vectors pLitmus29 was obtained from New England Biolabs (Beverly, MA, USA). Two loxP sites were generated by insertion of the synthetic oligonucleotides, GTACATAACTTCGTATAGCATACATTATACGAAGTT ATAATGA and CTAGTCATTATAACTTCGTATAAT GTATGCTATACGAAGTTAT, which when annealed formed loxP sites with XbaI and Acc65I sticky ends.

Plasmid construction

The transfer plasmid pBac-B4, which was used to generate the Bac/Ad hybrid virus, was constructed as follows. The baculovirus transfer vector, pFastBac1 (Gibco BRL), was modified by the insertion of two loxP sites between the SpeI and AvrII sites, resulting in pBac-B1. pBac-B1 was further modified by the introduction of ClaI and XbaI sites between the loxP sites yielding pBac-B2. The plasmid pBHG10,39 containing a circularized Ad5 genome with fused inverted terminal repeats (ITRs), was digested with ClaI and XbaI, which removed all of the bacterial sequences. The resulting 32.46 kb fragment was inserted between the ClaI and XbaI sites of pBAC-B2, generating pBac-B3. Finally, pBac-B3 was further modified by the insertion of a fragment containing the VSVG coding sequence and the SV40 polyadenylation signal into a unique BstBI site located immediately downstream of the polyhedrin promoter in pBac-B3. This produced the plasmid pBac-B4, which was then used to generate the corresponding recombinant baculovirus, Bac-B4.

To construct the FD-AdV-containing plasmid pAVec9-2, the left 356 bp terminal fragment of the Ad5 genome including the adenovirus packaging signal was PCR-amplified using two primers, VK-67 (CCTCTA GACCAGATTTAAATCATCAATAATATACCTTATTTT GG), which was specific for the very end of Ad5 ITR (sequences specific to Ad5 underlined) and contained a SwaI site (shown in bold) at the end of ITR, and VK-69 (CCATCGATCGGCCCTAGACAAATATTACGCGC), which annealed to the base pairs 356-333 in Ad5 genome. The PCR product was cloned in the vectors pZeRO-2 (Invitrogen, Carlsbad, CA, USA) and pNEB193 (New England Biolabs), resulting in pVNK1 and pVNK2, respectively. Following sequencing of the inserts, these two plasmids were used to generate pNC1.2, which contained two copies of the terminal fragment in opposite orientations with the beta-lactamase gene in the middle. Finally, the beta-lactamase gene was removed and replaced by stuffer sequences obtained from the introns of the human phenylalaninehydroxylase gene41 and an expression cassette in which the transcription of beta-galactosidase was driven by the EF1alpha promoter.44 In this final construct, pAVec9-2, two SwaI sites were located at both ends of the ITRs flanking the FD-AdV genome. Thus, SwaI digestion of pAVec9-2 generated a linear FD-AdV genome in which the ITR terminal sequences were only slightly altered, containing the sequences AAATCAT instead of the CATCAT that is found at the ends of the wild-type Ad5 genome. It had been previously demonstrated that slight alterations of the terminal sequences did not affect the functionality of ITRs in adenovirus replication.45,46

