Defining a murine ovarian cancer model for the evaluation of conditionally-replicative adenovirus (CRAd) virotherapy agents

The human ovarian carcinoma cell line SKOV3.ip1 was obtained from Janet Price (M. D. Anderson Cancer Center, Houston, Tex.). The mouse epithelial ovarian cancer cell line ID8, which was originally established by Dr. Kathy Roby (Kansas University Medical Center) and its derivative expressing firefly luciferase were obtained from Dr. Robin Bachelder (Duke University Medical Center) and were maintained in DMEM (high glucose, Gibco-Life Technologies) containing 4% FBS, 1× penicillin and streptomycin as described elsewhere [33]. The ID8-Trp53^−/− [34] mouse ovarian surface epithelial cells (F3) were kindly provided by Dr. Iain McNeish (University of Glasglow) and labeled to express mCherryLuciferase (F3mCherryLuc). Both cell lines were grown in DMEM (high glucose, Gibco-Life Technologies) containing 4% FBS, 1× penicillin and streptomycin and 1× ITS (Insulin-Transferrin-Selenium solution) (Gibco-Life Technologies). The 911 human embryonic retinoblasts derived by transformation with a plasmid containing 79–5789 bp of the Ad5 genome [35] were obtained through Crucell Holland B.V. (Leiden, The Netherlands). The human lung carcinoma cell line A549 were obtained from American Cell Type Culture Collection (ATCC, Manassas, Virginia USA). All cell lines were grown at 37 °C in medium recommended by the suppliers in a humidified atmosphere of 5% CO₂.

The replication-competent CRAd Delta 24 was provided by Dr. J. Fueyo. (The University of Texas M. D. Anderson Cancer Center, Houston, TX). This virus contains a 24-nucleotide deletion, from Ad5 bp 923 to 946 (both included), corresponding to the amino acid sequence L₁₂₂TCHEAGF₁₂₉ of the E1A protein known to be necessary for Rb protein binding [36]. Details of the tumor-specific replication of this virus are presented elsewhere [37, 38]. The incorporation of the RGD-4C motif, known to interact with α_v integrins, into the HI loop of the fiber knob (T₅₄₆CDCRGDCFCP₅₄₇) to enhance Delta24 CRAd infection efficiency of tumor cells was described previously [15, 39]. The construction of Ad5/3-Delta24 CRAd, which contains the fiber knob domain replaced with its counterpart from Ad serotype 3 (Ad3) was described elsewhere [9, 40]. The construction of Ad5/3-Delta24-based CRAd-IL24 and CRAd-ING4 vectors expressing human IL-24 [41] or ING4 (the inhibitor of growth 4) [42] gene, respectively, under transcriptional control of human cytomegalovirus (CMV) immediate-early promoter/enhancer incorporated in place of the deleted E3B region was described in detail recently [43]. Non-armed control CRAd that encodes the secreted Gaussia luciferase (Gluc) from the copepod Gaussia princeps (New England BioLabs Inc., Ipswich, MA USA) driven by CMV promoter in place of E3B region was described recently [43]. Wild-type Ad5 was kindly provided by Dr. H Ugai (Washington University in St Louis, St Louis, MO).The replication incompetent Ad5∆E1 containing the CMV promoter-driven firefly luciferase reporter gene in place of the deleted E1A/B genes was described before [44] and propagated using 911 cells. All CRAd vectors and wild type Ad5 were propagated using A549 cells, purified by centrifugation on CsCl gradients according to standard protocol, and dialyzed against phosphate-buffered saline (PBS) (8 mM Na₂HPO₄, 2 mM KH₂PO₄ [pH 7.4], 137 mM NaCl, 2.7 mM KCl] containing 10% glycerol. The titers of physical viral particles (vp) were determined by the methods of Maizel et al. [45]. The titers of infectious viral particles were determined by plaque assay using 911 cells as described by Mittereder et al. [46].

