In children and adolescents, cancers are the diseases with the highest mortality [1
]. Osteosarcoma is the most common type of primary bone cancer and the eighth most common form of pediatric cancer [3
]. Current treatment strategies, introduced in the 1980s, include neo-adjuvant chemotherapy, resection of all primary and metastatic lesions, and subsequent adjuvant chemotherapy [4
]. These strategies achieve average 5-year survival rates of 60–70%. Approximately 20% of osteosarcoma patients initially present with metastatic disease, predominantly in the lungs. In metastatic osteosarcoma patients, the 5-year survival rate is drastically reduced to 15–30% [6
]. Especially for patients with high-risk or metastatic disease, novel treatment approaches are urgently needed. Among the targeted therapies now available to refractory or relapsed osteosarcoma patients and having shown first survival benefits [8
], immunomodulatory approaches stimulating the antitumor immune response are considered the most promising [9
Oncolytic viruses are a class of promising, potent, and highly specific anticancer agents, combining specific cytotoxicity towards transformed cells with an immunotherapeutic action [11
]. These viruses induce lytic infection of malignant cells and thereby stimulate the innate and adaptive immune systems by promoting the availability of tumor antigens which initiate cross-priming and vaccination effects [11
]. The use of such viruses can also be synergistically combined with other treatment modalities inducing immunogenic cancer cell death [15
]. The efficacy of several oncolytic viruses has been proven through in vivo preclinical osteosarcoma studies. In a study where wild-type oncolytic Seneca Valley Virus was tested on six human osteosarcoma–xenograft severe combined immuno-deficient mouse models, only two models showed significantly extended event-free survival and no model displayed any effect on overall survival [16
]. In another study, the wild-type Semliki Forest Virus was found to increase dramatically the survival of osteosarcoma–xenograft nude mice [17
]. In xenograft mice with localized or metastatic osteosarcoma, the recombinant, conditionally replicating adenoviruses Ad5-Δ24RGD, Ad-OC-E1a, and Adenovirus VCN-01 significantly delayed tumor growth (localized tumors) or reduced number of metastatic lesions (pulmonary metastases) [18
]. In the first phase I clinical trial of the naturally occurring reovirus isolate Reolysin in pediatric sarcoma patients (including osteosarcoma patients), safety data have been published [22
The rodent protoparvovirus H-1 (H-1PV) is another promising therapeutic agent. Its clinical safety upon intratumoral or intravenous injection has been demonstrated in a recently published first phase I/IIa clinical trial recruiting adult glioblastoma patients [23
]. H-1PV is a wild-type oncolytic virus occurring naturally in rats. A natural tropism of this virus for the developing skeleton has been hypothesized, as infection in utero in embryonic rodents induces craniofacial dysmorphisms in successfully infected animals [24
]. First-in-man applications of this wild-type rodent parvovirus in two adolescent osteosarcoma patients in the 1960s were prompted by the discovery that has oncolytic potential [25
]. In this compassionate use study, no severe virus-induced toxicity was reported. Thereafter, more than four decades elapsed before the first phase I virotherapy trials with other oncolytic viruses started to recruit osteosarcoma patients [27
]. To prepare for a clinical trial of H-1PV in osteosarcoma patients, systematic preclinical research is needed.
Protoparovirus H-1PV consists of an approximately 5.1-kb single-stranded DNA genome enclosed in a 25-nm non-enveloped icosahedral shell. The viral genome contains two main transcriptional units: one encoding the non-structural (NS) proteins and one encoding the viral capsid proteins (VP) and the small alternatively translated (SAT) [27
]. The major large non-structural protein NS1 is essential to virus replication and cytotoxicity in permissive cells [28
In H-1PV-infected animals and humans, the immune system produces virus-specific antibodies 5–7 days after infection, and these at least partially neutralize the virus [25
]. Additionally, H-1PV triggers anticancer vaccination effects, whereby animals cured of cancer by the oncolytic virus remain immune to the same malignant disease even in the absence of the virus [29
In recent years, H-1PV fitness mutants have been isolated and tested for antineoplastic efficacy [30
]. It has been hypothesized that the greater egress of infective viral progeny viruses and the higher infectivity observed in vitro with these mutants may lead to higher antineoplastic efficacy in vivo. This has been confirmed for Del H-1PV in pancreatic cancer and cervix carcinoma xenograft models [30
Here, as a first step in assessing the responsiveness of osteosarcomas to parvovirus treatment, preclinical testing of wild-type H-1PV and of two derived mutants, Del H-1PV, and DM H-1PV, has been performed in vitro.
