Here, we present a protocol for adenovirus production using the pAdEasy system. The technology includes the recombination of the pAdTrack and pAdEasy-1 plasmids, the adenovirus packaging and amplification, the purification of the adenoviral particles from cell lysate and culture medium, the viral titration, and the functional testing of the adenovirus.
Adenoviral transduction has the advantage of a strong and transient induction of the expression of the gene of interest into a broad variety of cell types and organs. However, recombinant adenoviral technology is laborious, time-consuming, and expensive. Here, we present an improved protocol using the pAdEasy system to obtain purified adenoviral particles that can induce a strong green fluorescent protein (GFP) expression in transduced cells. The advantages of this improved method are faster preparation and decreased production cost compared to the original method developed by Bert Vogelstein. The main steps of the adenoviral technology are: (1) the recombination of pAdTrack-GFP with the pAdEasy-1 plasmid in BJ5183 bacteria; (2) the packaging of the adenoviral particles; (3) the amplification of the adenovirus in AD293 cells; (4) the purification of the adenoviral particles from cell lysate and culture medium; and (5) the viral titration and functional testing of the adenovirus. The improvements to the original method consist of (i) the recombination in BJ5183-containing pAdEasy-1 by chemical transformation of bacteria; (ii) the selection of recombinant clones by “negative” and “positive” PCR; (iii) the transfection of AD293 cells using the K2 transfection system for adenoviral packaging; (iv) the precipitation with ammonium sulfate of the viral particles released by AD293 cells in cell culture medium; and (v) the purification of the virus by one-step cesium chloride discontinuous gradient ultracentrifugation. A strong expression of the gene of interest (in this case, GFP) was obtained in different types of transduced cells (such as hepatocytes, endothelial cells) from various sources (human, bovine, murine). Adenoviral-mediated gene transfer represents one of the main tools for developing modern gene therapies.
Adenoviruses are nonenveloped viruses containing a nucleocapsid and a double-stranded linear DNA genome1,2,3. Adenoviruses can infect a broad range of cell types and infection is not dependent on active host cell division. After infection, the adenovirus introduces its genomic DNA into the host cell nucleus, where it stays epichromosomal and is transcribed together with the genes of the host. Thus, a minimal potential risk for insertional mutagenesis or oncogenes regulation is attained4,5,6. The adenoviral genome is not replicated together with the host genome and thus, the adenoviral genes are diluted in a dividing cell population. Among the advantages of adenoviral transduction, there are: (i) high levels of transgene expression; (ii) reduced risks related to the integration of the viral DNA into the host genome, due to episomal expression; (iii) transduction of a wide variety of dividing and non-dividing cell types. Most adenoviruses used in biomedical research are non-replicative, lacking the E1 region7,8,9. For their production, a cell line supplying the E1 sequence (such as HEK293) is required. Besides, a non-essential region for the viral life cycle (E3) was deleted to allow the insertion of a transgene in the viral genome; other regions (E2 and E4) were further deleted in some adenoviruses, but in these cases, a decreased yield of adenoviral production and low expression of the transgene were reported7.
Here, we present an improved protocol for constructing, packaging, and purifying the adenoviruses using the AdEasy System. These improvements allowed the packing of the adenovirus in a faster and more economical way as compared to the original method developed by Bert Vogelstein2,10, due to the following advantages: (i) the recombination in BJ5183-containing pAdEasy-1 by chemical transformation of bacteria; (ii) the selection of the recombinant clones by PCR; (iii) the transfection of AD293 cells using the K2 transfection system for adenoviral packaging; (iv) the precipitation of adenoviral particles from culture medium after viral packaging and amplification; (v) the adenoviral purification using one-step cesium chloride (CsCl) gradient ultracentrifugation.
The protocol for adenovirus production using the AdEasy system (Figure 1) comprises the following steps:
(1) Recombination of pAdTrack-GFP with pAdEasy-1 in BJ5183 bacteria
(2) Packaging of the adenoviral particles
(3) Amplification of the adenovirus
(4) Purification of the adenoviral particles from cell lysate and culture medium
(5) Adenovirus titration.
