Synthetic virology approaches to improve the safety and efficacy of oncolytic virus therapies

Azad, Taha; Rezaei, Reza; Singaravelu, Ragunath; Pelin, Adrian; Boulton, Stephen; Petryk, Julia; Onsu, Kemal Alper; Martin, Nikolas T.; Hoskin, Victoria; Ghahremani, Mina; Marotel, Marie; Marius, Ricardo; He, Xiaohong; Crupi, Mathieu J. F.; Hoang, Huy-Dung; Nik-Akhtar, Abolfazl; Ahmadi, Mahsa; Zamani, Nika Kooshki; Golshani, Ashkan; Alain, Tommy; Greer, Peter; Ardolino, Michele; Dickinson, Bryan C.; Tai, Lee-Hwa; Ilkow, Carolina S.; Bell, John C.

doi:10.1038/s41467-023-38651-x

Download PDF

Article
Open access
Published: 26 May 2023

Synthetic virology approaches to improve the safety and efficacy of oncolytic virus therapies

Nature Communications volume 14, Article number: 3035 (2023) Cite this article

7390 Accesses
9 Citations
24 Altmetric
Metrics details

Subjects

Abstract

The large coding potential of vaccinia virus (VV) vectors is a defining feature. However, limited regulatory switches are available to control viral replication as well as timing and dosing of transgene expression in order to facilitate safe and efficacious payload delivery. Herein, we adapt drug-controlled gene switches to enable control of virally encoded transgene expression, including systems controlled by the FDA-approved rapamycin and doxycycline. Using ribosome profiling to characterize viral promoter strength, we rationally design fusions of the operator element of different drug-inducible systems with VV promoters to produce synthetic promoters yielding robust inducible expression with undetectable baseline levels. We also generate chimeric synthetic promoters facilitating additional regulatory layers for VV-encoded synthetic transgene networks. The switches are applied to enable inducible expression of fusogenic proteins, dose-controlled delivery of toxic cytokines, and chemical regulation of VV replication. This toolbox enables the precise modulation of transgene circuitry in VV-vectored oncolytic virus design.

Chemogenetic ON and OFF switches for RNA virus replication

Article Open access 01 March 2021

A simeprevir-inducible molecular switch for the control of cell and gene therapies

Article Open access 27 November 2023

Unconventional viral gene expression mechanisms as therapeutic targets

Article 19 May 2021

Introduction

Tumors are heterogeneous cellular ecosystems that continuously evolve in the face of therapeutic intervention limiting the efficacy of conventional treatments for metastatic disease. Replicating cancer therapeutics like engineered bacteria¹, immune cells² and viruses³ comprise an emerging class of modalities that, in principle, can work in harmony with a patient’s immune system to initiate robust, systemic anti-tumor immune responses capable of keeping pace with and attacking evolving tumor ecosystems. These forms of therapies can be thought of as biological machines that to be optimally deployed, will require control strategies coordinating and regulating their activity within the patient^4,5. Oncolytic viruses (OVs) have the potential to be replicating cancer gene therapy vectors that reproduce in and destroy tumor cells (oncolysis) and in addition have the capacity to dispense therapeutic payloads to remodel the tumor microenvironment³ by impacting tumor vasculature⁶ and reshaping extracellular matrix components⁷. However, this remodeling process, especially the generation of robust adaptive and innate immune responses, can sometimes be at odds with optimal virus spread within and between tumors. Incorporation of genetic circuits into oncolytic virus backbones that regulate the timing/magnitude of expression of virulence genes and therapeutic payloads could facilitate safer, more effective therapeutics. In this study, we used synthetic virology approaches to develop a new generation of OVs equipped with multiple inducible systems, which can be used as “safety switches” for providing user defined control of both virus replication and transgene expression. To facilitate clinical translation of our products we tested orally available FDA-approved small-molecule responsive transcription systems. We show through combinatorial application of discrete regulatory elements we can increase the safety and efficacy of an oncolytic vaccinia virus in xenograft and syngeneic mouse tumor models. These systems will pave the way for developing a generation of OVs in which the timing and level of both OV replication and their therapeutic payloads can be regulated.

Results

Generation of a rapamycin-inducible expression system in an oncolytic vaccinia virus vector

We designed a series of chemogenetic switches to enable temporal control of the expression of multiple therapeutic payloads. Vaccinia virus (VV) is a cytoplasmic virus and while is able to infect and replicate in mammalian cells, its transcriptional machinery and genomic structure is not optimized for the use of mammalian promoters. Vaccinia virus has its own set of regulatory sequences and transcription factors that are adapted to its specific needs, and the use of mammalian promoters is not compatible with these requirements (Supplementary Fig. 1a). Previous studies have demonstrated that the VV-encoded T7 polymerase can drive the expression of gene cassettes under the control of a T7 promoter element and so we incorporated a recently developed small-molecule activated, proximity-dependent split T7 RNA polymerase (RNAP)⁸. This system consists of T7 RNA polymerase divided into two domains (N term-1-29 and C term-30-181 of T7 RNAP), that assemble into a functional polymerase in a proximity-dependent fashion. The T7 components were fused to a rapamycin-induced dimerization system, consisting of FKBP-rapamycin binding domain (FRB) and FK506 binding protein (FKBP). Upon introduction of rapamycin, FRB is recruited to FKBP, which brings the C-terminus in close proximity to N-terminal T7 RNAP. T7 RNAP is then assembled and able to transcribe a gene regulated by a T7 promoter (Fig. 1a). We engineered VV to constitutively express mCherry and the rapamycin-controlled, split T7 RNA polymerase-based transcription system (ST7) in the Thymidine Kinase (TK) coding region. In the same TK locus, we also engineered a T7 promoter-driven firefly luciferase and GFP (GFPLuc) cassette (VV-ST7-iGFPLuc; Fig. 1a). In VV-ST7-iGFPLuc infected U2OS osteosarcoma cells, GFP and luciferase expression were induced dramatically in the presence of 10 nM rapamycin over a range of different virus concentrations (Fig. 1b, c). Infected U2OS cells showed a greater than tenfold increase in luciferase activity 24 h post infection relative to uninduced infected controls (Fig. 1c). Similar results were observed in HeLa cervical carcinoma cells, HT29 colorectal adenocarcinoma cell line, and A549 lung carcinoma cells (Fig. 1d). In order to confirm that rapamycin and the inducible split T7 polymerase system did not impact viral growth kinetics, we performed multistep virus growth curves. Rapamycin and ST7 expression did not have a significant effect on VV-ST7-iGFPLuc growth in U2OS, HeLa, or A549 cells (Fig. 1e; Supplementary Fig. 1b, c).

**Fig. 1: Integration, functional characterization, and optimization of the split T7 inducible system in vaccinia virus.**

Rapamycin (sirolimus) is approved for use as an immunosuppressant and treatment of lymphangioleiomyomatosis, a rare lung disease⁹. Its actions are mediated through inhibition of the mTOR protein kinase and associated with anti-proliferative and anti-cancer activity. Rapamycin suffers from poor solubility, stability and pharmacokinetic properties. However, rapamycin analogs, or “rapalogs”, (e.g. everolimus, temsirolimus, and ridaforolimus) have been approved for the treatment of a wide spectrum of cancers^10,11. We examined if these approved rapalogs were compatible with the VV-adapted ST7 expression system. In infected U2OS cells, we observed a robust increase in GFP and luciferase expression during treatment with rapalogs, in line with our observations with rapamycin (Fig. 1f–g). We also demonstrated a dose-dependent increase of luciferase activity in infected U2OS cells treated with rapalogs, suggesting the system is tunable (Fig. 1g). Similar to rapamycin, rapalogs did not significantly influence viral growth (Fig. 1h). To examine whether rapamycin and rapalogs can be used in vivo to control the virally encoded transgenic RNA polymerase system, we injected HT-29 xenograft bearing mice with VV-ST7-iGFPLuc intratumorally and treated mice with rapamycin or rapalogs via different routes of administration. Temsirolimus¹² is more water soluble and suited for intravenous injection, whereas everolimus¹³ is available for oral administration. In this study, we infected tumors and applied a single dose of the rapalog at the time of infection using the clinically appropriate route for each rapalog. As shown in Fig. 1i-j, we observed robust stimulation of luciferase 24 h later. The animals received no further treatments with the rapalogs and then at 72 h post initial infection the tumors were once again imaged to determine the level of luciferase expression. At this time, the level of luciferase expression driven by the initial induction with the rapalog decreased significantly (Fig. 1i, j), This decrease in signal is not related to a clearance of virus as we determined the level of infectious particles at day 10 and it was comparable in all of the tumors sampled (Supplementary Fig. 1d). Furthermore, in animals that were infected but not treated with rapalog there was minimal luciferase signal which was robustly induced upon rapalog induction. Among the conditions tested, rapamycin (i.p.) and temsirolimus (i.v.) produced the highest induction of luciferase signal in the tumors. These results demonstrate that the rapamycin-inducible ST7 system can be used to temporally control viral transgene expression in vivo.