Construction of a recombinant baculovirus

The recombinant baculovirus Bac-B4 was generated using the Bac-to-Bac system (Gibco BRL) according to the manufacturer's manual. This system utilizes Tn7-mediated site-specific recombination between the bacmid and the transfer vector (pFastBac), modified to contain one or more transgene-expressing cassettes, to generate a recombinant bacmid in E. coli DH10Bac carrying the episomal copy of the baculovirus genome (bacmid). The plasmid pBac-B4 was used to transform competent DH10Bac cells, in which the transposition occurred. The resulting recombinant bacmid bBac-B4 was then isolated, and following the confirmation of the genetic structure was transfected into Sf9 cells using the Effectene transfection reagent (Qiagen, Valencia, CA, USA). At 96 h after transfection the recombinant infectious baculovirus Bac-B4 was recovered and plaque-purified as previously described.47 Following verification of the genetic structure, the subcloned Bac-B4 was expanded in Sf9 cells. Large-scale virus preparation was performed in spinner cultures according to the standard protocol supplied with the Bac-to-Bac system. Briefly, Sf9 cells were grown in suspension in SF-900 II medium (1 liter of medium in 3000 ml spinner flasks with a rotation speed of 100 r.p.m./min) to a density of 3 ´ 106 viable cells/ml. Cells were infected at an MOI of 3. Culture medium was harvested at 72 h and clarified by low-speed centrifugation. The resulting baculovirus titers varied from 2 ´ 108 to 1 ´ 109 plaque-forming units (p.f.u.)/ml. High-titer baculovirus stocks were generated by concentrating the virus from the culture media using ultrafiltration in Centricon-80 or Centricon-20 Plus centrifugal filter devices. The 25-fold to 100-fold concentration achieved in this step resulted in baculovirus titers as high as 4 ´ 1010 p.f.u./ml. The titers of the recombinant baculoviruses were determined using the BacPAK Baculovirus Rapid Titer Kit (Clontech Laboratories, Palo Alto, CA, USA) and were further confirmed by plaque assays performed according to the standard protocol supplied with the Bac-to-Bac system.

Detection of replication-competent adenoviruses

The appearance of replication-competent adenovirus (RCA) or helper virus in all FD-AdV preparations was monitored by plaque assay in 293 cells and by PCR. Plaque assay was performed as previously described.48 For PCR detection, the rightward primer 5'-GACGCCCGACATCACCTGTG-3', which anneals to sequence 1313-1338 in Ad5, and the leftward primer 5'-CGGCGAGCGCCTTCTGGCGG-3', which anneals to sequence 5072-5053 in Ad5, were designed in such a way as to allow detection of the adenovirus generated from the recombination between the helper genome and 293 cells.

Rescue and propagation of the FD-AdV

AdAVec9-2 was rescued in 293B5B1 cells as follows. pAVec9-2 DNA was digested with SwaI, purified using QIAquick Gel Extraction Kit (Qiagen) and transfected into 293B5B1 cells using a modified calcium-phosphate precipitation procedure. Briefly, 10 mug of this digested DNA was mixed with 62 mul of 2 M CaCl2, and sterile water in a final volume of 500 mul, and the precipitate was formed by addition of 500 mul of 2´ HBS (50 mM Hepes, pH 7.5; 10 mM KCl; 12 mM Dextrose; 280 mM NaCl; 1.5 mM Na2HPO4). The mixture was incubated for 2-3 min at room temperature and the resulting precipitate was added to a 60 mm dish of 80% confluent 293B5B1 cells, which contained 4 ml of the freshly changed medium with 25 muM chloroquine. Following 6 h of incubation at 5% CO2, 37°C, transfected cells were washed by PBS, the medium was replaced with MEM supplemented with 5% horse serum (Gibco BRL) and cells were returned to 5% CO2, 37°C incubator for further 18 h. Growth medium was removed and cells were infected by 0.5 ml of concentrated Bac-B4 with MOIs of 100 or 2000 for 1 h at room temperature followed by the addition of 5 ml of fresh MEM supplemented with 5% horse serum. Cells were returned to 5% CO2, 37°C incubator for an additional 96 h, then harvested, subjected to three rounds of freeze and thaw, and the AdAVec9-2 yields in the resulting crude extracts were assessed by titration.

Titration of the FD-AdV

The titers of the AdAVec9-2 were determined by X-gal staining of 293 cells infected by the vector preparations. Cells were grown in 35 mm dishes to a confluency of 80%. Growth medium was removed and cells were infected by 0.2 ml of preparations of the vector serially diluted in PBS. Following a 45 min incubation virus was removed and replaced by 3 ml of growth medium. At forty-eight hours after infection, cells were washed twice in PBS, fixed with 0.5% glutaraldehyde for 5 min at room temperature, washed again in PBS to remove the fixative and stained for 2-4 h at 37°C with 1 mg/ml X-gal (5-bromo-4-chloro-3-indolyl-beta-d-galactoside) in PBS buffer supplemented with 20 mM MgCl2, 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide.