Figure 1: The adenovirus production technology. The main steps of the adenoviral technology are: (1) The recombination of the pAdTrack-GFP with the pAdEasy-1 plasmid in BJ5183 bacteria. The selected recombined plasmids are amplified in DH5α bacteria and then purified; (2) The packaging of the adenoviral particles in AD293 cells, that are producing adeno-E1 proteins; (3) The amplification of the adenovirus in AD293 cells; (4) The purification of the adenoviral particles from the cell lysate and the culture medium by ultracentrifugation on a CsCl density gradient; (5) The titration of the adenovirus and the functional testing. Please click here to view a larger version of this figure.
In this protocol, we exemplified the technology for the production of the adenovirus, which can induce the expression of GFP in the host cells. GFP is already inserted in the backbone of the pAdTrack-CMV shuttle vector (Addgene #16405), under a second CMV promoter and is used as a reporter gene (Figure 1). For this reason, here we designated the pAdTrack-CMV vector as pAdTrack-GFP and we assessed the expression of GFP for demonstrative purposes. Besides GFP expression, the system can be used to overexpress a gene of interest, which may be cloned in the multiple cloning sites of the pAdTrack-CMV. A gene or a minigene cloned in the pAdTrack-CMV is usually more efficient for expression induction as compared with the cDNA11. The data showed a strong GFP expression in transduced cells (such as hepatocytes, endothelial cells) from various sources (human, bovine, murine). Adenoviral-mediated gene transfer represents one of the main tools for developing modern gene therapies.
Safety Note: In general, adenoviruses are classified as biosafety level 2 organisms and thus, all manipulations must be done in a Class II biosafety cabinet by a trained person, wearing biohazard protective equipment (including gloves, face-mask for biological aerosols, lab coat, etc.). All solid materials contaminated with the adenovirus must be disinfected with a 10% bleach solution for 30 min and autoclaved for 30 min at 121 °C and 1 bar. Depending on the gene inserted, the adenovirus created may have dangerous potential and may be classified in other biosafety levels.
1. Experimental preparation
2. Recombination of pAdTrack-GFP viral vector with pAdEasy-1 plasmid in BJ5183 bacteria
3. Packaging the adenoviral particles
4. Amplification of the adenovirus
NOTE: If the AD293 cells did not reach the necessary confluence, the aliquots of the adenoviral stocks (lysate obtained from the virus-producing cells) to be used for infection may be stored at -80 °C.
5. Purification of the adenovirus from cell lysate and culture medium
6. Adenovirus titration
Figure 2: Titration plate design. Please click here to view a larger version of this figure.
7. Adenoviral transduction of target cells and testing of the induced protein expression
We modified and improved the original Vogelstein’s protocol in order to attain faster and more efficient adenovirus production. First, we revised the methodology to achieve an easier selection of recombinants. After recombination, the BJ5183 bacterial clones were tested by “negative PCR” to assess the integrity of pAdTrack-GFP as an indicator of the lack of recombination (Figure 3A), or by “positive PCR” to identify the gene of interest, assimilated in our case to GFP (Figure 3B). In both “negative” and “positive” PCRs, we used pAdTrack-GFP as a control template, which gave a band of 986 bp for pAdTrack integrity (Figure 3A, lane 1), and a band of 264 bp for GFP (Figure 3B, lane 3). The primers used for the “negative PCR” were designed to amplify a fragment of 986 bp containing the PmeI site in pAdTrack-GFP. This DNA fragment is drastically enlarged after recombination and is not amplified in the positive recombinant clones. Negative clones for recombination, in which pAdTrack-GFP remained intact, are represented in Figure 3A, lanes 3, 4, and 6. The primers anneal on the DNA sequences adjacent to the recombination site. Potential positive recombinant clones (Figure 3A, lanes 2 and 5) expressed GFP as shown in Figure 3B, lane 1, and 2. Plasmid DNA from these clones was isolated and used for DH5α transformation to obtain a higher amount of DNA. These preselected recombinant plasmids amplified in DH5α were then tested by enzymatic digestion. In Figure 3C-E are illustrated the results of the enzymatic digestion of one recombinant-positive clone digested with Hind III, PstI, BamHI restriction enzymes (Figure 3C, D, E lane 2). The HindIII and PstI digestion patterns of the recombinant clone were similar to those obtained for pAdEasy-1 since HindIII and PstI cut the pAdEasy-1 plasmid 24 and 25 times, respectively, (Figure 3C and D, lane 3); HindIII cut once and PstI cut four times the pAdTrack-GFP vector (Figure 3C and D, lane 1). BamHI cut twice pAdEasy-1 vector (Figure 3C, lane 3), and once pAdTrack-GFP (Figure 3C, lane 1).
