Here we describe a protocol based only on cell infection, which improves the efficiency of recombinant parvovirus production by more than 100 fold in comparison to other protocols in use. This protocol relies on the use of a novel adenovirus 5-based helper containing the parvovirus VP transcription unit (Ad-VP).
Rodent parvoviruses (PV) such as rat H-1PV and MVM, are small icosahedral, single stranded, DNA viruses. Their genome includes two promoters P4 and P38 which regulate the expression of non-structural (NS1 and NS2) and capsid proteins (VP1 and VP2) respectively1. They attract high interest as anticancer agents for their oncolytic and oncosuppressive abilities while being non-pathogenic for humans2. NS1 is the major effector of viral cytotoxicity3. In order to further enhance their natural antineoplastic activities, derivatives from these vectors have been generated by replacing the gene encoding for the capsid proteins with a therapeutic transgene (e.g. a cytotoxic polypeptide, cytokine, chemokine, tumour suppressor gene etc.)4. The recombinant parvoviruses (recPVs) vector retains the NS1/2 coding sequences and the PV genome telomeres which are necessary for viral DNA amplification and packaging. Production of recPVs occurs only in the producer cells (generally HEK293T), by co-transfecting the cells with a second vector (pCMV-VP) expressing the gene encoding for the VP proteins (Fig. 1)4. The recPV vectors generated in this way are replication defective. Although recPVs proved to possess enhanced oncotoxic activities with respect to the parental viruses from which they have been generated, their production remains a major challenge and strongly hampers the use of these agents in anti-cancer clinical applications.
We found that introduction of an Ad-5 derived vector containing the E2a, E4(orf6) and the VA RNA genes (e.g. pXX6 plasmid) into HEK293T improved the production of recPVs by more than 10 fold in comparison to other protocols in use. Based on this finding, we have constructed a novel Ad-VP-helper that contains the genomic adenoviral elements necessary to enhance recPVs production as well as the parvovirus VP gene unit5. The use of Ad-VP-helper, allows production of rec-PVs using a protocol that relies entirely on viral infection steps (as opposed to plasmid transfection), making possible the use of cell lines that are difficult to transfect (e.g. NB324K) (Fig. 2). We present a method that greatly improves the amount of recombinant virus produced, reducing both the production time and costs, without affecting the quality of the final product5. In addition, large scale production of recPV (in suspension cells and bioreactors) is now conceivable.
Note that a laboratory with a biosafety level 2 is required for the production of recombinant parvoviruses (recPV).
The protocol is subdivided in two main parts. The first part (production of recPV via transfection) is required to produce the minimal amount of recPVs that serves as inoculum in the second part of the protocol (production of recPVs via infection). Once a small amount of recPVs is produced, the first part of the protocol can be omitted, and the recombinant parvovirus can be amplified only via infection providing the gene encoding for the parvovirus capsid proteins through the adenovirus helper (see below).
1. Production of recPVs via Transfection
1.1 Viral DNA transfection
It is possible to use any established DNA transfection method in this section of the protocol either based on cationic lipids or calcium phosphate. We generally use Fugene HD Transfection Reagent. To control transfection efficiency, we recommend to transfecting additional plate with a plasmid expressing enhanced Green Fluorescent Protein (e.g. pEGFP-N1, Clontech). Transfection is considered efficient and optimal for virus production when at least 50% of the cell population results EGFP-positive 24 hours after transfection.
1.2 Virus production
1.3 Virus harvest
1.4 Virus storage
2. Production of recPVs via Infection
Before starting with the recPV production via infection, produce and purify Adenovirus 5 carrying the parvovirus VP gene (Ad-VP helper, described in5) according to standard protocol8. It is also necessary to have a minimal amount of recPVs carrying the transgene of interest (e.g., produced via transfection as described above).
Below, we describe the recPV production in 175 cm2 flask (T175) format. Further amplification of the virus stock produced in this way can be conducted in 10 multilayer CellSTACK culture chambers (Corning) with minor adjustments of the protocol proposed.