Control of vaccinia virus promoter elements with doxycycline

We investigated the utility of repressor/operator systems to create virally encoded chemogenetic switches. Doxycycline (Dox) is an approved orally delivered tetracycline-class antibiotic with well-established pharmacokinetic and pharmacodynamic properties. The Dox-inducible expression system has previously been applied in the context of a VV genome to enable conditional expression of viral factors¹⁴. We carried out a more detailed analysis of this technology by applying it to a variety of vaccinia promoter elements. Vaccinia genes have stage-specific transcription factors and corresponding promoter elements which mediate temporal regulation of their expression. We screened a diverse set of VV promoters encompassing the range of temporal expression patterns exhibited by VV proteins. Using ribosome profiling in VV infected HeLa cells, we identified viral promoters associated with genes encoding transcripts that are efficiently translated, and thus highly expressed, at different times during the virus replication cycle (Fig. 2a). We generated a series of recombinant VV strains encoding tet operator controlled VV candidate promoters to conditionally control the expression of the GFPLuc cassette (Fig. 2b). In the same locus (TK), the strains encoded blue fluorescence protein (BFP) and the tetracycline repressor protein (TetR), which binds TetO in the absence of Dox to impair expression. We performed an arrayed screen of these viruses to identify which TetO-controlled VV promoter resulted in the highest level of induced GFP and luciferase expression (Fig. 2c, d). There was considerable variability in the ability to control vaccinia virus promoters using the Dox-inducible gene cassettes. We found that the p11 and PEL promoters were most amenable to this form of transcriptional control with robust induction in the presence of Dox and minimal background expression in its absence. A closer examination of the luciferase assay results (Fig. 2d) revealed that the use of p11 promoter resulted in maximal reporter gene expression in the presence of Dox; therefore, we selected the TetO-controlled P11 promoter for further characterization studies. We first tested the kinetics of inducible expression with the newly developed VV strain encoding TetR and p11-TetO driven GFPLuc gene cassettes (VV-TetR-iGFPLuc). We observed induction of a GFP signal in infected U2OS cells within 3 h of Dox treatment (Supplementary Fig. 2a) and similar high inducibility (>30-fold) in a panel of other cancer cell lines (A549, HeLa, HT-29, and SKOV3) (Fig. 2e and Supplementary Fig. 2b). We conducted a comparison of luciferase expression at various time points using the wild-type vaccinia virus promoter and the doxycycline-controlled promoter with and without induction and demonstrated that both promoter systems achieve a similar maximum level of reporter expression (Supplementary Fig. 2c). Neither expression of TetR, incorporation of the TetO-controlled expression system, or Dox influenced viral growth – illustrating compatibility of the Dox-controlled chemogenetic switch in VV-based vectors (Fig. 2f and Supplementary Fig. 2d, e). Under in vitro conditions, 10 ng/mL of Dox was sufficient to induce detectable levels of luciferase activity in VV-TetR-iGFPLuc infected U2OS cells, while 25 ng/mL produced the maximal signal (Supplementary Fig. 2f).

**Fig. 2: Optimization of the doxycycline-inducible system for controlling gene expression in vaccinia virus.**

We then characterized VV-TetR-iGFPLuc in vivo using a HT-29 colon cancer xenograft model. We infected the mice intratumorally with VV-TetR-iGFPLuc and performed 1 mg/kg intraperitoneal treatments. In vivo imaging of bioluminescence, 24 h post infection, demonstrated robust luciferase activity in the tumors (Supplementary Fig. 2g). We repeated the experiment with one alteration—introduction of Dox via diet to simulate the potential use of oral Dox in combination with a VV vector bearing Dox-controlled transgene expression (Fig. 2g, h, Supplementary Fig. 2h). We were able to conditionally control transgene expression through feeding and withdrawal of a Dox-containing diet (Fig. 2h; Supplementary Fig. 2g). We noted that Dox had no influence on viral growth in vivo, in line with our in vitro results (Fig. 2g), and no negative effect on luciferase activity in vivo (Supplementary Fig. 2i, j). Collectively, these results establish the utility of a Dox-controlled gene cassette as a tool for temporal control of payload expression in VV vectors in vivo.

A cumate-controlled chemogenetic switch for vaccinia virus

We sought to identify a repressor-based operator system to generate another VV-adapted inducible expression system to further enable independent temporal control of multiple payloads. The cumate-controlled operator system derived from the Pseudomonas putida p-cmt and p-cym operons^15,16, has been used in gene therapy vectors^17,18 and is similar to the tetracycline-controlled operator system. It consists of the cumate operator (CuO) sequence which is bound by a repressor (CymR) in the absence of cumate. We engineered VV to constitutively express CymR and BFP. In the same locus (TK), we incorporated the coding sequence of GFPLuc under the control of the cumate operon fused to a VV promoter (Fig. 3a). Similar to the development of the Dox-based chemogenetic switch, we screened VV promoters to identify which worked best in combination with the CuO to mediate cumate-inducible expression in infected U2OS cells (Fig. 3b, c). As with the Dox system, there was considerable variability in the control of vaccinia promoters by the cumate operon. While the early-late promoters H5R, LEO, and LEO160 resulted in maximal luciferase activity and GFP expression in the presence of cumate, these promoters also resulted in the highest basal reporter signal in the absence of cumate. The use of the synthetic early-late promoter PEL in tandem with CuO resulted in the highest inducibility (450X) while maintaining minimal basal expression (Fig. 3c); therefore, we selected the PEL-Cu promoter/operator combination for further evaluation.

**Fig. 3: Cumate-inducible system controls viral gene expression in vitro and in vivo.**

We examined the kinetics of inducible expression with our VV strain encoding CymR and PEL-CuO driven GFPLuc gene cassettes (VV-CymR-iGFPLuc). With cumate concentrations as low as 10 µg/mL, GFP signal is observed within 6 h of cumate treatment of VV-CymR-iGFPLuc infected cells (Supplementary Fig. 3a). Consistent with this, significant luciferase activity is also observed in infected cells with 10 µg/mL cumate (Supplementary Fig. 3d). We confirmed that cumate treatment does not influence viral growth in U2OS cells at concentrations up to 500 µg/mL (Supplementary Fig. 3b, c). Similarly, multistep viral growth studies confirmed that cumate or CymR expression does not impact VV replication in U2OS, HeLa, and A549 cells (Supplementary Fig. 3e–g). We next compared head-to-head the cumate-inducible expression system with the ST7 and Dox-inducible systems (Supplementary Fig. 3h–k). In infected U2OS cells, the Dox-inducible system produced the highest reporter signal in the presence of its inducer; however, the cumate-inducible system provided the greatest inducibility in the presence of its ligand while maintaining minimal basal expression in the absence of cumate (Supplementary Fig. 3h).