Propagation of the FD-AdV

293B5B1 cells (80% confluent) were infected by 0.5 ml of a Bac-B4 preparation at MOIs of 100 or 2000 p.f.u. per cell for 1 h at room temperature. The virus suspension was removed and replaced by 0.5 ml of a titered AdAVec9-2-containing crude lysate. Following 45 min incubation at 5% CO2 and 37°C, crude lysate was removed, fresh growth medium was added, and cells were returned to the incubator for a further 96 h. Cells were then harvested and used to prepare crude lysates. The vector yields were evaluated by titration in 293 cells followed by X-gal staining.

Southern blot analysis

Baculoviral DNA was prepared from high-titer stocks of BAC-B4 virus using a DNA Blood Mini Kit (Qiagen, Chatsworth, CA, USA). Plasmid DNA from E. coli was prepared by using the Concert High Purity Plasmid Miniprep System (Gibco BRL). DNA samples were digested by HindIII restriction endonuclease, separated in 0.8% agarose gels, and transferred to SuPerCharge nylon membranes (Schleicher & Schuell, Keene, NH, USA) according to the standard protocol supplied by the manufacturer. Radioactive probes were generated by random prime labeling system Rediprime II (Amersham Pharmacia Biotech, Bucks, UK) of wild-type adenovirus Ad5 DNA. Hybridized filters were exposed to Kodak Biomax MS film (Kodak, Rochester, NY, USA).

Excision assay

In order to test the Cre-mediated excision of adenovirus genes from the Bac/Ad genome, 293B5B1 cells (1 ´ 106 cells per 60 mm dish) were infected with Bac-B4 at a MOI of 100. Forty-eight hours after infection, the total DNA was extracted from the cells using QiaAmp DNA Blood kit. Aliquots of DNA were analyzed by 35 cycles of PCR using the oligonucleotide primers CACAGCCTGGCG ACGCCCACC and GGACCATGTGGTCACGCTTTTCG. These primers were specific to the regions adjacent to the loxP sites in Bac-B4, and the presence of a 1735 bp PCR fragment was indicative of the Cre-mediated excision of Ad sequences from the hybrid genome.


This work was supported by NIH grants DK51700 to RCE and DK5333 to VAK.


1 Mullbacher A, Bellett AJ, Hla RT. The murine cellular immune response to adenovirus type 5. Immunol Cell Biol 1989; 67: 31-39, MEDLINE

2 Yang Y, Ertl HC, Wilson JM. MHC class I-restricted cytotoxic T lymphocytes to viral antigens destroy hepatocytes in mice infected with E1-deleted recombinant adenoviruses. Immunity 1994; 1: 433-442, MEDLINE

3 Yang Y et al. Cellular immunity to viral antigens limits E1-deleted adenoviruses for gene therapy. Proc Natl Acad Sci USA 1994; 91: 4407-4411, MEDLINE

4 Juillard V et al. Long-term humoral and cellular immunity induced by a single immunization with replication-defective adenovirus recombinant vector. Eur J Immunol 1995; 25: 3467-3473, MEDLINE

5 Yang Y et al. Inactivation of E2a in recombinant adenoviruses improves the prospect for gene therapy in cystic fibrosis. Nat Genet 1994; 7: 362-369, MEDLINE

6 Engelhardt JF, Ye X, Doranz B, Wilson JM. Ablation of E2A in recombinant adenoviruses improves transgene persistence and decreases inflammatory response in mouse liver. Proc Natl Acad Sci USA 1994; 91: 6196-6200, MEDLINE

7 Armentano D et al. Characterization of an adenovirus gene transfer vector containing an E4 deletion. Hum Gene Ther 1995; 6: 1343-1353, MEDLINE

8 Krougliak V, Graham FL. Development of cell lines capable of complementing E1, E4, and protein IX defective adenovirus type 5 mutants. Hum Gene Ther 1995; 6: 1575-1586, MEDLINE