2.1 Infection
2.2 Virus production
Incubate the flasks at 37 °C with 92% humidity and 5% CO2 for 48 hours. As indication of efficient viral production, clear signs of cytotoxicity should be observed in the flask containing viral infected cells, starting 36-48 hours post-infection.
2.3 Virus harvest
2.4 Virus storage
3. RecPV Purification
4. Recombinant Parvovirus Titration
5. Quality Controls
6. Representative Results
An example of recPVs production via transfection in the presence or absence of adenovirus genomic elements is shown in Figure 3. Cells were transfected with pCMV/VP (plasmid carrying the gene encoding for the VP parvovirus capsid proteins) together with phH-1-GFP (a recPV harbouring the GFP gene) or phH-1-luciferase (a recPV harbouring the firefly luciferase gene), with (+ pXX6) or without (-pXX6) the pXX6 plasmid (carrying the adenovirus E2A, E4(orf6) and VA RNA genes). Equal volume of crude cell extracts were applied to NB324K cells and GFP transduction or luciferase assays were performed as reported in El-Andaloussi et al.5. A clear increase in the recPVs production was obtained in the presence of pXX6 with production increasing from 0.3 GFP transductional units (TU)/cell obtained according to conventional protocols to approximately 5 TU/cell obtained following our method (Fig. 3A, B). A significant increase (about 24-fold) of recPV production was also observed in the case of phH-1-luciferase (Fig. 3C).These results indicate that the genetic material contained in pXX6 is able to boost parvovirus production.
A representative example of recPVs production via infection is shown in Figure 4. NB324K cells were co-infected with various recPVs (as indicated in the figure) and Ad-VP-helper (harbouring the gene encoding for the PV VP capsid proteins). Ad-VP helper further enhanced recPV production up to >70 TU per seeded cell (Fig. 4A) without increasing the occurrence of undesirable replication competent viral particles (Fig. 4B).
Figure 1. Vectors based on autonomous parvoviruses. (A) Top: The transgene replaces part of the VP-coding genes and is under the control of the viral promoter P38. The genes for NS1/2 are retained and their expression is controlled by the viral promoter P4. The vector genome is flanked by the parvoviral ITRs, which contain cis-acting elements that are required for replication and packaging of the recombinant genome. Bottom: A plasmid carrying the VP-gene under either a heterologous (e.g. CMV), or autologous (e.g. P38) promoter (Px), is supplied in trans during recombinant parvovirus production in order to compensate for the disruption of the structural genes in the recombinant genome. ITR, inverted terminal repeat. Figure adapted from 4. (B) Schematic view of the classical protocol used for the production of recPVs. HEK293T cells are transiently transfected with viral DNA (vector and helper plasmids) and after three days, cells are collected and viruses harvested.
Figure 2. Production of recPVs with the help of the Ad-VP. (A) Schematic maps of recPV and Ad-VP genomes. The recPV contains a heterologous transgene which replaces part of the VP region. The Ad-VP harbors the parvovirus VP gene. (B) Schematic view of the protocol described in this manuscript. NB324K cells are co-infected with recPV and Ad-VP viruses. After three days cells are harvested and recPV particles recovered from cell lysate.
Figure 3. Stimulation of recPV production by adenovirus-based plasmids pXX6 . HEK293T cells, seeded in 10 cm dishes, were transfected with pCMV/VP in combination with phH-1-GFP (A and B) or phH-1-luciferase (C) to produce recPVs. Simultaneously, cells were co-transfected with the adenovirus-derived helper plasmids pXX6 or not, as indicated. Three days post-transfection, cells were harvested and lysed by three freeze and thaw cycles. Equal volumes of crude virus extracts were applied to NBK indicator cells, and transduction assays were carried out. (A) Representative micrographs showing GFP positive cells within confluent NB324K monolayers. (B) Quantification of the GFP transduction assays expressed in transduction unit (TU) per seeded cell. (C) Quantification of the luciferase activity expressed as relative luciferase units (RLU). The luciferase assay was performed as described in El-Andaloussi et al.5. Columns represent average values from three replicates with standard deviation bars. Number on top of the + pXX6 column, in (B and C), indicates the fold increase in the recPV virus titers, obtained in the presence of pXX6 versus without.