2. Materials and Methods
2.1. Ethics Statement
Human neonatal foreskin fibroblasts were supplied by CET celleng-tech (Coralville, IA, USA). Primary human osteoblasts were obtained from PromoCell GmbH (Heidelberg, Germany). All other osteosarcoma cell lines (CAL 72, H-OS, MG-63, SaOS-2, and U2-OS) were obtained from Cell Line Service GmbH (Eppelheim, Germany). The WAC-2 clone, derived from the neuroblastoma cell line SH-EP by stable transfection, contains an ectopic Cytomegalovirus N-myc proto-oncogene (CMV-MYCN)
enhanced expression cassette [31
] was kindly provided by Prof. Dr. med. O. Witt, Clinical Cooperation Unit Pediatric Oncology, German Cancer Research Center (Heidelberg, Germany). For data confirmation, a second batch was obtained from Prof. A. Schramm, Department of Pediatric Hematology and Oncology, University Hospital Essen (Essen, Germany).
2.2. Mammalian Cell Culture
All cell cultures were maintained in 5% CO2
at 37°C and 100% relative humidity. Human neonatal foreskin fibroblast cells were propagated in Human Foreskin Fibroblast Expansion Medium (Cellular Engineering Technologies, Coralville, IA, USA) containing 10% fetal calf serum (FCS), 100 U/mL penicillin, and 100 μg/mL streptomycin. Non-transformed human osteoblasts were grown in Osteoblast Growth Medium (PromoCell GmbH, Heidelberg, Germany). The culture medium for osteosarcoma cell lines was Dulbecco’s Modified Eagle’s Medium (DMEM) or Minimum Essential Medium (MEM) for H-OS cells, supplemented with 2 mM L-glutamine, 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin (final concentrations). The human neuroblastoma cell line WAC-2 was cultured in Roswell Park Memorial Institute (RPMI-1640) medium containing 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin. For passaging, cells were detached in 0.05% or 0.25% Trypsin-EDTA solution and then resuspended in fresh culture medium. All cell lines and non-transformed cell cultures were routinely checked for contamination [32
] and genomic identity [33
], using previously established methods. Osteosarcoma cell lines used in this study are listed in Table S1
2.3. Viruses and Virus Production
Wild-type H-1 parvovirus (H-1PV) and the recombinant H-1 parvovirus (Chi-hH-1/EGFP) expressing enhanced green fluorescent protein (EGFP) were produced at the Virus Production & Development Unit, Division of Tumor Virology, German Cancer Research Center, Germany. The recombinant parvovirus Chi-hH-1/EGFP was obtained by co-transfecting HEK-293T cells with the corresponding recombinant vector DNA and a helper plasmid expressing the viral capsid genes in trans [34
]. It was purified in the same manner as the wild-type H-1PV. H-1PV was produced by infecting human newborn embryonic kidney NBK-324K cells at a multiplicity of infection (MOI) of 10−2
plaque-forming units per cell (PFU/cell). Four to five days after infection, the crude virus was extracted from infected cells and purified by filtration (pore diameter: 0.2 μm) and by iodixanol gradient centrifugation as previously described [35
]. Contamination of virus stocks with endotoxins was below 2.5 U/mL. The Del H-1PV mutant was produced as previously described [30
2.4. Detection of Infectious H-1PV Particles
Viral titers were determined by means of infected cell hybridization assays or by plaque assay as previously described [36
]. Titration experiments were carried out in triplicates. For the hybridization assay, NB-324K cells (7.6 × 103
cells/well) were seeded into 96-well plates. The cells, 24 h after seeding, were infected with 10-fold serial dilutions of the virus sample and incubated for 72 h under 5% CO2
, at 37 °C and 100% relative humidity. Next, the cells were lysed with 0.75 M NaOH. The DNA was transferred to a nylon membrane, UV-cross-linked, hybridized with a 32
P-labeled NS-specific DNA fragment, and used to expose X-ray films for autoradiography.
2.5. Western Blot Analysis
Western blotting was performed as described [37
]. The cells, 12 h after seeding, were either mock-infected or exposed to wild-type H-1PV (MOI: 1 PFU/cell). Osteosarcoma cells were harvested at 24-h intervals over a 5-day period post-infection. Briefly, approximately 106
mock- or H-1PV-infected cells were harvested at the time points indicated, collected by centrifugation, and washed with phosphate-buffered saline (PBS). Cell pellets were kept on ice for 1 h in RIPA lysis buffer with freshly added protease inhibitor (complete Mini, EDTA-free, Roche Diagnostics, Indianapolis, IN, USA) and then centrifuged at 15,000× g
for 10 min at 4 °C. The supernatants were stored at −80 °C until further analysis. Protein concentrations of the cell lysates were determined photometrically with the PierceTM
BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer’s protocol. Twenty micrograms of each cell lysate was resolved by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a nitrocellulose membrane (Schleicher & Schüll, Kassel, Germany).