To monitor cytotoxic effects induced by Ad5, Delta-24, or Delta-24-RGD vector the cell monolayers grown in 96-well plates (3 × 10³ to 5 × 10³ cells/well) were infected in triplicates with each virus at MOI of 100 vp/cell. The infected and mock-infected cells were subjected to CellTox™ Green Cytotoxicity assay as recommended by the manufacturer (Promega Corporation, Madison, WI) by adding DNA-binding cyanine dye on day 1 and monitoring the increase in fluorescent signal intensity, which is proportional to cell lysis, till day 5–6 post-infection. The degree of virus-mediated cell killing was measured using the Synergy-HT plate reader (Bio-Tek Instruments, Winooski, VT) equipped with 485 nm excitation and 520 nm emission wavelength filters and the average values of relative fluorescent units (RFU) are shown. The relative cell viability on day 5–6 post-infection with either Ad5, Delta-24, Delta-24-RGD or non-replicating Ad5∆E1 control was determined using the Cell Proliferation Assay (Promega Corporation, Madison, WI) as recommended by the manufacturer. Assay was performed by adding 10 μL CellTiter 96 AQ_ueous One Solution Reagent directly to culture wells containing red phenol free media supplemented with 2% FBS, incubating for 2 h and then recording the absorbance at 490 nm with a plate reader (Synergy HT, Bio-Tek Instruments, Winooski, VT). The data are presented as the percentages of viable cells in monolayers infected with each viral dose that were determined with respect to the uninfected control set as 100%.

Female 5–7 week-old athymic nu/nu mice (The Jackson Laboratory) were kept under pathogen-free conditions according to the American Association for Accreditation of Laboratory Animal Care guidelines. Animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee of the Washington University in Saint Louis, School of Medicine. Five million SKOV3.ip1 cells were xenografted s.c. into the right flank of the mice under anesthesia. When the nodules reached a volume of 60–70 mm³, a single Ad dose (10¹⁰ vp in 20 μl of PBS) or the same volume of PBS was administered intratumorally (n = 9–10 animals/group) and injections were repeated weekly for 3 consecutive weeks. Tumor size was monitored twice a week, and fractional volume was calculated by standard technique for volume determination of subcutaneously xenografted tumors in vivo using external caliper measurements and the modified ellipsoid formula 1/2 (Length × Width²) [47, 48]. Animal were euthanized according to IACUC policy “Maintaining Tumors in Rodents Policy” if the tumor became greater than 2 cm in any dimension or showed clinically significant cutaneous ulceration.

Female 5–7 week-old C57Bl/6 mice (The Jackson Laboratories) were injected i.p. with 5 × 10⁶ F3mCherryLuc cells in 200 μL PBS. After one week, mice were injected i.p. with 1 × 10¹⁰ vp of wild-type Ad or replication-incompetent Ad vectors in 100 μL PBS (n = 9). Mice were euthanized after 17 days of tumor inoculation. In vivo bioluminescence imaging was performed on days 7 and 17 after tumor inoculation on an IVIS Lumina (PerkinElmer, Waltham, MA; Living Image 3.2, 1 s exposure(s), 8 bin. Mice were injected intraperitoneally with D-luciferin (150 mg/kg in PBS; Gold Biotechnology, St. Louis, MO) and imaged using isoflurane anesthesia (2% vaporized in O₂). Total photon flux (photons/sec) was measured from fixed regions of interest (ROIs) over the entire mouse abdomen using Living Image 2.6. Animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee of the Washington University in Saint Louis, School of Medicine.

All data are presented as the mean ± SD. The Student’s two-tailed t-test was used to determine statistical significance at the 95% confidence level, with p ≤ 0.05 being considered significantly different for in vitro data. The animal survival data from the tumor xenograft model were analyzed using product-limit method to estimate the survival function S(t) for each treatment and control group with IBM-SPSS statistics software. The long rank test with family-wise significance level α = 0.05 was employed to compare animal survival between control group treated with non-armed CRAd and each armed CRAd treatment group. Mann-Whitney test was used to determine statistical significance in the syngeneic orthotopic model groups with non-parametric distribution, with p ≤ 0.05 being considered significantly different for the syngeneic orthotopic model.