PacI cut out a fragment of 4.5 kb from the recombinant plasmid (Figure 3F, lane 2), a fragment of 2863 bp from pAdTrack-GFP (Figure 3F, lane 1), and linearized the pAdEasy-1 vector (Figure 3F, lane 3). The DNA ladder is represented in Figure 3C-F, in lanes 4. The recombinant plasmid was digested with Pac I for further use for AD293 transfection.
Figure 3: The recombination of pAdTrack-GFP with the pAdEasy-1 plasmid. The plasmids obtained after the recombination of pAdTrack-GFP and pAdEasy-1 were tested by “negative” PCR for the pAdTrack-GFP integrity (A). The non-recombinant clones were evidenced by the presence of a 986 bp band corresponding to the sequence amplified from the pAdTrack-GFP plasmid (A, lanes 3, 4, and 6). The clones potentially positive for recombination (A, lanes 2 and 5) were also obtained. When the pAdTrack-GFP vector was used as a template, a band of 986 bp for pAdTrack-GFP (A, lane 1) was obtained. The potentially positive recombinant clones were tested for GFP expression by “positive” PCR (B); a band of 264 bp appears for both potentially recombined clones (B, lane 1 and 2), as well as for the pAdTrack-GFP plasmid. The DNA from one potential recombinant clone was tested with HindIII, PstI, BamHI, and PacI restriction enzyme (C-F, lanes 2). In the controls, the pAdEasy-1 vector (C-F, lanes 3) and the pAdTrack-GFP plasmid (C-F, lanes 1) were digested with the same enzymes. The DNA ladder is represented in C-F lane 4. Please click here to view a larger version of this figure.
The adenoviral packaging and amplification were performed in AD293 cells. The adenoviral particles (AdV-GFP) were purified from the AD293 cell lysate as well as from the cell culture medium, where they had been released by the infected cells. To concentrate the adenovirus found in the cell culture medium, the particles were precipitated with ammonium sulfate and then resuspended in 10 mM Tris HCl pH 8 with 2 mM MgCl2, the same buffer as that used for cell lysis. Subsequently, the adenoviral particles from the cell lysate and from the culture medium were purified by CsCl discontinuous gradient ultracentrifugation. After ultracentrifugation, a strong band of purified AdV-GFP was obtained, as shown in Figure 4.
Figure 4: The adenoviral purification by ultracentrifugation on a discontinuous CsCl gradient. The cell homogenate and the adenovirus precipitated from the medium were subjected to ultracentrifugation on a discontinuous gradient formed by low and high-density CsCl solutions. Strong bands of GFP- adenovirus were evidenced in both cases. Please click here to view a larger version of this figure.
To determine the viral titer expressed in transducing units per one mL (TU/mL), the AD293 cells were infected with serial dilutions of the AdV-GFP. After 48 hours, the infected cells expressed GFP, in an inverse correlation with the dilution factor of the viral suspension. This was observed by fluorescence microscopy and the percentage of GFP-positive cells was determined by flow cytometry (Figure 5). To calculate the titer, the viral dilution that induced 5 – 20% of GFP-positive cells was considered (Figure 5C). Usually, we obtain a viral titer of ~1010 (TU/mL) for GFP-adenovirus.