Figure 4. Stimulation of recPV production by means of recombinant Ad-VP helper virus . (A) NB324K cells, were infected with purified Ad-VP helper virus at MOI of 10 (Ad-X unit/cell, titrated with Adeno-X Rapid Titer Kit), and then superinfected with either of the following recombinant parvovirus Chi-hH-1-EGFP11 (0.1 TU/cell) or H-1-GFP (0.5 TU/cell)5. One day post-infection, medium was changed and two days later, cells were collected and lysed via three freeze-and-thaw cycles. Crude cell extracts were used to determine virus titers by transduction assay according to El-Andaloussi et al.5. TU, transduction unit. (B) Viral batches produced in the presence of absence of Ad-VP were analyzed for their content of replication competent viral particles (RCV) by plaque assay on NB324K indicator cells.
We have shown that recPV production can be enhanced by the presence of adenoviral genomic elements. We have increased the recPV yields by more than 10 fold (from 0.3 to 5 TU/cell) by providing adenovirus genomic element via transfection and by more than 100 fold co-infecting the cells with Ad-VP-helper in combination with the recPV in comparison to conventional protocols. The protocol described here can be further optimized by determining the most appropriate timing for the delivery of the adenovirus genomic elements and the optimal concentration of Ad-VP helper to be used for the infection.
RecPVs with larger transgenes (more than 1,600 bases) are less efficiently produced. This is probably due to the fact that by inserting the transgene, cis-acting elements necessary for PV proper viral DNA packaging are removed. The optimal size of transgene that can be inserted into the PV genome without affecting its production is up to 700 bases 6. Also the sort of the transgene inserted may affect recPV production (e.g. insertion of genes encoding for cytotoxic proteins may result in lower recPV yields).
Another important aspect to take in consideration during recPVs production is the possible occurrence of replication competent viruses (RCV) that could spontaneously be formed through homologous recombination between the recombinant parvovirus and VP-containing plasmids. The present protocol does not enhance the recombination risk. Nevertheless a RCV quality control should be routinely performed (e.g. plaque assay in NB324K indicator cells5) at the end of the production in order to assure that viral stocks contain only a negligible amount of RCVs.
The protocol described overcomes previous limitations in the choice of the packaging cell line used, as it renders possible the use of cell lines that are difficult to transfect (and therefore not suitable with protocols based on transfection) but are good producers of PVs. e.g. NB324K cells10 that in our hands were more efficient in producing recPVs than HEK293T (data not shown). A screening of different cell lines is now possible applying the new protocol, and could identify more efficient cells for recPV production. This also paves the way to further exploration of the system for large-scale production of recombinant parvovirus e.g. in bioreactors with cells in suspension.
The authors have nothing to disclose.
We thank the team of the DKFZ virus production and development unit, in particular Marcus Müller, Silvia Münstermann, Barbara Liebetrau, and Mandy Roscher. This study has been partly supported by grants from the Federal Ministry of Education and Research (BMBF) and the Helmholtz Association in the framework of the Deutsches Krebsforschungszentrum / Cancéropôle du Grand-Est joint Programme in Applied Tumour Virology.
Name of the reagent | Company | Catalogue number |
DMEM | Sigma | D5796 |
MEM | Sigma | M4655 |
Foetal Bovine Serum | PAA | A15-101 |
L-Glutamine | Gibco | 25030-024 |
Fugene | Roche | 047097050001 |
Trypsin-EDTA | Invitrogen | 25300062 |
Benzonase Nuclease | Sigma | E8263 |
Adeno-X Rapid Titer Kit | Clontech | 632250 |