We have advanced the development of advanced generation CRAd agents which embody the capacity to accomplish CAR-independent infection of tumor cell targets for enhanced oncolytic potency. Infectivity enhancement was initially embodied into CRAd design via incorporation of the integrin-binding peptide RGD-4C in the HI loop of the fiber knob [49, 50]. We next pursued a strategy of chimerism for the fiber knob by replacing the CAR-binding adenovirus serotype 3 fiber knob with the corresponding fiber knob domain of the human adenovirus serotype 3 [9, 40]. Our studies confirmed that this latter strategy also allowed significant enhancement of oncolytic potency in murine orthotopic xenograft models of ovarian cancer virotherapy. On the basis of these findings, we sought to advance the design of our knob 3 infectivity-enhanced CRAd via an “arming” strategy. Specifically, we sought to configure within the adenovirus genome the immunoregulatory molecules ING and IL-24.

On this basis we constructed Ad5/3-Delta24 CRAds armed with either ING4 or IL-24 [43]. These earlier studies confirmed that the armed CRAds expressed the encoded arming genes at a high level. Importantly, modification of the Ad5/3-Delta24 CRAds in this manner was not deleterious to their anti-tumor oncolytic properties. We thus evaluated these armed CRAds in an established murine orthotopic xenograft model of human cancer of the ovary which employed the human ovarian cancer cell line SKOV3.ip1. Analysis of treated animals was endeavored for tumor growth and survival. As can be seen, the armed versions of Ad5/3-Delta24 CRAd exhibited less anti-tumor activity than the parental agent. Importantly, this observation was noted with both assays of tumor growth (Fig. 1a) and survival (Fig. 1b).

The findings noted in our study of the armed Ad5/3-Delta24 CRAds in the murine orthotopic xenograft model highlighted the limits of immunodeficient models for analysis of the immunobiologic activities of adenovirus-based virotherapy agents. To address this limit, we considered the potential utility of the murine immunocompetent syngeneic ID8 model of human carcinoma of the ovary. Of note, an expanding repertoire of reports have recently highlighted the utility of this model for study of ovarian cancer immunobiology. More importantly, this model has demonstrated utility for analysis of immunotherapy interventions for cancer of the ovary [51].

As noted, a number of recent reports have reported the utility of immunocompetent murine models to study the ability of CRAds to accomplish effective anti-tumor immunization [29, 30]. Whereas not fully permissive for productive human adenovirus replication, selected murine tumor cells targets appear to allow a level of human adenovirus replication of sufficient magnitude to activate anti-tumor immunity in the context of implantation in corresponding immunocompetent syngeneic murine models. On this basis, we studied the cytolysis in the murine cell lines ID8 and ID8luc, which were employed widely as a murine immunocompetent model of ovarian cancer [51].

We also employed in these studies the human ovarian cancer cell line SKOV3.ip1, with the wild type adenovirus Ad5 and the armed CRAd agents. These adenoviruses achieved prominent cytotoxic effects in this replication permissive tumor cancer cell line, as predicted (Fig. 2). The specificity of this effect was confirmed by the lack of toxicity induced by the replication-incompetent control adenovirus. Of note, similar effects were noted with analysis of the murine ovarian cancer cell lines ID8 and ID8luc. Indeed, the levels of induced tumor cell cytotoxicity were nearly comparable in the murine targets as noted in the human tumor targets.

Of note, it has recently been shown that knockout derivatives of ID8 exhibited significant alterations of the tumor microenvironment and reduced survival, with an overall closer resemblance with the human disease. Although the p53-knockout ID8 derivative cell line, F3, exhibited a lower virus-induced cytotoxicity and replication compared to a highly susceptible human cancer line, A549 (Fig. 3), the levels were comparable to those obtained with GL261, a mouse murine cell line that has shown utility as a syngeneic model vis-à-vis induction of anti-tumor immunization.

Finally, we evaluated the of fully replication competent wild type adenovirus effect in a syngeneic murine model generated by the administration of F3mCherryLuc. Analysis of treated animals after 7 days of Ad5 injection showed a significant reduction on luminescence signal (Fig. 4). The ID8 line, and its derivative, thus exhibited: (1) a level of permissivity to adenovirus-mediated cytotoxicity in vitro comparable to GL261, a murine model useful for study of CRAd-induced anti-tumor immunization, and (2) susceptibility to adenovirus-mediated oncolysis in vivo. This initial study thus supports the employment of the F3 syngeneic murine model as a potentially robust system for the study of the therapeutic effect of CRAds for cancer of the ovary.