9 Wang Q, Jia XC, Finer MH. A packaging cell line for propagation of recombinant adenovirus vectors containing two lethal gene-region deletions. Gene Therapy 1995; 2: 775-783, MEDLINE

10 Fang B et al. Lack of persistence of E1- recombinant adenoviral vectors containing a temperature-sensitive E2A mutation in immunocompetent mice and hemophilia B dogs. Gene Therapy 1996; 3: 217-222, MEDLINE

11 Gorziglia MI et al. Elimination of both E1 and E2 from adenovirus vectors further improves prospects for in vivo human gene therapy. J Virol 1996; 70: 4173-4178, MEDLINE

12 Wang Q et al. Persistent transgene expression in mouse liver following in vivo gene transfer with a delta E1/delta E4 adenovirus vector. Gene Therapy 1997; 4: 393-400, MEDLINE

13 Amalfitano A et al. Production and characterization of improved adenovirus vectors with the E1, E2b, and E3 genes deleted. J Virol 1998; 72: 926-933, MEDLINE

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

15 Mitani K, Graham FL, Caskey CT, Kochanek S. Rescue, propagation, and partial purification of a helper virus- dependent adenovirus vector. Proc Natl Acad Sci USA 1995; 92: 3854-3858, MEDLINE

16 Fisher KJ et al. Recombinant adenovirus deleted of all viral genes for gene therapy of cystic fibrosis. Virology 1996; 217: 11-22, Article MEDLINE

17 Kochanek S et al. A new adenoviral vector: Replacement of all viral coding sequences with 28 kb of DNA independently expressing both full-length dystrophin and beta-galactosidase. Proc Natl Acad Sci USA 1996; 93: 5731-5736, Article MEDLINE

18 Lieber A, He CY, Kirillova I, Kay MA. Recombinant adenoviruses with large deletions generated by Cre- mediatedexcision exhibit different biological properties compared withfirst-generation vectors in vitro and in vivo. J Virol 1996; 70: 8944-8960, MEDLINE

19 Parks RJ et al. A helper-dependent adenovirus vector system: removal of helper virus by Cre-mediated excision of the viral packaging signal. Proc Natl Acad Sci USA 1996; 93: 13565-13570, Article MEDLINE

20 Parks RJ, Graham FL. A helper-dependent system for adenovirus vector production helps define a lower limit for efficient DNA packaging. J Virol 1997; 71: 3293-3298, MEDLINE

21 Schiedner G et al. Genomic DNA transfer with a high-capacity adenovirus vector results in improved in vivo gene expression and decreased toxicity published erratum appears in Nat Genet 1998 Mar;18(3): 298). Nat Genet 1998; 18: 180-183, MEDLINE

22 Morsy MA et al. An adenoviral vector deleted for all viral coding sequences results in enhanced safety and extended expression of a leptin transgene. Proc Natl Acad Sci USA 1998; 95: 7866-7871, Article MEDLINE

23 Parks RJ et al. Effects of stuffer DNA on transgene expression from helper-dependent adenovirus vectors. J Virol 1999; 73: 8027-8034, MEDLINE

24 Morral N et al. Administration of helper-dependent adenoviral vectors and sequential delivery of different vector serotype for long-term liver-directed gene transfer in baboons. Proc Natl Acad Sci USA 1999; 96: 12816-12821, Article MEDLINE

25 Morsy MA, Caskey CT. Expanded-capacity adenoviral vectors - the helper-dependent vectors. Mol Med Today 1999; 5: 18-24, Article MEDLINE

26 Sandig V et al. Optimization of the helper-dependent adenovirus system for production and potency in vivo. Proc Natl Acad Sci USA 2000; 97: 1002-1007, Article MEDLINE

27 Hofmann C et al. Efficient gene transfer into human hepatocytes by baculovirus vectors. Proc Natl Acad Sci USA 1995; 92: 10099-10103, MEDLINE

