2018 Volume 41 Issue 10 Pages 1615-1619
In recent times, oncolytic viruses expressing an extraneous gene have attracted great interest; in fact, they have been engaged in multiple applications, such as medicine for cancer. Our group made an oncolytic adenovirus, namely, OBP-301, for use in treating solid cancers and press clinical trial to get approval for a pharmaceutical product. In this study, we applied a flow cytometry-based method to determine the titer of adenoviruses expressing an extraneous gene as well as assess their quality. We considered using the green fluorescent protein (GFP)50 titer as a measure of viral quality. The GFP50 titer (GFP50/mL) is the viral load required to render the HeLa S3 cell line 50% GFP-positive by analysing flow cytometry data. We measured the GFP50 titers for three types of recombinant adenoviruses (OBP-401, OBP-1101, and OBP-1106). We compared GFP50/mL and tissue culture infectious dose (TCID50/mL), a conventional titration index, and found that these titers showed a linear correlation, with a correlation coefficient of >0.9. Moreover, GFP50/mL showed high repetitive accuracy. We expect this flow cytometry-based method to be useful in case of clinically relevant viruses expressing an extraneous gene, in particular, to control viral quality.
OBP-301 (Telomelysin) is a genetically engineered serotype 5 adenovirus containing the hTERTp-E1A-IRES-E1B cassette inserted into the E1 region, which can replicate efficiently and induce cytopathic effect (CPE) on a panel of human cancer cell lines; however, its replication as well as cytotoxicity was highly attenuated in normal human fibroblasts lacking telomerase activity. The clinical safety of using this serotype for various solid tumours has been determined by phase I clinical trials.1,2)
Use of viruses such as genetically modified viruses in the research and development of pharmaceutical products deems it essential to estimate their quality. Viral titer is one of the most important factors used to evaluate viral quality. The assays involving determination of plaque-forming units (PFU) or the tissue culture infectious dose (TCID50) are conventional methods based on titration and depend on microscopic observations of CPE of cells.3) However, these titration-based methods have low repetitive accuracy, especially because of the subjective nature of CPE reporting. Previous study suggested enumerating the cells expressing a reporter protein such as β-galactosidase4) or green fluorescent protein (GFP)-positive cells,5,6) instead of determining the number of plaques formed. Most of these methods are relatively rapid but require multiple steps; therefore, the accuracy and reproducibility are commonly dependent on experimental conditions and operator.5,6) Some studies showed that the titer value obtained by analysing flow cytometry data was almost identical to the PFU values.5,6)
To develop a model of adenoviruses expressing an extraneous gene, we infected HeLa S3 cells with three kinds of recombinant adenoviruses expressing GFP (OBP-401,7) OBP-1101,8) and OBP-1106) (Fig. 1), and the population of GFP-expressing cells was analysed by flow cytometry. Quality control of such oncolytic viruses includes viral titer as the most important characteristic, which can be determined by a precise titration assay.
(A) OBP-401 has the hTERTp-E1A-IRES-E1B cassette inserted into the E1 region and the CMVp-GFP cassette in the E3 region.7) (B) OBP-1101 is a partially modified form of OBP-401, possessing four copies of a target sequence perfectly complementary to the miR-142-3p present in the 3′-UTR of the E1 region and the EGFP gene in the E3 region. In addition, the part of fibre protein is substituted from type 5 to type 35.8) (C) OBP-1106, a by-product of OBP-1101 generated during the plaque purification process, has the same genome structure as OBP-1101, except for the CMV promoter, and is thought to express GFP by promoting the residual E3 region after infection of the cell line.
The objectives of this study were to propose a new method to determine the viral titer using GFP50 value (GFP50/mL) by analysing the flow cytometry data and to compare GFP50/mL with TCID50/mL, a conventional evaluation index. Their correlation and precision values were also considered.
The three kinds of recombinant adenovirus (OBP-401,7) OBP-1101,8) and OBP-1106) (Fig. 1) were propagated in HeLa S3 cells, purified by ion-exchange chromatography and several steps of filtration, and stored at −80°C in the Kobe Clinical Laboratory Center, Oncolys BioPharma Inc.