HT-29 xenografts were injected intratumorally with VV-CymR-iGFPLuc and cumate (1 mg/kg) was administered intraperitoneally. In vivo imaging of bioluminescence 24 h post infection revealed robust tumor-localized reporter signal in a cumate-dependent fashion (Supplementary Fig. 3l). In parallel, we demonstrated that cumate treatment in vivo did not influence VV replication (Supplementary Fig. 3l, lower panels). To examine the potential for more accessible administration of cumate, we repeated the experiment and delivered cumate (6000 mg/kg) to mice via their diet. We observed cumate-dependent activation of luciferase activity in the tumor, and, confirmed that cumate does not influence VV replication (Supplementary Fig. 3m, n). As this is the first report, to the best of our knowledge, of feeding mice cumate, we examined the effect of feeding cumate on mouse health and viral growth. At all concentrations tested (100–6000 mg/kg), cumate did not impact viral replication in vivo (Fig. 3d). Treatment for more than twenty-five days with a diet containing less than 2000 mg/kg cumate did not cause significant loss of weight in immunocompetent mice (Fig. 3e). For a more comprehensive toxicity analysis, we also examined the activity of the serum enzymes alkaline phosphatase, asparatate aminotransferase, and glutamate dehydrogenase. Twenty-five days of cumate diet, at all cumate concentrations tested (100-6000 mg/kg), did not impact on the levels of these enzymatic markers of liver toxicity (Fig. 3f and Supplementary Fig. 3o–r). The data suggests that the cumate-inducible system has an adequate safety profile for use in vivo.

Lastly, we examined the sensitivity of the cumate-inducible system in vivo (Fig. 3g–j). We were able to rapidly induce luciferase expression in VV-CymR-iGFPLuc infected mice within 1 day of treatment. Upon removal of the cumate, the luciferase signals dissipated within 96 h. Overall, the data establish VV-CymR-iGFPLuc as a recombinant virus enabling tunable expression of a target transgene or viral factor.

Combinatorial application of chemogenetic switches

We noted that relative to the TetO or CuO controlled VV promoters, the VV-adapted T7 system resulted in higher basal expression of transgenes. We reasoned that combining the ST7 system with either operator would result in a more selective safety switch controlled by the use of two small molecules. We generated a virus encoding a TetO immediately downstream of a T7 promoter driving GFPLuc expression (Fig. 4a). The virus also encoded the TetR and ST7 (VV-ST7-TetR-iGFPLuc) to mediate Dox- and rapamycin-inducible expression of the reporter genes, respectively. In VV-ST7-TetR-iGFPLuc infected U2OS cells, we demonstrated that treatment with Dox alone increased luciferase activity 7-fold, whereas treatment with rapamycin alone induced luciferase activity by 19-fold (Fig. 4b). A combination of Dox and rapamycin treatment synergized to activate luciferase activity by 48-fold. Similar trends were observed using the GFP reporter system (Fig. 4c).

**Fig. 4: Double inducible chemogenetic switches overcome major challenges with single molecule systems.**

Analogous to VV-ST7-TetR-iGFPLuc, we developed a recombinant VV encoding ST7, CymR, and a CuO immediately downstream of a T7 promoter driving GFPLuc expression (VV-ST7-CymR-iGFPLuc; Fig. 4d). In VV-ST7-CymR-iGFPLuc infected U2OS cells, treatment with cumate alone, rapamycin alone, or both in combination induced luciferase activity 26-fold, 54-fold, and 97-fold respectively (Fig. 4e, f). Importantly, incorporation of either the TetO or CuO operators into the ST7 system significantly reduced basal “leaky” expression associated with the ST7 system as illustrated by the complete loss of basal luciferase activity in the absence of small-molecule inducers.

During our generation of a VV-adapted cumate-inducible expression system, we noted that while H5R, LEO, and LEO160 promoters produce the strongest luciferase activity in the presence of cumate (Fig. 3c), but also produced the highest basal expression in the absence of cumate. We hypothesized that combining CuO with TetO would produce a more tightly regulated expression system (Cu-Tet double inducible expression system). We engineered seven different recombinant VV expressing CymR, TetR, as well as a CuO and TetO placed immediately downstream of unique VV promoters (p11, H5R, LEO, LEO160, O2LA12L, E3LF17R and P11-NS) driving GFPLuc expression (Fig. 4g). Of the VV promoters screened, LEO in concert with a TetO and CuO produced the highest inducible luciferase activity in the presence of Dox and cumate, while maintaining low basal activity in the absence of small-molecule inducers (Fig. 4h) in infected U2OS cells. Treatment with Dox alone or cumate alone induced intermediate levels of luciferase activity. Analyses of GFP fluorescence revealed similar trends (Fig. 4i). We utilized VV expressing CymR, TetR, and CuO-Tet-O-LEO driven GFPLuc (VV-CymR-TetR-iGFPLuc) for additional in vitro characterization studies. To confirm that this level of tunability was not specific to U2OS cells, we examined the inducibility of luciferase activity in VV-CymR-TetR-iGFPLuc infected HEK293, Vero, HT-29, 786-0, and HeLa cells (Fig. 4j). In all tested cell lines, Dox induced 3- to 7-fold increase in reporter signal, whereas cumate induced a 11- to 23-fold activation of luciferase activity. Combined treatment with Dox and cumate produced maximal increases in luciferase activity ranging from 23- to 45-fold across the five tested cell lines. As a proof-of-concept, we also demonstrated that a 3rd generation Dox-inducible expression system (Tet-ON using reverse tetracycline-controlled transactivator 3, rTA3) can be combined with the cumate-inducible expression system (CymR/CuO) to generate a highly tunable expression system in an oncolytic herpes virus platform (HSV: Fig. 4k–m)—illustrating that the chemogenetic switches developed for VV have the potential to be adapted for other viruses.

The development of the VV-CymR-TetR, VV-ST7-TetR and VV-ST7-CymR expression systems enable the introduction of additional tunable functionality into poxviruses. Further, these inducible expression systems enable multiplexed control of multiple virally encoded transgenes through the addition of various combinations of rapamycin, cumate, and doxycycline.

Chemically regulated control of oncolytic virus replication and spread

Oncolytic viruses with deletions in critical virulence genes have a very strong safety record but this often comes at the price of compromised therapeutic potency and challenges to manufacturing. An alternative strategy for selective replication of an OV is to control expression of key replication gene products. Recent studies have described that deletion of the D13L gene in VV produces a severely attenuated vector incapable of producing infectious virus in healthy volunteers^19,20. Thus, we replaced the endogenous vaccinia virus D13L promoter with a TetO-controlled p11 promoter. (VV-TetR-iD13; Fig. 5a). As expected, in VV-TetR-iD13 infected U2OS cells, the addition of Dox increased plaque size, as observe by GFP fluorescence (Fig. 5b) and Dox dependent cell killing (Supplementary Fig. 4a). These results were observed in a panel of infected human cancer cell lines of a variety of histologies (A549, HeLa, HT-29, SKOV3: Fig. 5c, d and Supplementary Fig. 4b, c). While vaccinia replication as measured by a multistep growth curve analysis was severely compromised in the absence of Dox, the growth rate of VV-TetR-iD13L was indistinguishable from wild-type vaccinia virus in the presence of the inducer (Fig. 5e, f). Furthermore, our observations indicate that this virus replicates 100–1000 times less efficiently in human primary cells when compared to cancer cells (Supplementary Fig. 4d).

**Fig. 5: Application of the tetracycline inducible system as a safety switch to generate a conditionally replicating vaccinia virus.**

We next examined the temporal control of VV-TetR-iD13 in HT-29 xenograft bearing mice in vivo. Bioluminescence imaging of HT-29 xenografts revealed that in the presence of Dox-containing diet, robust tumor selective viral replication was observed within 12 h of viral infection (Fig. 5g–h). However, after changing back to normal diets viral replication was extinguished. Similarly, it was possible to delay the introduction of a Dox-containing diet to 2 days post infection, and robust tumor-localized luciferase activity would be observed 4–6 days post infection. Collectively, the data demonstrates that the chemogenetic switches (e.g., TetR/TetO) can be applied as a safety switch to generate VV vectors with chemically regulated replication kinetics in vivo.