The following antibodies were used: monoclonal mouse anti-actin clone C4 (MP Biomedicals, Illkirch, France), rabbit polyclonal antiserum MK3 raised against the viral NS1 protein, kindly provided by N. Salomé, INSERM U1109, Strasbourg, France [36
]. Binding of antibodies, detection with horseradish peroxidase-conjugated anti-rabbit or anti-mouse IgGs (dilution 1:5000), and chemiluminescence assays were performed as previously described [38
Comparative fluorescence and phase contrast images were recorded under an Olympus CKX41 inverted phase contrast microscope (Olympus Corporation, Tokyo, Japan) using the Cell B software from Olympus. Phase contrast images were captured with the Keyence BZ 9000 microscope (KEYENCE Microscope Europe NV/SA, Mechelen, Belgium) and the imaging softwares BZ II Viewer and BZ II Analyzer supplied with it by the manufacturer.
2.7. Viral DNA Extraction and Quantitative Real-Time Polymerase Chain Reaction (qPCR)
One-tenth of the total volume, i.e., 1 mL, of the culture medium of adherently growing cells was collected from each cell culture dish at regular intervals post-infection before harvesting the cells for protein extraction and subsequent western blot analysis (see Section 2.5
). This supernatant was subjected to alkaline lysis in 1 M NaOH in TE buffer for 30 min at 56 °C.
After neutralization, samples were diluted 1:100 in sterile water. Quantification of viral DNA in these solutions was carried out by real-time qPCR with an NS1-specific TaqManTM
probe (from Applied Biosystems, Thermo Fisher Scientific), according to previously published procedures [31
2.8. Cell Viability and Cell Death Assessment
Proliferation of bone tumor cells was tested in a 96-well plate format as published, using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) as a test reagent (from Sigma-Aldrich®
, Inc., St. Louis, MO, USA) [39
]. To assess the viability of osteosarcoma cell lines, cells were seeded at 2,500 cells per well and the assays were performed on days 3 and 6 post-infection. For the MTT tests pertaining to the positive control cell line (WAC-2 neuroblastoma cells), 1000 cells per well were seeded and MTT-tests were performed on days 3 and 6 post-infection. To determine the viability of primary osteoblasts and fibroblasts on day 7 after infection, cells were seeded into 96-well plates at 2500 cells per well. 12 h after cell seeding, the medium was removed and virus inoculum or buffer was added to the cells in 50 μL serum-free medium at the indicated MOI. Two hours after infection, 50 μL culture medium supplemented with 20% FCS was added in order to achieve culture conditions appropriate for the periods mentioned above. At the end of the infection period, cells were incubated for 1 h with 0.5 μg/mL MTT solution. After removing the solution, the cells were allowed to dry and isopropanol was added at 100 μL per well. Absorbance values were photometrically determined at 570 nm with a Multiscan PlusTM
Microplate Reader (from Titertek Instruments, Huntsville, AL, USA).
Cell lysis was determined by the amount of lactate dehydrogenase (LDH) released into the culture medium. For this, the Cytotox 96 cytotoxicity assay kitTM (from Promega, Mannheim, Germany) was used according to the manufacturer’s instructions. The absorbance at 490 nm of red formazan generated by an LDH-catalyzed reaction was measured with the Multiscan PlusTM Microplate Reader. Both cell viability tests and cell lysis assays were carried out in five replicates. The median lethal dose (LD50) of input virus was defined as the MOI at which cell viablitiy was reduced by at least 50% (as determined by the MTT test) and cell lysis reached at least 50% (as determined in the LDH release assay).
2.9. Flow Cytometric Characterization of the Cell-Cycle Distribution of Cells and of the Sub-G1 Apoptotic Cell Population
Osteosarcoma cells or WAC-2 neuroblastoma cells were seeded in triplicate (respectively, at 2 × 105 or 1 × 106 cells per 10-cm-diameter dish), infected with H-1PV (1 PFU/cell), and cultured for up to 120 h. Control (mock-infected) cells were harvested at the corresponding time points. Cells were washed twice with PBS and then fixed with 0.7 mL ice-cold 100% ethanol and 0.2 mL PBS. Fixed cells were stored at −20 °C. For analysis, the cells were pelleted for 10 min at 400× g and 4 °C and washed twice with PBS. Cell pellets were resuspended in PBS containing 100 mg/mL RNase H and 5 μg/mL propidium iodide (Sigma-Aldrich Inc., St. Louis, MO, USA). The stained cells were filtered through a 41-μm nylon mesh, incubated on ice for 1 h in the dark, and their DNA content was measured on a FACSort flow cytometer (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). A minimum of 10,000 events were recorded and analyzed with the Cell-QuestTM software (from Becton, Dickinson, NJ, USA). Differences in cell cycle distribution were tested for statistical significance by a two-sided Student’s t-test.