28 Boyce FM, Bucher NL. Baculovirus-mediated gene transfer into mammalian cells. Proc Natl Acad Sci USA 1996; 93: 2348-2352, Article MEDLINE

29 Sandig V et al. Gene transfer into hepatocytes and human liver tissue by baculovirus vectors. Hum Gene Ther 1996; 7: 1937-1945, MEDLINE

30 Graham FL. Covalently closed circles of human adenovirus DNA are infectious. EMBO J 1984; 3: 2917-2922, MEDLINE

31 McGrory WJ, Bautista DS, Graham FL. A simple technique for the rescue of early region I mutations into infectious human adenovirus type 5. Virology 1988; 163: 614-617, MEDLINE

32 Sauer B, Henderson N. The cyclization of linear DNA in Escherichia coli by site-specific recombination. Gene 1988; 70: 331-341, MEDLINE

33 Sternberg N. Bacteriophage P1 cloning system for the isolation, amplification, and recovery of DNA fragments as large as 100 kilobase pairs. Proc Natl Acad Sci USA 1990; 87: 103-107, MEDLINE

34 Groner A, Granados RR, Burand JP. Interaction of Autographa californica nuclear polyhedrosis virus with two nonpermissive cell lines. Intervirology 1984; 21: 203-209, MEDLINE

35 Carbonell LF, Miller LK. Baculovirus interaction with nontarget organisms: a virus-borne reporter gene is not expressed in two mammalian cell lines. Appl Environ Microbiol 1987; 53: 1412-1417, MEDLINE

36 Hartig PC, Chapman MA, Hatch GG, Kawanishi CY. Insect virus: assays for toxic effects and transformation potential in mammalian cells. Appl Environ Microbiol 1989; 55: 1916-1920, MEDLINE

37 Hartig PC, Cardon MC, Kawanishi CY. Insect virus: assays for viral replication and persistence in mammalian cells. J Virol Meth 1991; 31: 335-344,

38 Hartig PC, Cardon MC, Kawanishi CY. Effect of baculovirus on selected vertebrate cells. Dev Biol Stand 1992; 76: 313-317, MEDLINE

39 Bett AJ, Haddara W, Prevec L, Graham FL. An efficient and flexible system for construction of adenovirus vectors with insertions or deletions in early regions 1 and 3. Proc Natl Acad Sci USA 1994; 91: 8802-8806, MEDLINE

40 Barsoum J, Brown R, McKee M, Boyce FM. Efficient transduction of mammalian cells by a recombinant baculovirus having the vesicular stomatitis virus G glycoprotein. Hum Gene Ther 1997; 8: 2011-2018, MEDLINE

41 DiLella AG et al. Molecular structure and polymorphic map of the human phenylalanine hydroxylase gene. Biochemistry 1986; 25: 743-749, MEDLINE

42 Ayres MD et al. The complete DNA sequence of Autographa californica nuclear polyhedrosis virus. Virology 1994; 202: 586-605, Article MEDLINE

43 Barsoum J. Concentration of recombinant baculovirus by cation-exchange chromatography. Biotechniques 1999; 26: 834-840, MEDLINE

44 Guo ZS, Wang LH, Eisensmith RC, Woo SL. Evaluation of promoter strength for hepatic gene expression in vivo following adenovirus-mediated gene transfer. Gene Therapy 1996; 3: 802-810, MEDLINE

45 Berkner KL, Sharp PA. Generation of adenovirus by transfection of plasmids. Nucleic Acids Res 1983; 11: 6003-6020, MEDLINE

46 Lippe R, Graham FL. Adenoviruses with nonidentical terminal sequences are viable. J Virol 1989; 63: 5133-5141, MEDLINE