While our primary interest is devising strategies to manipulate oncolytic virus replication and transgene expression, it seems reasonable to expect that conditional virus gene expression could be also used to create effective vaccine vectors for either infectious diseases or certain cancer indications. The use of replication competent VV-vectored vaccines in the clinic to-date has been limited due to potential adverse events, including progressive vaccinia and post-vaccination encephalitis. We hypothesized that the use of the conditionally replicating VV-TetR-iD13 as a backbone vector for immunogen delivery could help reduce safety concerns. To examine the safety of VV-TetR-iD13, we compared the effects of VV-TetR-iD13 and VV-CTRL infection in SCID mice after intravenous injection. VV-TetR-iD13 infection was followed by two days of Dox treatment to allow for initial replication to enhance any antigen expression in the context of a vaccine vector. Five days post infection, we noted a significant number of pox lesions in VV-CTRL infected mice; however, VV-TetR-iD13 infected mice had drastically lower number of lesions (Fig. 6a, d). This result was consistent with bioluminescent imaging of virus localization (Fig. 6b, e). Whereas in the VV-TetR-iD13 infected mice, there was no detected signal, the VV-CTRL infected mice had bioluminescent signals localized to tails, paws, and snouts. Furthermore, VV-CTRL infected mice demonstrated increasing weight loss from 4 to 8 days post infection (Fig. 6c). Virus biodistribution in known mouse VV reservoirs were analyzed via plaque assay and there was more a than tenfold decrease in viral titers in lungs, livers, and spleen. Collectively, the data demonstrate that conditionally replicating VV-TetR-iD13 represents a safer vector for antigen delivery.

**Fig. 6: Conditionally replicating vaccinia virus as a safe alternative for vaccine development.**

We also evaluated the immunogenicity of VV-TetR-iD13 delivering a clinically relevant immunogen. The SARS-CoV-2 spike glycoprotein (VV-S-TetR-iD13) or a prefusion-stabilized spike variant (VV-S-HexaPro-TetR-iD13) was encoded into VV-TetR-iD13 to generate SARS-CoV-2 vaccine candidates. We confirmed that these viruses maintained Dox-inducible replication (Supplementary Fig. 5a, b) and expressed their respective antigens (Supplementary Fig. 5c). Subsequently, we analyzed the humoral responses induced in mice infected with these vaccine candidates in the presence of 48 h Dox treatment. A single dose of VV-S-TetR-iD13 or VV-S-HexaPro-TetR-iD13 was sufficient to induce robust Spike-targeted IgG response (Fig. 6f). However, an examination of neutralizing antibody titer levels via SARS-CoV-2 S pseudovirus neutralization assay suggested that HexaPro more consistently elicited a stronger response (Fig. 6g). In the absence of Dox treatment, there was a detectable, but significantly impaired antibody response elicited by VV-S-HexaPro-TetR-D13 (Fig. 6h, i). Further, we confirmed that VV-S-HexaPro-TetR-D13 induced higher antibody responses relative to a non-replicating vaccinia virus vector (MVA) delivering the same antigen (Fig. 6j, k). Taken together, these results demonstrate that chemogenetic switches can be applied to increase the safety of replicating vaccine VV vectors while maintaining their ability to induce strong immunogen specific antibody responses.

Chemogenetic control of virus directed membrane fusion

There is accumulating evidence that encoding membrane fusion proteins into an oncolytic virus backbone can improve therapeutic activity by enabling neutralizing antibody-resistant viral spread, immunogenic cell death, and minimizing the release of virus into healthy tissue or systemic circulation²¹. However, expression of heterologous fusogenic proteins can also be counter-productive to OV replication, impacting virus manufacturing and in some circumstances causing off-target toxicity^22,23,24,25. We hypothesized that integration of a chemogenetic switch to mediate temporal control of fusogenic protein expression would overcome some of these limitations. To identify an optimal fusion protein, we performed an overexpression screen in HEK293 cells of arenavirus and reovirus fusion proteins. To this end, HEK293 cells expressing GFP were transfected with a mammalian expression plasmid encoding a range of fusion proteins and then co-cultured with HEK293 cells stably expressing mCherry. Twenty-four hours post transfection cells were analyzed for co-localization of mCherry and GFP (indicative of fusion between heterotypic cells). (Fig. 7a; Supplementary Fig. 6a). The screen revealed that two fusion proteins produced significant amounts of syncytia—the Tamiami virus glycoprotein (TAMV-GP) and reovirus p14 in a variety of cell lines (U2OS, Vero, A549 and HeLa, Supplementary Fig. 6b). We performed an Alamar Blue assay to analyze the effects of the GPs on cell viability (Fig. 7b). In a panel of cell lines (HEK293, U2OS, Vero, A549, and HeLa), overexpression of TAMV-GP or p14 resulted in a greater than 70% reduction of cell viability, whereas overexpression of p15, a reovirus fusion protein produced low levels of syncytia and resulted in minimal cell death (Fig. 7b). Previous work has demonstrated that expression of p14-induced cell fusion results in increased caspase 3 activation²⁵. Consistent with this, we observed increased caspase activity in cells overexpressing p14 or TAMV-GP (Supplementary Fig. 6c). As both TAMV-GP and p14 showed robust induction of fusion and cell death, we generated recombinant VV strains encoding each fusion protein under the control of our Dox-inducible expression system (VV-TetR-iP14 and VV-TetR-iTAMV-GP; Supplementary Fig. 6d). Vaccinia virus has a complex replication cycle with multiple viral isoforms that contribute to its ability to disperse throughout an infected individual. To examine the contribution of membrane fusion to local virus spread within the tumor microenvironment, we analyzed virus-induced syncytia formation in the presence of ST-246, an established inhibitor of extracellular isoforms of orthopoxvirus²⁶ (Fig. 7c). While both viruses induced robust cell fusion in all tested cell lines in the presence of Dox, the addition of ST-246 significantly impaired the ability of VV-TetR-iP14 to induce syncytium. On the other hand, VV-TetR-iTAMV-GP’s ability to induce syncytia was not impacted by ST-246 treatment – suggesting VV-TetR-iTAMV-GP, in the presence of Dox treatment, has an enhanced ability to spread via cell fusion relative to VV-TetR-iP14. We analyzed the impact of TAMV-GP expression on the growth of the recombinant virus. Multistep viral growth curve analysis in A549 and HeLa cells showed that VV-TetR-iTAMV-GP infection with simultaneous Dox treatment results in decreased viral growth (Fig. 7d, e); however, delaying expression of TAMV-GP by treating with dox 48 h post infection resulted in similar viral growth kinetics to untreated VV-TetR-iTAMV-GP infected cells. This suggests that while constitutive expression of the fusion protein impairs viral growth, the use of a chemogenetic switch will enable temporal control of TAMV-GP expression allowing productive virus infection of the tumor before initiating membrane fusion and its downstream therapeutic sequelae. To test this hypothesis, we evaluated the anti-tumor effects of VV-TetR-iTAMV-GP in vivo using the A549 xenograft model infected in the presence or absence of Dox (Fig. 7f–i). Without a Dox-containing diet, VV-TetR-iTAMV-GP had decreased anti-tumor activity which was significantly enhanced in mice fed a Dox-containing diet 48 h post infection. Tumor volumes and weight were significantly reduced in mice co-treated with VV-TetR-iTAMV-GP and Dox relative to the virus alone. Immunohistochemistry analysis of tumors revealed Dox treatment-specific induction of TAMV-GP and caspase activation in virus infected cells, consistent with our in vitro results (Fig. 7i, Supplementary Fig. 6e). We repeated the study utilizing intravenous delivery, and our results demonstrated that systemic delivery of VV-TetR-iTAMV-GP in the presence of a Dox-containing diet was superior to virus alone (Supplementary Fig. 6f–h). Collectively, our results demonstrate that the application of chemogenetic switches in VV enables the chemically regulated delivery of virus-inhibitory payloads strategically delaying transgene expression while optimizing therapeutic effect.

