Abstract
Background/Aim: Efficient production of adenovirus vectors is crucial for their clinical use. Adenovirus death protein (ADP), which is encoded in the E3 region of the adenovirus genome, is involved in host-cell lysis and the subsequent release of progeny virus; however, the ADP gene is often removed from the adenovirus vector genome. Materials and Methods: We have developed adenovirus vectors that possess the ADP gene and maintain a relatively large insertion capacity for foreign genes by deleting the partial E3 region. Adenovirus vector-mediated transgene expression levels and virus titers were examined. Results: The adenovirus vectors maintaining the ADP gene showed cytopathic effect earlier than conventional adenovirus vector without the ADP gene following treatment of HEK293 cells, although there were no significant differences in total virus titers. Conclusion: The adenovirus vectors possessing the ADP gene showed efficient spread of progeny virus infection following transduction in HEK293 cells.
Recombinant adenovirus vectors (AVs) are widely used not only as gene delivery vehicles but also as vaccine vectors due to their several advantages, including high transduction efficiencies, high titer production, ability for efficient induction of both innate and adaptive immune responses, and relatively large capacity for insertion of transgene expression cassettes (1–4). More than 80 serotypes have been identified so far (5). Human adenovirus serotype 5 (Ad5), which belongs to species C, is commonly used as a framework for AVs.
The E3 region is commonly deleted from the AV genome in order to enlarge the capacity for insertion of foreign genes because the E3 genes are dispensable for the self-replication of adenoviruses. The E3 region encodes several genes, including the adenovirus death protein (ADP) gene (bp 29485-29766) (6, 7). The ADP gene is expressed mostly in the late phase of infection and is involved in the host cytopathic effect (CPE), host cell lysis, and the subsequent release of progeny virus (8–12). ADP is expressed at low levels from the E3 promoter early in infection, and at high levels from the major late promoter at the late stage of infection. While the deletion of the ADP gene from the adenovirus genome has no effect on viral replication, overexpression of this gene results in accelerated CPE, host cell lysis, and plaque formation. There is thus a correlation between ADP expression levels and cell killing efficiencies following infection with adenovirus (6). Insertion of the Ad5-derived ADP gene into the E3 region of adenovirus serotype 41 (Ad41), which does not carry ADP, has been shown to result in a 10- to 50-fold increase of virus titers (13). Several studies reported that an adenovirus containing the ADP gene was promising as an oncolytic virus (8, 11, 12, 14–17), but no previous studies evaluated whether the ADP gene provides any benefits for replication-incompetent AVs.
One of the major issues with AVs is the appearance of replication-competent adenovirus (RCA) during their production (18). Not only can RCA cause inflammatory responses, but it can also function as a helper virus to amplify an AV (19). Our group previously developed an AV system that can suppress the appearance of RCA by inserting foreign genes into regions other than the E1 region, taking advantage of the fact that the adenovirus genome, which is more than 105% of the length (approximately 37,700 bp) of the wild-type adenovirus genome, cannot be encapsulated in virus particles (20). HEK293 cells, a packaging cell line commonly used to propagate AVs, possess the adenovirus E1 gene (bp 1-4344) in the genome (21). When a transgene expression cassette is inserted into the deleted E1 region of an AV genome, homologous recombination can occur between the AV genome and the E1 gene in HEK293 cells, resulting in the emergence of RCA. When a transgene expression cassette is inserted into regions other than the E1, even if homologous recombination occurs, the viral genome exceeds the adenovirus packaging size limit and thus cannot be incorporated into the virus particle. Nonetheless, to our knowledge, no study has evaluated which region in the AV genome is suitable for the insertion of a transgene expression cassette.
In this study, we developed AVs possessing the ADP gene but lacking part of the E3 region for efficient production of AVs. We inserted the transgene expression cassette into the E1 region, the E3 region, or the region downstream of the fiber gene in order to optimize the insertion site of a transgene expression cassette. This study provides important information for the design of AVs with efficient transgene expression.
Materials and Methods
Cell lines. HEK293 cells (a human embryonic kidney cell line) were cultured with Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, and antibiotics. H1299 cells (a human non-small cell lung carcinoma cell line) were cultured with RPMI-1640 supplemented with 10% FBS and antibiotics. Panc-1 cells (a human pancreatic cancer cell line) and SK HEP-1 cells (a human hepatic adenocarcinoma cell line) were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% FBS and antibiotics. Cells were obtained from the American Type Culture Collection (Manassas, VA, USA) or official cell bank.