Cell CultureFor the GFP50 titer assay, the HeLa S3 human cervical cancer cell line was obtained from the American Type Culture Collection (ATC C, Manassas, VA, U.S.A.; CCL-2.2™) and adapted at Introgen (Houston, TX, U.S.A.) to grow in serum-free CD293 medium without L-glutamine and phenol red (Thermo Fisher Scientific; 03-0021). The medium was supplemented with 2% Glutamax 1 (Life Technologies; 35050061) and equilibrated at 37°C in the presence of 10% CO2 before use. The HeLa S3 cells were cultured for 5 d at 37°C in an incubator containing 10% CO2, with continuous agitation at 100 rpm before use; the cell density was maintained at 0.4–1.6×106 cells/mL. Cell enumeration was performed using the standard trypan blue method and the cell counter (model R1; Olympus, Tokyo, Japan).
For the TCID50 titer assay, HEK-293 cell line (CRL-1573™) was obtained from ATC C. These cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Thermo Fisher Scientific; 11995065) with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific; 10270-106) (10% FBS/DMEM) containing 1% penicillin/streptomycin (Merck; P0781) for 7 d at 37°C in the presence of 5% CO2 before use.
Both cell lines express both receptors coxsackie and adenovirus receptor (CAR) and CD46, and hence adenovirus serotype 5 and 35 can transduce the both cell lines.
Flow Cytometry and GFP50 Titer DeterminationHeLa S3 cells were counted and seeded into each well of a 24-well plate, at a density of 4×105 cells in a volume of 0.5 mL. The three types of viruses were serially diluted in the culture medium, and each 0.02 mL of the virus dilutions were independently added to the cells in the wells. The assay was individually performed three times, and each viral dilution was assayed in triplicate. The cells were incubated at 37°C, in the presence of 10% CO2, with continuous agitation at 100 rpm. After a 24-h-long incubation period, the cells were collected, fixed by 4% paraformaldehyde prepared in phosphate-buffered saline (PBS) (Nakalai Tesque, Kyoto, Japan) and re-suspended in FACS buffer (2% FBS/PBS containing 1% penicillin/streptomycin).
The GFP-positive rate of HeLa S3 cells was determined using the MACS Quant flow cytometer (Miltenyi Biotec, Bergisch Gladbach, Germany). Thirty-thousand events were counted for each sample. The analytical gates were set such that the GFP-positive rate of the uninfected cell sample is ≤0.5%. The GFP-positive rate was determined for each viral dilution by using the FlowJo software (FlowJo, LLC, Ashland, OR, U.S.A.). The GFP-positive rates were normalized, so that the positive rate reached the peak at a multiplicity of infection (MOI) of 100% (based on the concept of ‘relative EC50’ described by Sebaugh9)). Non-linear regression curve fitting and EC50 calculation were performed using GraphPad Prism (GraphPad Software Inc., La Jolla, CA, U.S.A.). The EC50 value is considered as the GFP50 value of the virus.
GFP50/mL were calculated from the viral dilution factor at 50% GFP-positive rate and the corresponding volume (Fig. 2). The functional viral concentration was defined according to the following formula5,6):
 |
GFP-positive rate of infected cells shows sigmoidal curve. The dilution factor when relative GFP-positive rate is 50% is calculated by non-linear regression.
Viruses were serially diluted ten-fold by using 5% culture medium, and a 105 dilution of OBP-1101 and OBP-1106 and a 107 dilution of OBP-401 were prepared. The viral solutions were further diluted three-fold on a 96-well plate. About 5000 HEK-293 cells were added to each well and cultured for 11 d. The CPE was observed, and TCID50/mL and VP/TCID50 ratio were calculated.
One of the traditional methods used for physical titration is the measurement of the optical density (OD260)10) in order to determine the DNA content of the stock solution. This method assumes that an OD260 of 1 is equivalent to 1.1×1012 viral particles.5) The total number of viral particles is determined in terms of viral particle per millilitre (VP/mL). The OD260 of each sample was measured by absorption spectrophotometry, after preparing the sample to contain 0.1% sodium dodecyl sulfate (SDS). The VP/mL value was determined for all three types of recombinant adenoviruses (Table 1).
| Virus name | VP/mL | TCID50/mL (%CV) | GFP50/mL (%CV) | VP/TCID50 | VP/GFP50 |
|---|---|---|---|---|---|
| OBP-401 | 3.7×1012 | 5.3×1010±2.2×1010 | 1.4×1011±1.0×1010 | 70 | 26 |
| (41%) | (7%) | ||||
| OBP-1101 | 6.2×1011 | 6.1×108±2.1×108 | 1.4×109±2.2×108 | 1016 | 443 |
| (34%) | (15%) | ||||
| OBP-1106 | 1.2×1012 | 2.1×109±4.3×108 | 7.5×109±7.0×108 | 571 | 156 |
| (20%) | (9%) |
Date are presented as the mean±standard deviation and the %CV (n=3). VP, virus particle; TCID50/mL, tissue culture infectious dose per mL; GFP50/mL, green fluorescent protein titer units per mL.
