A lentiviral vector-based adenovirus fiber-pseudotyping approach for expedited functional assessment of candidate retargeted fibers

Reovirus Type 3 Dearing Variants Do Not Induce Necroptosis in RIPK3-Expressing Human Tumor Cell Lines

Reoviruses are used as oncolytic viruses to destroy tumor cells. The concomitant induction of anti-tumor immune responses enhances the efficacy of therapy in tumors with low amounts of immune infiltrates before treatment. The reoviruses should provoke immunogenic cell death (ICD) to stimulate a tumor cell-directed immune response. Necroptosis is considered a major form of ICD, and involves receptor-interacting protein kinase 1 (RIPK1), RIPK3 and phosphorylation of mixed-lineage kinase domain-like protein (MLKL). This leads to cell membrane disintegration and the release of damage-associated molecular patterns that can activate immune responses. Reovirus Type 3 Dearing (T3D) can induce necroptosis in mouse L929 fibroblast cells and mouse embryonic fibroblasts. Most human tumor cell lines have a defect in RIPK3 expression and consequently fail to induce necroptosis as measured by MLKL phosphorylation. We used the human colorectal adenocarcinoma HT29 cell line as a model to study necroptosis in human cells since this cell line has frequently been described in necroptosis-related studies. To stimulate MLKL phosphorylation and induce necroptosis, HT29 cells were treated with a cocktail consisting of TNFα, the SMAC mimetic BV6, and the caspase inhibitor Z-VAD-FMK. While this treatment induced necroptosis, three different reovirus T3D variants, i.e., the plasmid-based reverse genetics generated virus (T3D^K), the wild-type reovirus T3D isolate R124, and the junction adhesion molecule-A-independent reovirus mutant (jin-1) failed to induce necroptosis in HT29 cells. In contrast, these viruses induced MLKL phosphorylation in murine L929 cells, albeit with varying efficiencies. Our study shows that while reoviruses efficiently induce necroptosis in L929 cells, this is not a common phenotype in human cell lines. This study emphasizes the difficulties of translating the results of ICD studies from murine cells to human cells.

Induction of Robust Type I Interferon Levels by Oncolytic Reovirus Requires Both Viral Replication and Interferon-α/β Receptor Signaling

Oncolytic viruses are promising agents for cancer therapy because they selectively infect and kill tumor cells, and because they trigger immune responses that can boost anticancer immunity. Key to the latter process is the production of type I interferons (IFN-Is) that can turn noninflamed "cold" tumors into "hot" ones. Besides this desired anticancer effect, IFN-Is are antiviral and successful oncolytic virotherapy thus relies on tightly controlled IFN-I levels. This requires a profound understanding of when and how tumor cells induce IFN-I in response to specific viruses. In this study, we uncovered two key factors that augment IFN-I production in transformed human myeloid cells infected with a tumor-selective reovirus. Viral replication and IFN-α/β receptor (IFNAR) signaling progressively reinforced the levels of IFN-I expressed by infected cells. Mechanistically, both augmented the activation of interferon regulatory factor 3, a key transcription factor for IFNβ expression. Our findings imply that reovirus-permissive tumor cells themselves are a major source of IFN-I expression. As tumors can perturb the IFNAR pathway for their own survival, reovirus-exposed IFNAR-unresponsive tumors may need additional therapeutic intervention to promote the secretion of sufficient IFN-I into the tumor microenvironment. Our increased understanding of the parameters that affect reovirus-induced IFN-I levels could aid in the design of tailored virus-based cancer therapies.

Development of a lentivirus vector-based assay for non-destructive monitoring of cell fusion activity

Cell-to-cell fusion can be quantified by endowing acceptor and donor cells with latent reporter genes/proteins and activators of these genes/proteins, respectively. One way to accomplish this goal is by using a bipartite lentivirus vector (LV)-based cell fusion assay system in which the cellular fusion partners are transduced with a flippase-activatable Photinus pyralis luciferase (PpLuc) expression unit (acceptor cells) or with a recombinant gene encoding FLPe^NLS+, a nuclear-targeted and molecularly evolved version of flippase (donor cells). Fusion of both cell populations will lead to the FLPe-dependent generation of a functional PpLuc gene. PpLuc activity is typically measured in cell lysates, precluding consecutive analysis of one cell culture. Therefore, in this study the PpLuc-coding sequence was replaced by that of Gaussia princeps luciferase (GpLuc), a secretory protein allowing repeated analysis of the same cell culture. In myotubes the spread of FLPe^NLS+ may be limited due to its nuclear localization signal (NLS) causing low signal outputs. To test this hypothesis, myoblasts were transduced with LVs encoding either FLPe^NLS+ or an NLS-less version of FLPe (FLPe^NLS−) and subsequently co-cultured in different ratios with myoblasts containing the FLPe-activatable GpLuc expression cassette. At different times after induction of cell-to-cell fusion the GpLuc activity in the culture medium was determined. FLPe^NLS+ and FLPe^NLS− both activated the latent GpLuc gene but when the percentage of FLPe-expressing myoblasts was limiting, FLPe^NLS+ generally yielded slightly higher signals than FLPe^NLS− while at low acceptor-to-donor cell ratios FLPe^NLS− was usually superior. The ability of FLPe^NLS+ to spread through myofibers and to induce reporter gene expression is thus not limited by its NLS. However, at high FLPe concentrations the presence of the NLS negatively affected reporter gene expression. In summary, a rapid and simple chemiluminescence assay for quantifying cell-to-cell fusion progression based on GpLuc has been developed.