47 O'Reilly DR. Use of baculovirus expression vectors. Meth Mol Biol 1997; 62: 235-246,

48 Graham FL, Prevec L. Methods for construction of adenovirus vectors. Mol Biotechnol 1995; 3: 207-220, MEDLINE


Figure 1 The principal design of the helper-free system. The baculovirus/adenovirus hybrid containing a packaging-deficient adenovirus helper genome is rescued in insect cells as a recombinant baculovirus. Upon infection of Cre-expressing 293 cells, the circular adenovirus genome is excised from the hybrid genome. This excision generates two circular molecules, one containing an adenovirus genome and the other containing a baculovirus genome. While the baculovirus genome would remain transcriptionally and replicatively inactive in mammalian cells, the adenovirus genome would be capable of replicating and providing all of the helper functions that are necessary for the propagation of FD-AdVs. Since the adenovirus genome released from the Bac/Ad hybrid lacks the adenovirus packaging signal, only the FD-AdV is packaged in the adenovirus particles, resulting in helper-free vector preparation.

Figure 2 Genetic stability of baculovirus Bac-B4, which carries a 38 kb insert. Bac-B4 DNA, isolated from different baculovirus preparations that underwent 2, 3 or 4 passages, was rescued in E. coli DH10B cells. The virus and bacmid DNA was digested with HindIII and assayed by Southern analysis using 32P-labeled DNA of the transfer vector pBac-B4 as a probe.

Figure 3 Excision of the Ad genome from Bac-B4 in Cre-expressing 293B5B1 cells. 293B5B1 cells were transfected with pBac-B4 DNA or infected by Bac-B4 at a MOI of 100. The low molecular weight DNA was extracted 48 h later and analyzed by PCR using primers specific to the excision product. Lane 1: DNA from 293B5B1 transfected with pBac-B4; lane 2: DNA from 293B5B1 infected by Bac-B4; lane 3: DNA from the mock-infected 293B5B1 cells; lane 4: DNA from Bac-B4-infected 293 cells. The position of the expected 1735 bp PCR product is indicated by the arrow.

Figure 4 The structure of the plasmid pAVec9-2 (a) and of the FD-AdV, AdAVec9-2 (b) generated from this plasmid. The plasmid pAVec9-2 was constructed so that it contained two copies of the adenovirus type 5 left inverted terminal repeat with its adjacent packaging signal These two ITRs were positioned at the ends of the vector genome in opposite orientations to form functional adenovirus origins of replication. The Ad5 coding region was replaced by stuffer DNA and a cassette in which the expression of lacZ was driven by the human EF1alpha promoter. ITR: adenovirus inverted terminal repeat; psi : adenovirus packaging signal. To rescue the FD-AdV AdAVec9-2, the sequence between two SwaI sites was first liberated from pAVec9-2 by SwaI digestion. The excised DNA fragment was transfected into 293B5B1 cells followed by infection by the Bac-B4, which resulted in the replication and packaging of AdAVec9-2.

Figure 5 Southern analysis of the DNA of AdAVec9-2 produced using a helper-free or helper-dependent system. HindIII + SwaI-digested DNA of the RCA-negative preparation of the HF AdAVec9-2 (lane 1) and HD AdAVec9-2 (lane 2). HindIII + SwaI-digested DNA of and pAVec9-2 (lane 3) was used as a positive control. DNAs were separated in 1% agarose gel, transferred to the Hybond NX nylon membrane (Amersham Pharmacia, Little Chalfont, UK) and hybridized with 32P-labeled DNA of pAVec9-2. The position of the 1.8 kb SwaI fragment of pAVec9-2 containing the prokaryotic origin of replication and an antibiotic-resistance gene is indicated by the arrow.

Figure 6 Histochemical staining for beta-galactosidase after infection of 293 Cells by the helper virus-free and RCA-free AdAVec9-2. 293 cells were infected by diluted crude lysates of HF AdAVec9-2 obtained from passages 2 and 3 and 48 h after infection stained by X-gal. (a) FD-AdV from passage 2 diluted 1:100. (b) AdAVec9-2 from passage 3 diluted 1:10 000.


Table 1 Production of helper virus-free FD-AdV using the hybrid virus BAC-B4

Received 1 August 2000; accepted 6 February 2001
June 2001, Volume 8, Number 11, Pages 846-854
Table of contents    Previous  Article  Next    [PDF]