Plasmid and replication-incompetent AVs. AV plasmids possessing the ADP gene with partial deletion of the E3 region, pAdHM309-E1, -E3, and -F, were constructed using pAdHM41-E3+ (20), based on the sequence of the E3 region of the dl309 mutant adenovirus (22). PI-SceI/I-CeuI-digested pAdHM309-E1, -E3, and -F were ligated with PI-SceI/I-CeuI-digested pHM5-RBG-CMVL1 and pHM5-RBG-CMVL2, which contained the cytomegalovirus promoter-driven firefly luciferase expression cassette. An AV plasmid for a conventional AV containing the intact E3 region, pAdHM1-L2, was constructed by ligation of PI-SceI/I-CeuI-digested pAdHM1 (23) and PI-SceI/I-CeuI-digested pCMVL1 (24). Details of the plasmid-construction methods are available upon request.
The AVs were constructed by means of an improved in vitro ligation method described previously (23, 25). Each AV plasmid was digested with PacI to release the recombinant viral genome and was transfected into HEK293 cells plated on 60-mm dishes. All AVs were propagated in HEK293 cells, purified by two rounds of cesium chloride-gradient ultracentrifugation, dialyzed, and stored at −80°C. An E3-deleted conventional AV expressing firefly luciferase, Ad-L2, which was produced from pAdHM4-L2 and has deletions in the E1 gene (bp 342-3523) and E3 gene (bp 28133-30818), was previously prepared (24). The virus particles produced were determined using a spectrophotometric method (26), and biological titers were measured using an Adeno-X-rapid titer kit (Clontech, Mountain View, CA, USA). The ratio of the particle-to-biological titer was between 2.5 and 6.7 for each AV used in this study.
Western blotting. HEK293 cells were seeded on a 12-well plate at 5×105 cells/well. On the following day, cells were transduced with AVs at a multiplicity of infection (MOI) of 5. Whole-cell extracts were prepared 48 h after transduction and electrophoresed on 15% sodium dodecyl sulfate-polyacrylamide gels under reducing conditions, followed by electrotransfer to polyvinylidene difluoride membranes (Millipore, Darmstadt, Germany). After blocking with 5% skim milk prepared in Tris-buffered saline with Tween (Tween-20, 0.1%), the membrane was incubated with a rabbit ADP antibody (dilution 1:1,000; kindly provided by Ya-Fang Mei, Umeå University, Sweden), or rabbit anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody (dilution 1:5,000; R&D Systems, Minneapolis, MN, USA), followed by incubation with horseradish peroxidase-labeled anti-rabbit IgG antibody (Cell Signaling Technology, Danvers, MA, USA). The membrane was developed with a chemiluminescence kit (ECL Plus Western blotting detection system; Amersham Biosciences, Piscataway, NJ, USA), and then the signals were read with an LAS-3000 imaging system (Fujifilm, Tokyo).
Observation of CPE. HEK293 cells were seeded on a 6-well plate at a density of 5×105 cells/well. On the following day, cells were transfected with PacI-digested AV plasmids using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Cells were observed under a microscope every day from 7 days after transfection. For observation of CPE after transduction with the AVs, HEK293 cells were seeded into 12-well plates at a density of 4×105 cells/well. On the following day, cells were transduced with AVs at an MOI of 0.25, and then observed under a microscope.
Adenovirus-mediated gene transduction into cultured cells. Cells were seeded at a density of 1×104 cells/well on a 96-well plate. On the following day, cells were transduced with the AVs at MOIs of 20 and 100. Luciferase production was determined using a luciferase assay system (PicaGene LT 2.0; Toyo Inki, Tokyo, Japan) at 48-h after transduction. In AV-mediated transduction in HEK293 cells, HEK293 cells were seeded into 96-well plates at a density of 1×104 cells/well. On the following day, cells were transduced with AVs at an MOI of 0.25. Luciferase production was determined as described above at 24-h and 72-h after transduction.
Results
Production of AVs possessing the ADP gene. In order to develop an AV possessing the ADP gene with part of the E3 region deleted, we constructed the AV plasmids pAdHM309-E1, -E3 and -F (Figure 1), based on the sequence of the E3 region of the dl309 mutant adenovirus (22). pAdHM309-E1, -E3, and -F had a combined deletion of 3,984 bp, which consisted of a 3,182 bp deletion (bp 342-3523) in the E1 region and 752 bp deletion (bp 28597-28602, 30005-30750) in the E3 region, in the AV genome, indicating that a transgene expression cassette of less than 5.9 kb in length can be inserted in the AV genome. An adenovirus genome having a size approximately 105% of the size of the wild-type adenovirus genome can be incorporated into the virion (27). pAdHM309-E1, -E3, and -F have I-CeuI/SwaI/PI-SceI sites in the E1 region, the E3 region, and the region downstream of the fiber gene, respectively, for insertion of a transgene expression cassette. The AVs used in this study are shown in Table I. These AVs were normally propagated in HEK293 cells and exhibited titer productions comparable to those of a conventional AV, Ad-L2, which has deletion of the E1 region (bp 342-3523) and the E3 region (bp 28133-30818) (Table I and Figure 2). There were no significant differences in the virus titers between the AVs possessing the ADP gene and the conventional AV, Ad-L2 (Figure 2).