**Fig. 7: Doxycycline-inducible system as an effective strategy to regulate expression of a toxic fusogenic protein.**

Orchestrating transgene expression for safety and efficacy

Several studies have attempted to deliver cytokines using viral vectors^27,28, however, for a subset of potent, pro-inflammatory cytokines, the safety and efficacy of cytokine therapy is heavily influenced by the timing and intensity of dosage. Our chemogenetic switches provide an opportunity to adjust both timing and magnitude of payload expression. For these experiments we used the cumate-based operator as we had established the dynamic range of the VV-adapted cumate-regulated expression system in vivo (Fig. 3c–j; Supplementary Fig. 3h). Numerous lines of evidence indicate that IL-12, IL-2 and IL-18 give complementary immunoregulatory signals^29,30,31,32. It has been demonstrated that these three cytokines exert synergistic anti-tumor action in preclinical tumor models by activating the innate immune response, increasing the cytotoxicity of natural killer cells, natural killer T cells and granulocytes, and potently shutting down the tumor vasculature and reversing the immune suppression mediated by dysfunctional myeloid cells in the tumor^29,33,34,35. We generated several different PEL-CuO driven gene cassettes for combinatorial expression of the three cytokines, including a multi-cistronic construct with internal ribosome entry sites (VV-CymR-iIL2IRES-2-DIRES-18; Supplementary Fig. 7a), 2A self-cleaving peptide-based multi-gene expression system (VV-CymR-iIL12-P2A-2-T2A-18; Supplementary Fig. 7b), as well as independent PEL-CuO promoters driving each cytokine (Fig. 8a; VV-CymR-iIL12-2-18). Of the three tested systems, ELISA analyses revealed that three separate promoter systems resulted in the most robust, cumate-inducible expression of all three cytokines (Fig. 8b, Supplementary Fig. 7c). All three cytokines are known to activate IFN-γ and TNF-α expression in immune cells and so we cultured mouse splenocytes with the virus-filtered supernatants of VV-CymR-iIL12-2-18 infected U2OS cells and examined the levels of both cytokines in splenocyte supernatants. In the presence of cumate, VV-CymR-iIL12-2-18 infected cells’ supernatant produced robust increases in IFN-γ and TNF-α production (Fig. 8c, d). The functional role of each cytokine was examined through the use of neutralizing antibodies against IL-2, IL-12 or IL-18. Blocking of IL-2 or IL-12 in the supernatants impaired activation of IFN-γ and TNF-α production while blocking of IL-18 only impaired IFN-γ production from splenocytes. These in vitro results demonstrated that VV-CymR-iIL12-2-18 encodes a cumate-regulated multi-expression system for IL-2, IL-12, and IL-18.

**Fig. 8: Applications of the cumate “safety switch” in regulating expression of potentially toxic cytokines.**

We examined whether the chemical regulation of cytokine expression by VV-CymR-iIL12-2-18 circumvents the toxicity associated with systemic administration of these cytokines. For our toxicity studies, we utilized CD-1 nude mice to mimic an immunocompromised cancer patient, who may be more susceptible to cytokine-induced immunopathology due to persistence of the VV vector after administration. Whereas infection with VV-CTRL or VV-CymR-iIL12-2-18 in the absence of cumate resulted in no mouse mortality after 5 days, infection with VV constitutively expressing the three cytokines (VV-IL12-2-18) resulted 100% mortality (Fig. 8e). This is consistent with constitutive high levels of expression of all three cytokines being toxic. 80% mortality was observed in the mice infected with VV-CymR-iIL12-2-18 and fed a 6000 mg/kg cumate diet to induce comparable levels of cytokine production in mouse sera (Fig. 8f). Similar to VV-IL12-2-18, the 6000 mg/kg cumate combined with VV-CymR-iIL12-2-18 infection induced lung edema (Fig. 8g), and liver toxicity as evidenced by increases in serum alkaline phosphatase activity levels (Fig. 8h). In order to determine whether cumate dosage could be used to limit toxicity of VV-CymR-iIL12-2-18, we fed infected mice with a range of cumate doses (1000 mg/kg, 2000 mg/kg and 6000 mg/kg). At 1000 mg/kg cumate, we observed no mortality after 5 days in VV-CymR-iIL12-2-18 infected CD-1 nude mice (Fig. 8i). The 1000 mg/kg cumate diet induced a lower level of cytokine production in infected mice relative to the other tested doses and caused no significant differences in lung edema (Fig. 8k), or elevated liver markers (serum alkaline phosphatase or asparatate aminotransferase activity) (Fig. 8l, m). These results demonstrate that 1000 mg/kg is safe dosage of cumate in VV-CymR-iIL12-2-18 infected mice. Finally, we examined the anti-tumor effects of the combination of VV-CymR-iIL12-2-18 and 1000 mg/kg cumate in immunocompetent mice with peritoneal murine cancer (MC38) (Fig. 8n). As expected, mice treated with VV constitutively expressing the three cytokines while providing some limited benefit over VV lacking the transgenes, ultimately succumbed likely due to toxicity of the three cytokines. In contrast, with controlled cytokine expression using the VV-CymR-iIL12-2-18 virus and 1000 mg/kg cumate we observed increased mouse survival from the tumor challenge to 40%. These results demonstrate that the chemogenetic switch enables dosage control to mediate delivery of potentially toxic payloads at safe and efficacious levels.

Finally, we demonstrated that both virus replication and payload delivery can be placed under separate controls to increase safety of the VV vector (VV-CymR-iIL12-2-18/TetR-iD13). This vector replicates conditionally in the presence of Dox (Fig. 8o) and expresses the three cytokines in the presence of cumate (Fig. 8p). It should be noted that both cumate and dox treatment were required for VV-CymR-iIL12-2-18/TetR-iD13 infected cells to produce the same levels of cytokine as VV-CymR-iIL12-2-18 in the presence of cumate. This demonstrates that Dox-conditional replication adds another level control on payload delivery in addition to dox-inducible replication. Lastly, we examined the anti-tumor effects of VV-CymR-iIL12-2-18/TetR-iD13 in the murine model with peritoneal MC38. In line with the results obtained with 1000 mg/kg cumate and VV-CymR-iIL12-2-8 (Fig. 8n), VV-CymR-iIL12-2-18/TetR-iD13 infection with induced replication (Dox treatment) and payload delivery (cumate treatment) resulted in 40% mouse survival (Fig. 8q). Taken together, our results demonstrate the application of two chemogenetic switches in VV enables the independent chemical regulation of both replication and payload expression.

Discussion

Challenges to the development of replication competent viral vectors as therapeutics include safety concerns related to uncontrolled viral dissemination in cancer patients and healthcare workers as well as off-target delivery of potentially toxic payloads. Here we develop three VV-adapted chemogenetic switches: the rapamycin-inducible ST7 RNA polymerase expression system (FRB/FKBP; Fig. 1), Dox-inducible expression (TetR/TetO; Fig. 2), and cumate-inducible expression (CymR/CuO; Fig. 3) that can be used alone or in combination to control VV replication and/or transgene expression. This study describes, to the best of our knowledge, the first application of the cumate- or rapamycin-inducible expression systems in a replicating viral vector. These systems will facilitate the development of a range of vectors that can potentially deliver potent transgenes in a safe and controlled fashion to enhance therapeutic outcomes. The use of combinatorial chemogenetic switches is particularly appealing as they provide the clinician with opportunities to tune both the level of virus replication and therapeutic transgene expression. One of the challenges in the development of oncolytic virus vectors is finding the balance between safe selective replication and potency through expression of viral virulence gene products. A strategy that has shown promise is the development of vectors that control certain viral virulence genes using tumor selective promoter elements. For instance oncolytic herpes viruses have been developed for selective replication in glioma cells using the nestin promoter to control the virulence gene ICP 34.5³⁶. Our system has the advantage of providing control over the virus replication and gene expression in any tumor type as well as being able to both turn on and off virus gene expression. The cumate-controlled system demonstrated good dynamic range. However, it is not currently an FDA-approved compound thus limiting its rapid application into the clinic. On the other hand, both doxycline and the rapalogs used in this study are FDA-approved drugs and thus their corresponding chemogenetic switches could have a more straight forward regulatory approval path. Ultimately, the expansion of the available toolkits to regulate payload delivery will enable the incorporation of more complex synthetic circuits within viral vectors. Towards this goal, we combined different conditional expression systems (ST7-CymR, ST7-TetR, and Cu-Tet) to create expression systems with two levels of control and decreased basal level of expression. These systems will prove critical to the delivery of potent toxic payloads that can be locally therapeutically effective but have significant toxicity when delivered systemically. Lastly, while our studies have been focused on engineering tunable expression systems in VV, these conditional expression systems can be rapidly adapted to other viral vectors, as we demonstrate for HSV (Fig. 4k–m). We anticipate that each viral vector, with its own unique replication kinetics and replication cycle will have different selectivity and robustness of inducible transgene expression; therefore, expanding the available systems will enable the selection of an expression system tailored to the viral vector or application. Overall, the application of viral expression switches described herein should enable the development of conditionally replicating viral vectors or inducible transgene expression system. This technology will enable testing of highly potent transgenes in oncolytic virus vectors and expand their utility in the clinic.