On the other hand, the biological titration methods involve the estimation of the potency of a viral stock in transduced cells. The classical method is the PFU assay, in which the patches of lysed cells in a monolayer are counted to determine the viral titer in terms of PFU/mL.11) In the monolayers formed by the 293 cells, the plaques become visible in 4–5 d, and can be counted at 6–8 d.12) The determination of TCID50 using monolayer cultures in 96-well plates is a variation of this technique13); a TCID50/mL of 1 approximately corresponds to a PFU/mL of 1.3) We reported this value in terms of TCID50 value per millilitre in this study for the three types of adenoviruses, which are shown in Table 1.
The VP/TCID50 value obtained for OBP-1101 was found to be 13-times greater than that obtained for OBP-401. In addition, the VP/TCID50 value obtained for OBP-1101 was 2-times greater than that obtained for OBP-1101. The findings might be influenced by the different combinations of the gene sequences of CMVp, GFP, and mir-142-3 pT and/or the differences in the manufacturing methods, as discussed in a previous study.7)
Flow CytometrySeveral methods for virus titration have been reported to use the positive rate of reporter protein-expressing cells, with GFP often being used as the reporter protein.5,6) In these methods, however, the GFP-positive rate might not correctly reflect the viral titer at high or low infectious doses (i.e., non-exponential increasing part of sigmoidal curve in Fig. 2). We incorporated the theory of EC50 value and fixed the positive rate for titer evaluation at 50%, to increase the accuracy of the assay. The VP/TCID50 and VP/GFP50 values of the three viruses were proportional (Fig. 3). The relative standard deviation observed in the GFP50/mL was <15%, which falls within the acceptable range defined by the U.S. Food and Drug Administration (FDA) (25%) for cell-based bioassays, in contrast that the deviation of VP/TCID50 was larger than 25%.
In a series of in vitro experiments, each adenovirus stock was assayed by TCID50/mL and GFP50/mL. In addition, each value of VP/GFP50 and VP/TCID50 was calculated. Each set of three data points was used to calculate a linear curve and to calculate a correlation coefficient.
In this study, we proposed a modified method of virus titration, which showed proportional correlation and relatively high precision to the conventional method of titration (TCID50). OBP-401 and OBP-1101 have a GFP gene, which allows the development of a precise surgical navigation, a screening method for early detection of cancer, and/or a method for post-operative examination.7,8,14–21) In addition, OBP-702, a second-generation candidate of OBP-301 containing the hTERTp-E1A-IRES-E1B cassette inserted into the E1 region,1,2) carries the p53 expression cassette.22–25) In recent times, different types of recombinant oncolytic viruses have been developed, which enhance the functioning of the immune system against cancer, by attenuating the PD-1/PD-L1 axis and/or enhancing cancer immune system.26–28) The high precision of expression titers like a GFP50 titer helps accurate determination of the viral titer and transgene expression level, which, in turn, contributes to the quality control regimen for pharmaceutical products. Expression levels of transgene products can be measured by immunofluorescent staining using specific antibodies, instead of using reporter genes, including GFP.
The findings of this study confirmed that the VP/GFP50 and VP/TCID50 values were nearly identical on the basis of the flow cytometry data obtained for the cells, which were individually infected with OBP-401, OBP-1101, and OBP-1106 to generate different models of oncolytic adenoviruses expressing an extraneous gene. Moreover, because the GFP50/mL, have a high repetitive accuracy, we considered that the VP/GFP50 value could be an evaluation index superior to the VP/TID50 value for the qualitative determination of oncolytic adenoviruses expressing an extraneous gene. This method would be useful to control the quality of viruses expressing an extraneous gene because it can help confirm the validity of other genes, except GFP.
We would like to thank Shunsuke Iizuka (Osaka University) for his support in the use of the flow cytometer in this study.
Hiroyuki Mizuguchi is a consultant of Oncolys BioPharma Inc. Yasuo Urata is President and CEO of Oncolys BioPharma Inc. Akikazu Asada, Hideki Hayakawa, Natsuki Yanase and Kanae Abe are employees of Oncolys Biopharma Inc. Fuminori Sakurai and Hiroyuki Mizuguchi have the potential to receive patent royalties from Oncolys BioPharma Inc.