Schematic diagram of the adenovirus (Ad) vector plasmids used in this study. The luciferase expression cassette was inserted in the E1 region, the E3 region, or the region downstream of the fiber gene. CMV: Cytomegalovirus immediate-early promoter; Luc: firefly luciferase gene; pA: polyadenylation signal.
Adenovirus vectors produced in this study.
Dot plot showing the total virus titers of adenovirus (Ad) vectors used in this study. Each dot represents the virus titers produced after propagating viruses to 150Φ×5 dishes. The bars indicate the means of virus titers (n=3).
ADP expression during AV production. In order to examine whether the ADP gene was expressed from the AV genome during AV production, ADP expression was assessed by western blotting analysis (Figure 3). AdHM1-L2 is an AV with a deletion of the E1 region (bp 342-3523) but possessing the whole E3 region, including the ADP gene. ADP was expressed in all of the samples transduced with the AVs possessing the ADP gene, although the ADP expression levels in the cells treated with AdE3-L2, AdF-L1, and AdF-L2 were lower than those in the cells treated with the other AVs. This result indicated that the ADP gene was expressed during the production of AVs.
Adenovirus death protein (ADP) and hexon protein expression in HEK293 cells following treatment with adenovirus (Ad) vectors. HEK293 cells were treated with Ad vectors at a multiplicity of infection of 5. Whole-cell lysates were collected after a 48-h incubation, followed by western blotting analysis. Representative images from duplicate experiments are shown. GAPDH: Glyceraldehyde 3-phosphate dehydrogenase.
CPE after treatments with AV plasmids and AVs in HEK293 cells. In order to examine whether the ADP gene promoted the spread of infection with AVs in HEK293 cells, cells were observed following transfection with AV plasmids (Figure 4). HEK293 cells transfected with AV plasmids containing the ADP gene, pAdHM1-L2 (AV, AdHM1-L2), pAdHM309-E1-L2 (AV, AdE1-L2), and pAdHM309-F-L1 (AV, AdF-L1), showed apparent CPE on day 9, whereas only a small proportion of the HEK293 cells transfected with pAdHM4-L2 (AV, Ad-L2) showed CPE on day 9. This result suggested that the ADP gene enhanced the spread of infection with AVs following transfection with an AV plasmid.
Host cytopathic effect after transfection of adenovirus (Ad) vector plasmids. HEK293 cells were transfected with Ad vector plasmids. These images were taken 9 days post-transfection. Representative images from duplicate experiments are shown.
Next, in order to further examine the effects of ADP on AV production, the appearance of CPE was examined following transduction with AVs at an MOI of 0.25 in HEK293 cells (Table II). Most of the AVs maintaining the ADP gene started to show signs of CPE at 72-h post-transduction, while Ad-L2 showed no CPE at this time point. All the AVs containing the ADP gene mediated full CPE at 77-h post-transduction, while only about 30% of HEK293 cells transduced with Ad-L2 showed CPE. These results indicated that CPE occurred earlier in the AVs in which the ADP gene was retained, and that infection with the progeny virus spread faster to neighboring cells compared with a conventional AV.
Cytopathic effect (CPE) after transduction of adenovirus vectors at a multiplicity of infection of 0.25.
Spread of progeny virus after infection with adenovirus in HEK293 cells. In order to further examine whether the AVs retaining the ADP gene led to more efficient production and spread of progeny virus than a conventional AV without the ADP gene, the luciferase expression levels at 24-h and 72-h following transduction with luciferase-expressing AVs were determined in HEK293 cells at an MOI of 0.25 (Figure 5). Luciferase expression by all AVs maintaining the ADP gene, except for AdE3-L2, increased about 100-fold or more from 24-h to 72-h post-transduction, while Ad-L2-mediated luciferase expression increased by about 14-fold. These results indicated that the AVs maintaining the ADP gene mediated more efficient spread of infection of progeny virus than a conventional AV.
Increases in the luciferase production by HEK293 cells following 24-h and 72-h incubation with adenovirus (Ad) vectors at a multiplicity of infection of 0.25. Luciferase assay was performed after 24-h and 72-h of incubation. These data are expressed as the means±S.D. *Significantly different at p<0.05 (n=4).
Transduction efficiencies of AVs containing the transgene expression cassette in different sites. In order to determine the transduction efficiencies of the AVs with the transgene expression cassette in different sites, cells were transduced with AVs at MOIs of 20 and 100. Among the AVs examined, AdF-L1 showed the highest transduction efficiency in three human tumor cell lines (Figure 6). On the other hand, AdE1-L2 and AdE3-L2 mediated the lowest transduction efficiencies. These results indicate that the AV containing the transgene expression cassette in the region downstream of the fiber gene in a forward orientation mediated the highest transgene expression in cells.