Methods

Ethics

The study discussed here, adheres to all pertinent ethical guidelines at OHRI and the University of Ottawa (certificate for biohazardous material utilization GC317-125-12). The University of Ottawa’s institutional animal care committee approved all animal experiments (Protocol ID: OHRI2870 and MEe-2258) which were conducted following the National Institutes of Health and the Canadian Council on Animal Care standards.

Viruses

MVA (VR-1508) and VV Copenhagen strain were purchased from the ATCC. The Vaccinia TT strain was a gift from Dr. David Evans. HSV strain KOS was used.

Cell culture

All cell lines were purchased from the American Type Culture Collections (Manassas, VA). The cells were grown in Dulbecco’s Modified Eagle Medium/DMEM (GE Healthcare Life Sciences; ON, CAN) or Roswell Park Memorial Institute/RPMI 1640 Medium (Gibco; MA, USA), supplemented with 10% fetal bovine serum/FBS (Gibco). Cells were kept at 37 °C in a humidified environment with 5% CO₂. Cells co-cultured with effector cells were cultivated in RPMI medium supplemented with 10% FBS and 1% penicillin/streptomycin (volume/volume) (Gibco). Co-cultures were kept at 37 °C in a humidified environment with 5% CO₂. Cells were regularly checked for mycoplasma contamination using PCR (e-Myo VALiD Detection Kit, 25239, LiliF Diagnostics; South Korea) and were shown to be mycoplasma-free.

Construct design

The promoter constructs were designed using Snapgene software. codon-optimized inserts were ordered from GenScript (Piscataway, NJ, USA).

Bioluminescence Imaging

After lysis of virus infected cells with passive lysis buffer, the lysate was moved to 96-well plates and 1:1 ratio of the luciferase reagent (Promega) was added. A Synergy microplate reader (BioTek, Winooski, VT, USA) was used to measure luminescence. Results are presented as RLU normalized to control. The data presented are the mean of three independent experiments.

In vitro co-culture cell viability assays

The resazurin assay was used to determine cell cytotoxicity caused by viruses or transgenes. The metabolic activity of the cells was measured using the manufacturer’s protocol and resazurin sodium salt (R12204; ThermoFisher Scientific). In each well, treated and/or infected cells were given 10% (v/v, final) resazurin and incubated for 2–4 h, depending on the cell line. Using a BioTek Microplate Reader, fluorescence was measured at 590 nm after excitation at 530 nm (BioTek, Winooski, VT, USA). In a nutshell, freshly isolated splenocytes were co-cultured with target viruses in a 24-well plate or a flat-bottom 96-well plate with just media. In short, 24 h before, 200,000 (24-well plate) or 15,000 (96-well plate) target cells were seeded. Virus infections were performed on cells in serum-free media at the indicated MOIs. After 6 h, the media was removed and splenocytes in RPMI supplemented with 10% FBS and 1% (v/v) penicillin/streptomycin were added.

Reagents

Cumate solution (QM150A-1) was purchased from System Biosciences, doxycycline (D9891-25G) was purchased from Sigma-Aldrich, and Rapamycin analogs were purchased from Selleck Chemicals.

ELISA

Nunc Maxisorp 96-well flat-bottom plates were coated with 125 ng of RBD per well overnight at 4 °C (prepared in-house). The RBD solution was removed the next day, and the plates were washed three times with PBS-Tween (0.1% Tween 20) before being blocked for 1 h with a 3% skim milk solution. Vaccinated mouse sera were then serially diluted in 1% skim milk and added to the plates to incubate for 2 h at room temperature. In addition, a positive and negative control consisting of a monoclonal RBD antibody (1 g/mL; Cat No: MBS434247, Anti-RBD Domain [SARS-CoV-2 spike], monoclonal antibody, MyBioSource, CA, USA) and a pool of sera taken from mice prior to vaccination were added to each plate. Following the 2-h incubation, plates were washed with PBS-Tween and incubated for 1 h at room temperature with anti-mouse immunoglobulin (Ig)G conjugated to horseradish peroxidase (HRP) (1:3000; Cat No: 314930, Goat anti-Mouse IgG [H + L] Secondary Antibody, HRP, Invitrogen). After that, plates were developed with SigmaFast OPD solution and measured at 490 nm with a BioTek microplate reader.

All experimental absorbance measurements were normalized to the blank and the positive control (monoclonal RBD antibody at 1 g/mL) and fitted with a quadratic binding polynomial assuming 1:1 binding. A Monte Carlo simulation with the nonlinear curve-fitting tool in QtGrace was used to perform the fitting. The reciprocal antibody titer (LDF) was calculated by interpolating the dilution factor that intersected with a minimum detection threshold defined by 10x the standard deviation of responses from TT WT-vaccinated mice, or to a fixed value of 0.025 (whichever was larger).

Mouse experiments

Six- to 8-week-old female C57BL/6 or CD-1 nude mice (The Jackson Laboratory, Bar Harbor, ME) were obtained for studies. All experiments were approved by the University of Ottawa animal care and veterinary services (MEe-2258-R5, or OHRIe-3340-A), including periodic saphenous vein bleeds for serum collection.

Pseudovirus neutralization assay

At the time of infection, Vero E6 cells were seeded in 96-well plates with 40,000 cells per well. Serum was first diluted in serum-free DMEM at a 1:10 dilution in a separate 96-well plate, followed by a serial 1 in 2 dilution series. VSV pseudotyped with the SARS-CoV-2 spike glycoprotein and co-encoded with eGFP was then added to serum in an equal volume of serum-free DMEM for a final dilution of 2000 pfu per well and incubated at 37 °C for 1 h. After 1 h, the media on the cell was replaced with 60 L of the virus/serum and incubated at 37 °C for 1 h. The wells were then filled with carboxymethylcellulose (CMC) in DMEM (supplemented with 10% fetal bovine serum) to a final concentration of 3% CMC and incubated at 34 °C for 24 h. Using a Cellomics ArrayScan VTI HCS Reader, GFP foci were imaged and counted.

Western blot

Whole cell lysates were prepared in Cell Lysis Buffer II (Invitrogen™) with protease/phosphatase inhibitors (NEB). The BCA assay was used to determine protein concentrations (Pierce). Samples were loaded into precast SDS-PAGE gels after being mixed with NuPage LDS sample buffer (Invitrogen) (Bio-Rad). Proteins were transferred onto nitrocellulose membrane after SDS-PAGE, blocked for 1 h with 5% skim milk in TBS-Tween, and incubated with primary antibody. The following antibodies SARS-CoV-2 Spike (GeneTex, 1A9), HA (Thermo Fisher Scientific, 26183), VV (Abcam, ab117453), and GADPH (Cell Signaling, 2118L) antibodies, which were used at a dilution of 1:1000. Cell Signaling supplied HRP conjugated anti-mouse (7076S) and anti-rabbit (7074S) IgG secondary antibodies, which were used at a dilution of 1:2000. Westerns were created using Bio-Rad Clarity or Clarity Max ECL substrates and imaged with a Bio-Rad GelDoc imaging system. Uncropped western blots are shown in Supplementary Fig. 9.

Virus production and titration

HeLa cells in 850 cm² roller bottles were infected with virus at a MOI of 0.03 without removing inoculation media to produce vaccinia virus. Cells were cultured at 37 °C in 5% CO₂ for 72 h, or until a sufficient cytopathic effect was observed. Cells were pelleted and resuspended in 1 mM Tris at pH 9.0. Cells were frozen/thawed three times before being centrifuged at 700 × g for 10 min at room temperature. Before centrifugation at 20,666 × g for 1 h and 30 min at 4 °C, supernatant was collected and overlaid onto 36% sucrose cushions. Reconstituted viral pellets in 1 ml of 1 mM Tris and stored at −80 °C. Titration was used to determine viral titers. Virus stocks were serially diluted tenfold before infecting U2OS cells in 12-well plates. After 2 h at 37 °C in 5% CO₂, the media was replaced with overlay medium (1:1 of 3% CMC and 2x DMEM + FBS) and incubated for 48 h at 37 °C in 5% CO2. Plaques were stained with crystal violet after 48 h and plaque forming units per ml were calculated. Vero cells were used for HSV-1 virus production and titration under the same conditions as described above. Schematic illustration of VV-CTRL, VV-GFPLuc is shown in supplementary Fig. 8.