Below, we provide an example of an adenoviral titer calculation for a specific adenoviral batch in which 300000 cells (C) were transduced with 1 mL adenoviral solution (V), at a dilution factor of 106 (D), for which 6% GFP-positive cells (F) were obtained:
Titer (TU/mL) = D x F/100 x C/V = 106 x 6/100 x 300000/1 = 1.8 x 1010 TU/mL
Figure 5: The assessment of the adenoviral titer. AD293 cells were infected with various adenoviral dilutions. Forty-eight hours later, the cells were observed by fluorescence microscopy and analyzed by flow cytometry to determine the percentage of GFP positive cells induced by different adenoviral dilutions (A-D). To establish the gate for flow cytometry, non-transduced cells were also analyzed (E). The titer calculated for the dilution factor 106, when 6% of the cells were GFP positive was 1.8 x 1010 TU/mL. For panels A-E, bars: 100µm. Please click here to view a larger version of this figure.
To test the transduction potential of the prepared adenovirus, four cell lines were used: human endothelial cells (EA.hy926), bovine aortic endothelial cells (BAEC), murine hepatocytes (Hepa 1-6), and murine mesenchymal stromal cells (MSC). Endothelial cells (EA.hy926 and BAEC) were transduced with 25 TU/cell, the hepatocytes were transduced with 5 TU/cell and MSC were transduced with 250 TU/cell.
Here is an example of how the volume of adenoviral suspension needed to infect 3 x 106 cells with 25 TU/cell, using the adenoviral suspension with 1.8 x 1010 TU/mL, was calculated.
For 1 cell ………….. 25 TU
3 x 106 cells ………….. x TU x=75 x 106 TU
If the viral stock contains
1.8 x 1010 TU ………….. 1 mL
75 x 106 TU ………….. y mL y= 4.2 x 10-3 mL = 4.2µL of viral stock
Forty-eight hours after transduction, the cells were analyzed by fluorescence microscopy. As shown in Figure 6, human or bovine endothelial cells were transduced with good efficiency (~50%) for 25 TU/cell (Figure 6 EA.hy926 and BAEC). Murine hepatocytes (Hepa 1-6) were efficiently transduced by the adenovirus at a low amount of adenovirus particles (5 TU/cell), but they are also sensitive to the adenovirus since a higher percentage of dead cells (PI-positive cells) was recorded (~16%) as compared to the other cell types. Mesenchymal stromal cells were the most difficult to transduce (Figure 6), due to the lack of specific adenoviral receptors (unpublished data).
Figure 6: The infectivity of the adenovirus and the induction of GFP expression in transduced cells. Human endothelial cells (EA.hy926), bovine aortic endothelial cells (BAEC), murine hepatocytes (Hepa 1-6), and murine mesenchymal stromal cells (MSC) were transduced with the indicated amount of adenovirus. GFP was detected by fluorescence microscopy and the percentage of the GFP positive cells was analyzed by flow cytometry. PI-positive cells determined by flow cytometry show the cell mortality determined by viral transduction. EA.hy926 cells, bovine aortic endothelial cells, and Hepa 1-6 cells were highly transduced by the adenovirus, the yield of transduction ranging from 41 – 52%. For MSC, a higher amount of virus (250 TU/cells) induced only 27% GFP positive of the transduced cells. Bars: 100µm. Please click here to view a larger version of this figure.
Recombinant adenoviruses are a versatile tool for gene delivery and expression12,13,14. To induce strong protein expression by adenoviral transduction, the encoding sequence of the gene of interest is inserted in the genome of the adenovirus. The AdEasy adenoviral system, developed in the laboratory of Bert Vogelstein, comprises a backbone plasmid (pAdEasy-1) containing most of the wild-type adenovirus serotype 5 genome, and a shuttle vector (pAdTrack), designed for gene cloning2,10. The deletion of the adenoviral genes E1 (responsible for the assembly of infectious virus particles) and E3 (encoding proteins involved in evading host immunity) created a space in the adenoviral genome, in which a gene of interest of 6.5-7.5 kb can be inserted2,3. This size is sufficient for many genes, especially for those with shorter introns15,16,17. There are also researchers reporting the production of adenoviruses carrying the cDNA of a transgene18,19,20. However, we obtained a lower yield of transgene expression for cDNA-carrying adenoviruses than for their counterparts carrying a gene or a mini-gene (data not shown).