Transduction efficiencies of adenovirus (Ad) vectors in human tumor cell lines. Panc-1 (A), H1299 (B), and SK HEP-1 cells (C) were transduced with the Ad vectors at multiplicities of infection (MOI) of 20 and 100. Luciferase assay was performed at 48-h post-transduction. These data are expressed as the means±S.D. (n=4). RLU: Relative luciferase units.
Discussion
The aim of this study was to develop a replication-incompetent AV that can be easily produced at higher titers than conventional AVs. For this purpose, the ADP gene was left in the AV genome. A previous study demonstrated that a recombinant Ad41 in which the ADP gene of Ad5 was genetically inserted propagated more efficiently than the wild-type Ad41 (13). Doronin et al. demonstrated that dl309, which was a replication-competent adenovirus possessing the ADP gene but lacking part of the E3 genes (bp 28597-28602, 30050-30750), efficiently induced CPE, showed efficient cell lysis, and promoted release of progeny virus to neighboring cells (8). These data indicated that ADP was crucial for progeny virus production.
The AVs possessing the ADP gene showed CPE following transfection with AV plasmids, and transduction with AVs in HEK293 cells was faster than conventional AV without the ADP gene, Ad-L2. In addition, the increases in luciferase expression from 24-h to 72-h after transduction were significantly higher for the AVs containing the ADP gene than the conventional AV at an MOI of 0.25 in HEK293 cells. These data indicate that the ADP contributed to efficient spread of infection with progeny virus to neighboring cells, induction of CPE and cell lysis.
Even though the ADP gene was left in the AVs developed in this study, there were no significant differences in the total virus titers among these AVs. This was probably because, for the propagation of AVs, HEK293 cells were treated with a large quantity of AVs at high MOIs. Had HEK293 cells been treated with AVs at low MOIs, the vector titers would have differed between the viruses with and those without the ADP gene.
Western blot analysis of ADP expression demonstrated that all of the AVs containing the ADP gene expressed ADP during AV propagation in HEK293 cells. However, AdF-L1 and AdF-L2 showed lower ADP expression compared to the other AVs, despite the comparable levels of hexon expression in all AVs. It remains unclear why AdF-L1 and AdF-L2 mediated low ADP expression in HEK293 cells. The transgene expression cassette introduced in the region downstream of the fiber gene may somehow have influenced ADP expression.
The inclusion of the ADP gene is more important in an oncolytic adenovirus than a replication-incompetent AV. ADP has been reported to promote the host CPE, host cell lysis, and subsequent release of progeny virus (6). These are key points in the infection with an oncolytic adenovirus. Overexpressing the ADP gene is one of the promising strategies for an oncolytic adenovirus (8, 11, 12, 14–17).
The AVs developed in this study can theoretically carry a transgene expression cassette of up to 5.9 kb. However, when a transgene expression cassette of 2.8 kb or less is inserted in the E3 region or the region behind the fiber gene, RCA may appear because the AV genome size does not exceed the packaging size limit following homologous recombination between the AV genome and the adenovirus genes in the genome of HEK293 cells. The adjustment of the size of the transgene expression cassette is important to avoid RCA generation. In addition, when a transgene expression cassette longer than 5.9 kb is to be loaded, the E3 gene must be further deleted. In a previous report, expression of the ADP gene was found in an adenovirus in which the ADP gene was retained but the rest of the E3 region was deleted (8).
In summary, we have developed AVs possessing an ADP gene with deletion of part of the E3 region. A transgene expression cassette can be introduced into the E1 region, the E3 region, or the region downstream of the fiber region in the AVs. AdF-L1, an AV with a transgene expression cassette inserted in forward orientation in the region downstream of the fiber gene, showed the highest transduction efficiency among the AVs examined. This study provides crucial information for the design of an efficient AV.
Acknowledgements
The Authors thank Eiko Sakai and Yukiko Toba (Graduate School of Pharmaceutical Sciences, Osaka University, Osaka, Japan) for helpful discussion. This study was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI [grant numbers 20H00664] and AMED [grant number 20fk0108543h0001].
Footnotes
Authors’ Contributions
AS designed and performed the experiments, analyzed the data, and wrote the article. FN performed the experiments. FS supervised the project and wrote the article. HM conceived and designed the project, performed the experiments, analyzed data, and reviewed the article.
Conflicts of Interest
The Authors declare no competing interests.
- Received January 27, 2022.
- Revision received February 15, 2022.
- Accepted February 16, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.