Statistical analyses

GraphPad Prism v9 was used to create all graphs and statistical analyses. When appropriate, unpaired two-samples t-test was used. In figure legends, the n value represents all biological replicates. As indicated in the figure legends, error bars represent standard deviation (SD) or standard error of the mean (SEM). For all statistical tests, ns P > 0.05, *P < 0.05 **P < 0.003841, ***P < 0.000125, ****P < 0.001.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The authors declare that all data supporting the conclusions of this study are presented within the paper and the supplementary information files and are available from the authors. Schematic pictures for in vivo experiments were created with BioRender.com (Figs. 1i, 2h, 3g, 5g, and 6f). Source data are provided with this paper.

References

Fan, J.-X., Niu, M.-T., Qin, Y.-T., Sun, Y.-X. & Zhang, X.-Z. Progress of engineered bacteria for tumor therapy. Adv. Drug Deliv. Rev. 185, 114296 (2022).
Article CAS PubMed Google Scholar
Lim, W. A. & June, C. H. The principles of engineering immune cells to treat cancer. Cell 168, 724–740 (2017).
Article CAS PubMed PubMed Central Google Scholar
Twumasi-Boateng, K., Pettigrew, J. L., Kwok, Y., Bell, J. C. & Nelson, B. H. Oncolytic viruses as engineering platforms for combination immunotherapy. Nat. Rev. Cancer 18, 419–432 (2018).
Article CAS PubMed Google Scholar
Khalil, A. S. & Collins, J. J. Synthetic biology: applications come of age. Nat. Rev. Genet. 11, 367–379 (2010).
Article CAS PubMed PubMed Central Google Scholar
Cubillos-Ruiz, A. et al. Engineering living therapeutics with synthetic biology. Nat. Rev. Drug Discov. 20, 941–960 (2021).
Article CAS PubMed Google Scholar
Arulanandam, R. et al. VEGF-mediated induction of PRD1-BF1/Blimp1 expression sensitizes tumor vasculature to oncolytic virus infection. Cancer Cell 28, 210–224 (2015).
Article CAS PubMed Google Scholar
Everts, A., Bergeman, M., McFadden, G. & Kemp, V. Simultaneous tumor and stroma targeting by oncolytic viruses. Biomedicines 8, 474 (2020).
Article CAS PubMed PubMed Central Google Scholar
Pu, J., Zinkus-Boltz, J. & Dickinson, B. C. Evolution of a split RNA polymerase as a versatile biosensor platform. Nat. Chem. Biol. 13, 432–438 (2017).
Article CAS PubMed PubMed Central Google Scholar
Cai, X. et al. Sirolimus decreases circulating lymphangioleiomyomatosis cells in patients with lymphangioleiomyomatosis. Chest 145, 108–112 (2014).
Article CAS PubMed Google Scholar
Dancey, J. mTOR signaling and drug development in cancer. Nat. Rev. Clin. Oncol. 7, 209–219 (2010).
Article CAS PubMed Google Scholar
Su, D. W., Mita, M. & Mita, A. C. mTOR Inhibition for Cancer Therapy: Past, Present and Future 161–189 (Springer, 2016).
Bukowski, R. M. Temsirolimus: a safety and efficacy review. Expert Opin. Drug Saf. 11, 861–879 (2012).
Article CAS PubMed Google Scholar
Lebwohl, D. et al. Development of everolimus, a novel oral mTOR inhibitor, across a spectrum of diseases. Ann. N. Y. Acad. Sci. 1291, 14–32 (2013).
Article CAS PubMed ADS Google Scholar
Rodriguez, J. F. & Smith, G. L. Inducible gene expression from vaccinia virus vectors. Virology 177, 239–250 (1990).
Article CAS PubMed Google Scholar
Gossen, M., Bonin, A. L., Freundlieb, S. & Bujard, H. Inducible gene expression systems for higher eukaryotic cells. Curr. Opin. Biotechnol. 5, 516–520 (1994).
Article CAS PubMed Google Scholar
Kallunki, T., Barisic, M., Jäättelä, M. & Liu, B. How to choose the right inducible gene expression system for mammalian studies? Cells 8, 796 (2019).
Article CAS PubMed PubMed Central Google Scholar
Merten, O.-W., Hebben, M. & Bovolenta, C. Production of lentiviral vectors. Mol. Ther. Methods Clin. Dev. 3, 16017 (2016).
Article PubMed PubMed Central Google Scholar
Manceur, A. P. et al. Scalable lentiviral vector production using stable HEK293SF producer cell lines. Hum. Gene Ther. Methods 28, 330–339 (2017).
Article CAS PubMed PubMed Central Google Scholar
Prow, N. A. et al. The vaccinia virus based Sementis Copenhagen Vector vaccine against Zika and chikungunya is immunogenic in non-human primates. npj Vaccines 5, 1–10 (2020).
Article Google Scholar
Prow, N. A. et al. A vaccinia-based single vector construct multi-pathogen vaccine protects against both Zika and chikungunya viruses. Nat. Commun. 9, 1–12 (2018).
Article CAS ADS Google Scholar
Guedan, S. et al. Syncytia formation affects the yield and cytotoxicity of an adenovirus expressing a fusogenic glycoprotein at a late stage of replication. Gene Ther. 15, 1240–1245 (2008).
Article CAS PubMed PubMed Central Google Scholar
Brown, C. W. et al. The p14 FAST protein of reptilian reovirus increases vesicular stomatitis virus neuropathogenesis. J. Virol. 83, 552–561 (2009).
Article CAS PubMed Google Scholar
Le Boeuf, F. et al. Reovirus FAST protein enhances vesicular stomatitis virus oncolytic virotherapy in primary and metastatic tumor models. Mol. Ther. Oncolytics 6, 80–89 (2017).
Article PubMed PubMed Central Google Scholar
Krabbe, T. & Altomonte, J. Fusogenic viruses in oncolytic immunotherapy. Cancers 10, 216 (2018).
Article PubMed PubMed Central Google Scholar
Del Papa, J., Petryk, J., Bell, J. C. & Parks, R. J. An oncolytic adenovirus vector expressing p14 FAST protein induces widespread syncytium formation and reduces tumor growth rate in vivo. Mol. Ther. Oncolytics 14, 107–120 (2019).
Article PubMed PubMed Central Google Scholar
Jordan, R., Leeds, J. M., Tyavanagimatt, S. & Hruby, D. E. Development of ST-246® for treatment of poxvirus infections. Viruses 2, 2409–2435 (2010).
Article CAS PubMed PubMed Central Google Scholar
Liu, L. et al. Advances in viral-vector systemic cytokine gene therapy against cancer. Vaccine 28, 3883–3887 (2010).
Article CAS PubMed Google Scholar
Carew, J. F. et al. A novel approach to cancer therapy using an oncolytic herpes virus to package amplicons containing cytokine genes. Mol. Ther. 4, 250–256 (2001).
Article CAS PubMed Google Scholar
Gollob, J. A., Schnipper, C. P., Murphy, E. A., Ritz, J. & Frank, D. A. The functional synergy between IL-12 and IL-2 involves p38 mitogen-activated protein kinase and is associated with the augmentation of STAT serine phosphorylation. J. Immunol. 162, 4472–4481 (1999).
Article CAS PubMed Google Scholar
Lauwerys, B. R., Renauld, J.-C. & Houssiau, F. A. Synergistic proliferation and activation of natural killer cells by interleukin 12 and interleukin 18. Cytokine 11, 822–830 (1999).
Article CAS PubMed Google Scholar
Rodriguez-Galán, M. C., Bream, J. H., Farr, A. & Young, H. A. Synergistic effect of IL-2, IL-12, and IL-18 on thymocyte apoptosis and Th1/Th2 cytokine expression. J. Immunol. 174, 2796–2804 (2005).
Article PubMed Google Scholar
Pintarič, M., Gerner, W. & Saalmüller, A. Synergistic effects of IL-2, IL-12 and IL-18 on cytolytic activity, perforin expression and IFN-γ production of porcine natural killer cells. Vet. Immunol. Immunopathol. 121, 68–82 (2008).
Article PubMed Google Scholar
Wiltrout, R. H. & Wigginton, J. M. IL-12/IL-2 combination cytokine therapy for solid tumours: translation from bench to bedside. Expert Opin. Biol. Ther. 2, 513–524 (2002).
Article PubMed Google Scholar
Wigginton, J. M. et al. IFN-γ and Fas/FasL are required for the antitumor and antiangiogenic effects of IL-12/pulse IL-2 therapy. J. Clin. Investig. 108, 51–62 (2001).
Article CAS PubMed PubMed Central Google Scholar
D’Anna, F., Yamauchi, A., Yu, C.-R., Horvath-Arcidiacono, J. A. & Bloom, E. T. IL-12 synergizes with IL-2 to induce lymphokine-activated cytotoxicity and perforin and granzyme gene expression in fresh human NK cells. Cell. Immunol. 165, 33–43 (1995).
Article Google Scholar
Kambara, H., Okano, H., Chiocca, E. A. & Saeki, Y. An oncolytic HSV-1 mutant expressing ICP34. 5 under control of a nestin promoter increases survival of animals even when symptomatic from a brain tumor. Cancer Res. 65, 2832–2839 (2005).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We would like to thank other members of the Bell, Diallo, Ilkow, and Auer laboratories. We thank the personnel of the Flow Cytometry Core Facility, Histology Core Facility, and Animal Care and Veterinary Services of the Faculty of Medicine at the University of Ottawa for their support. We thank Dr. Keir Joe Menzies for providing differentiated Myoblast. We thank Bradley Austin, Clarence Choy, Georgia Cheung, Emily Wang, Tina Chen, Emma Le, and Kyle Nguyen for their support and help. Thanks to This Week in Virology for inspiration, specifically Dr. Vincent Racaniello. J.C.B. and C.S.I. support from CIHR, CCSRI and BioCanRx. J.C.B. also supported by Prostate Cancer Canada, and the Terry Fox Research Institute. M.J.F.C., S.B., and T. Azad received funding support from CanPRIME/Mitacs fellowships. M.J.F.C. is funded by the Taggart-Parkes Fellowship. T. Azad is funded by a CIHR postdoctoral award. A.P. was funded by the Lebovic Fellowship Funding. B.C.D. was supported by National Institute of General Medical Sciences (R35 GM119840) of the National Institutes of Health (NIH).