Improving and adapting the previous methods2,10,14,18,21, the technology for adenoviral production requires a shorter time, lower cost, and less effort. The full-length adenoviral DNA is obtained by recombination between the shuttle vector and the pAdEasy-1 plasmid in the homologous recombination prone E. coli strain, BJ5183. The protocol implies the chemical transformation of AdEasier-1 cells (BJ5183 bacteria containing pAdEasy-1). This technique does not require an electroporator that may not be available in some laboratories, is very simple, increases the recombination yield, and reduces the time necessary to obtain competent cells and to perform the transformation. The preselection of recombinant clones performed by PCR further shortens the time and eases the whole procedure. A similar procedure was used by Zhao and co-workers22, however, in the protocol, we optimized the sequences of the primers.
For the GFP-adenovirus packaging and amplification, a HEK293 derivative cell line was used, namely AD293 cells, which are more adherent to the culture plate. Other cell lines commonly used for adenoviral production are the following: 911, 293FT, pTG6559 (A549 derivative), PER.C6 (HER derivative), GH329 (HeLa derivative), N52.E6, and HeLa-E123,24,25,26. In our hands, no improvement in the adenoviral production was obtained when 911 cells were used (data not shown). The transfection of AD293 cells with the recombinant plasmid using K2 reagent highly increased the efficiency of the viral packaging step. After adenovirus production, up to ~70% of the adenovirus is still inside the cells and is released by three freezing and thawing cycles. Increasing the number of cycles is not suitable because it destroys the adenovirus.
Throughout the routine adenoviral production process, numerous viral particles are released in the cell culture medium. Discarding this cell culture medium during the harvesting of the infected AD293 cells would result in an important viral loss. We optimized the protocol described by Schagen and co-workers to purify the adenoviral particles from the cell culture medium by precipitation with ammonium sulfate27. This method has a higher efficiency in adenovirus recovery from the cell culture medium as compared with the method using polyethylene glycol28. The precipitated adenovirus should be purified immediately by ultracentrifugation or kept in the refrigerator for a couple of days but only after dialysis, to remove the salt excess. Keeping the precipitate longer than a few hours without dialysis is harmful to the virus.
Purification of the adenoviral particles by ultracentrifugation performed in one-step reduces the manipulation of the adenoviral stock and eases the procedure as compared with the protocols using successive ultracentrifugation steps14,29. Dialysis of the purified adenovirus is necessary to remove cesium chloride that may further affect transduction. In the protocol, we used Tris buffer containing MgCl2 but not sucrose for dialysis, since it requires a huge, unjustified amount of sucrose that is needed otherwise as a preservative for freezing. Thus, we added sucrose later, directly into the adenoviral stocks prepared for freezing. To avoid frequent freezing and thawing of the purified adenovirus, it is advisable to aliquot the adenoviral stocks and to store them at -80 °C. The adenoviral titer was evaluated by flow cytometry considering the GFP reporter gene and the percentage of transduced cells for a specific viral dilution. This method is faster as compared with the classical “plaque assay” and is more trustful as compared with the evaluation of the capsid proteins (by various methods such as ELISA or flow cytometry) which does not reveal the capacity of infection of the adenoviral particles. However, ELISA-based quantification, Q-PCR, or plaque assay using commercially available kits are alternative methods, especially useful for titration of adenoviruses which do not contain a fluorescent tracer.
Considering that pAdTrack adenoviruses are derived from human adenoviruses serotype 5 which is recognized by Coxsackievirus and Adenovirus Receptors (CAR), we demonstrated the capacity of the GFP-adenovirus to transduce cells of human origin (endothelial cells), but also cells of other origins: bovine (endothelial cells) and murine (mesenchymal stromal cells and hepatocytes). The data showed that the GFP-adenovirus can induce a high level of expression of a transgene.
In conclusion, we optimized this laborious technology to reduce the time, the costs, and the effort needed to obtain the adenoviral particles. The adenovirus prepared is able to infect various cell types and to induce the expression of the gene of interest. This protocol may be used in a variety of experiments since the adenoviral-mediated gene transfer represents one of the main tools for developing modern gene therapies.
ABBREVIATIONS: AdV-GFP, adenoviral particles; BAEC, bovine aortic endothelial cells; CsCl, cesium chloride; GFP, green fluorescent protein; MSC, mesenchymal stromal cells; TU, transducing units.