Author information

These authors contributed equally: Taha Azad, Reza Rezaei.

Authors and Affiliations

Ottawa Hospital Research Institute, Ottawa, ON, K1H 8L6, Canada
Taha Azad, Reza Rezaei, Ragunath Singaravelu, Stephen Boulton, Julia Petryk, Kemal Alper Onsu, Nikolas T. Martin, Victoria Hoskin, Marie Marotel, Ricardo Marius, Xiaohong He, Mathieu J. F. Crupi, Michele Ardolino, Carolina S. Ilkow & John C. Bell
Faculty of Medicine and Health Sciences, Department of Microbiology and Infectious Diseases, Université de Sherbrooke, Sherbrooke, QC, J1E 4K8, Canada
Taha Azad
Centre de Recherche du CHUS, Sherbrooke, QC, J1H 5N4, Canada
Taha Azad & Lee-Hwa Tai
Department of Biochemistry, Microbiology and Immunology, University of Ottawa, Ottawa, ON, K1H 8M5, Canada
Reza Rezaei, Ragunath Singaravelu, Marie Marotel, Mathieu J. F. Crupi, Huy-Dung Hoang, Abolfazl Nik-Akhtar, Tommy Alain, Michele Ardolino, Carolina S. Ilkow & John C. Bell
Quantitative Biosciences Institute (QBI), University of California San Francisco, San Francisco, CA, CA 94158, USA
Adrian Pelin
Department of Biology, University of Ottawa, Ottawa, ON, K1N 6N5, Canada
Mina Ghahremani
Center for Infection, Immunity, and Inflammation, University of Ottawa, Ottawa, ON, K1H, Canada
Marie Marotel & Michele Ardolino
Children’s Hospital of Eastern Ontario Research Institute, Ottawa, ON, K1H 8L1, Canada
Huy-Dung Hoang & Tommy Alain
Department of Biology, Ottawa Institute of Systems Biology, Carleton University, Ottawa, ON, K1S 5B6, Canada
Abolfazl Nik-Akhtar & Ashkan Golshani
Department of Microbiology, Faculty of Biological Sciences, Alzahra University, Tehran, Iran
Mahsa Ahmadi
Department of Biotechnology, College of Science, University of Tehran, Tehran, Iran
Nika Kooshki Zamani
Department of Pathology and Molecular Medicine, Queens University, Kingston, ON, K7L 3N6, Canada
Peter Greer
Department of Chemistry, The University of Chicago, Chicago, IL, USA
Bryan C. Dickinson
Department of Immunology & Cell Biology, Université de Sherbrooke, Sherbrooke, QC, J1E 4K8, Canada
Lee-Hwa Tai

Authors

Taha Azad
View author publications
You can also search for this author in PubMed Google Scholar
Reza Rezaei
View author publications
You can also search for this author in PubMed Google Scholar
Ragunath Singaravelu
View author publications
You can also search for this author in PubMed Google Scholar
Adrian Pelin
View author publications
You can also search for this author in PubMed Google Scholar
Stephen Boulton
View author publications
You can also search for this author in PubMed Google Scholar
Julia Petryk
View author publications
You can also search for this author in PubMed Google Scholar
Kemal Alper Onsu
View author publications
You can also search for this author in PubMed Google Scholar
Nikolas T. Martin
View author publications
You can also search for this author in PubMed Google Scholar
Victoria Hoskin
View author publications
You can also search for this author in PubMed Google Scholar
Mina Ghahremani
View author publications
You can also search for this author in PubMed Google Scholar
Marie Marotel
View author publications
You can also search for this author in PubMed Google Scholar
Ricardo Marius
View author publications
You can also search for this author in PubMed Google Scholar
Xiaohong He
View author publications
You can also search for this author in PubMed Google Scholar
Mathieu J. F. Crupi
View author publications
You can also search for this author in PubMed Google Scholar
Huy-Dung Hoang
View author publications
You can also search for this author in PubMed Google Scholar
Abolfazl Nik-Akhtar
View author publications
You can also search for this author in PubMed Google Scholar
Mahsa Ahmadi
View author publications
You can also search for this author in PubMed Google Scholar
Nika Kooshki Zamani
View author publications
You can also search for this author in PubMed Google Scholar
Ashkan Golshani
View author publications
You can also search for this author in PubMed Google Scholar
Tommy Alain
View author publications
You can also search for this author in PubMed Google Scholar
Peter Greer
View author publications
You can also search for this author in PubMed Google Scholar
Michele Ardolino
View author publications
You can also search for this author in PubMed Google Scholar
Bryan C. Dickinson
View author publications
You can also search for this author in PubMed Google Scholar
Lee-Hwa Tai
View author publications
You can also search for this author in PubMed Google Scholar
Carolina S. Ilkow
View author publications
You can also search for this author in PubMed Google Scholar
John C. Bell
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

J.C.B. proposed and supervised the project. T. Azad, R.R., R.S., S.B., K.A.O., N.T.M., V.H., M.G., M.M., M.J.F.C., H.D.H., A.N.A., M. Ahmadi, N.K.Z. conducted in vitro experiments. T. Azad, R.R., J.P., R.M., X.H. performed mouse experiments. T. Azad, R.R., J.C.B. engineered the viruses. T. Azad, R.R., R.S., A.P. performed data analyses. T. Azad, R.R., R.S. wrote the manuscript. A.G., T. Azad, P.G., T. Alain, M. Ardolina, B.C.D., L.H.T., C.S.I., J.C.B. contributed to the design of studies.

Corresponding author

Correspondence to John C. Bell.

Ethics declarations

Competing interests

We declare that J.C.B. has an interest in Turnstone Biologics, which develops the oncolytic Maraba MG1 virus as an OV platform. The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Zhen Xie and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Peer Review File

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Azad, T., Rezaei, R., Singaravelu, R. et al. Synthetic virology approaches to improve the safety and efficacy of oncolytic virus therapies. Nat Commun 14, 3035 (2023). https://doi.org/10.1038/s41467-023-38651-x

Download citation

Received: 15 December 2022
Accepted: 11 May 2023
Published: 26 May 2023
DOI: https://doi.org/10.1038/s41467-023-38651-x

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.