The authors have nothing to disclose.
This work was supported by a Project co-financed from the European Regional Development Fund through the Competitiveness Operational Program 2014-2020 (POC-A.1-A.1.1.4-E-2015, ID: P_37_668; acronym DIABETER), a grant of the Romanian Ministry of Research and Innovation PCCDI- UEFISCDI, Project number PN-III-P1-1.2-PCCDI-2017-0697 within PNCD III and by the Romanian Academy. The authors thank Kyriakos Kypreos (University of Patras, Greece) for his generous and relevant advice, Ovidiu Croitoru (University of Fine Arts, Bucharest, Romania) for filming, film editing, and graphical design, and Mihaela Bratu for technical assistance.
AD293 cells | Agilent Technologies | 240085 | |
AdEasier-1 cells | Addgene | 16399 | |
Agarose I (for electrophoresis) | Thermo Scientific | 17850 | |
Ammonium sulfate | Sigma | A4418 | |
Ampicillin sodium salt | Sigma | A0166 | |
BamH I | Thermo Scientific | FD0054 | |
Cell culture plates 100 mm | Eppendorf | 30702115 | |
Cesium chloride | Sigma | L4036 | |
DH5alpha bacteria | Thermo Scientific | 18265017 | |
DMEM (GlutaMAX, 4.5g/L D-Glucose) | Gibco | 3240-027 | |
EA.hy926 cells | ATCC | CRL-2922 | |
EDTA | Sigma | E5134 | |
Ethanol (99.8%) | Roth | 5054.2 | |
Fetal Bovine Serum | Sigma | F7524 | |
Flasks T25, T75, T175 | Eppendorf | 30712129 | |
Glucose | Sigma | G7021 | |
Hepa 1-6 murine hepatocytes | ATCC | CRL-1830 | |
Hind III | Thermo Scientific | FD0504 | |
Kanamycin Sulfate | Thermo Scientific | 15160054 | |
K2 Transfection System | Biontex | T060-5.0 | |
LB medium | Formedium | LBx0102 | |
LB-agar | Formedium | LBx0202 | |
Mix & Go E. coli Transformation kit | Zymo Research | T3001 | |
Midori Green Advanced DNA stain | Nippon Genetics Europe | MG-04 | |
NaOH | Sigma | S8045 | |
Opti-MEM | Thermo Scientific | 31985070 | |
Pac I | Thermo Scientific | FD2204 | |
pAdEasy-1 | Addgene | 16400 | |
pAdTrack-CMV | Addgene | 16405 | |
Phenol:chloroform:isoamyl alcohol (24:24:1) | Invitrogen | 15593-031 | |
Polymerase GoTaq | Promega | M3005 | |
Pme I (Mss I) | Thermo Scientific | FD1344 | |
Potassium acetate | VWR Chemicals | 43065P | |
Pst I | Thermo Scientific | FD0614 | |
Qiagen Midi Prep kit | Qiagen | 12125 | |
Cell Scraper | TPP | 99003 | |
SDS | Thermo Scientific | 28365 | |
Slide-A-Lyzer dialysis cassettes | Thermo Scientific | 66330 | |
Sodium pyruvate | SIGMA | P5280-100G | |
Syringe with 23G neeedle | B Braun | 464BR | |
Tris HCl | Sigma | 1185-53-1 | |
Trypan blue | Roth | CN76.1 | |
Tubes 50ml | TPP | 91050 | |
Ultra-Clear Tubes (14×89 mm) | Beckman Coulter | 344059 | |
Centrifuge (refrigerated) | Sigma Sartorius | 3-19KS | |
HeraeusFresco 17 Microcentrifuge | Thermo Scientific | 75002420 | |
Ultracentrifuge with SW41Ti rotor | Beckman Coulter | Optima L-80 XP | |
Culture Hood | Thermo Scientific | Class II | |
Pipettes (0-2µl, 1-10µl, 2-20µl, 10-100µl, 20-200µl, 100-1000µl) | Thermo Scientific | ||
Dry Block Heating Thermostat | Biosan | TDB-120 | |
Thermocycle | SensoQuest | 012-103 | |
Water Bath | Memmert | WNB 14 |