Optimizing reporter constructs for in vivo bioluminescence imaging of interferon-γ stimulated mesenchymal stromal cells

Mesenchymal stromal cells (MSCs) are a promising treatment modality for a variety of diseases. Strategies to investigate the fate of MSCs in vivo are important to unravel their therapeutic mechanisms. However, currently available techniques are hampered by their low sensitivity. We therefore aimed to optimize in vivo bioluminescence imaging of MSCs. We compared MSCs transduced with firefly luciferase (Fluc) and transmembrane-bound Gaussia luciferase driven by the human cytomegalovirus, spleen focus-forming virus (SFFV), and elongation factor 1-α (EF1α) promoters. Although cytomegalovirus-transmembrane-bound Gaussia luciferase-transduced MSCs showed the highest light intensity in vitro, the signal was almost undetectable in vivo. Spleen focus-forming virus-Fluc-transduced MSCs revealed a bright signal in vivo, but transgene expression was silenced upon in vitro stimulation with interferon (IFN)-γ. Therefore, the SFFV promoter was replaced by the EF1α promoter. Light emission of Fluc under the control of EF1α was similar to SFFV-Fluc. Although EF1α-Fluc light emission was decreased tenfold in the presence of IFN-γ when compared with unstimulated MSCs, the bioluminescent signal could still be detected and was clearly distinguishable from untransduced MSCs. Furthermore, stimulation of MSCs with tumor necrosis factor-α hardly affected transgene expression in EF1α-Fluc-transduced MSCs. Thus, the use of the EF1α promoter partially overcomes silencing and allows in vivo bioluminescence imaging of IFN-γ-stimulated MSCs.

A cathepsin-cleavage site between the adenovirus capsid protein IX and a tumor-targeting ligand improves targeted transduction

Human adenoviruses have a great potential as anticancer agents. One strategy to improve their tumor-cell specificity and anti-tumor efficacy is to include tumor-specific targeting ligands in the viral capsid. This can be achieved by fusion of polypeptide-targeting ligands with the minor capsid protein IX. Previous research suggested that protein IX-mediated targeting is limited by inefficient release of protein IX-fused ligands from their cognate receptors in the endosome. This thwarts endosomal escape of the virus particles. Here we describe that the targeted transduction of tumor cells is augmented by a cathepsin-cleavage site between the protein IX anchor and the HER2/neu-binding ZH Affibody molecule as ligand. The cathepsin-cleavage site did not interfere with virus production and incorporation of the Affibody molecules in the virus capsid. Virus particles harboring the cleavable protein IX-ligand fusion in their capsid transduced the HER2/neu-positive SKOV-3 ovarian carcinoma cells with increased efficiency in monolayer cultures, three-dimensional spheroid cultures and in SKOV-3 tumors grown on the chorioallantoic membrane of embryonated chicken eggs. These data show that inclusion of a cathepsin-cleavage sequence between protein IX and a high-affinity targeting ligand enhances targeted transduction. This modification further augments the applicability of protein IX as an anchor for coupling tumor-targeting ligands.

Cell entry and trafficking of human adenovirus bound to blood factor X is determined by the fiber serotype and not hexon:heparan sulfate interaction

Human adenovirus serotype 5 (HAdV5)-based vectors administered intravenously accumulate in the liver as the result of their direct binding to blood coagulation factor X (FX) and subsequent interaction of the FX-HAdV5 complex with heparan sulfate proteoglycan (HSPG) at the surface of liver cells. Intriguingly, the serotype 35 fiber-pseudotyped vector HAdV5F35 has liver transduction efficiencies 4-logs lower than HAdV5, even though both vectors carry the same hexon capsomeres. In order to reconcile this apparent paradox, we investigated the possible role of other viral capsid proteins on the FX/HSPG-mediated cellular uptake of HAdV5-based vectors. Using CAR- and CD46-negative CHO cells varying in HSPG expression, we confirmed that FX bound to serotype 5 hexon protein and to HAdV5 and HAdV5F35 virions via its Gla-domain, and enhanced the binding of both vectors to surface-immobilized hypersulfated heparin and cellular HSPG. Using penton mutants, we found that the positive effect of FX on HAdV5 binding to HSPG and cell transduction did not depend on the penton base RGD and fiber shaft KKTK motifs. However, we found that FX had no enhancing effect on the HAdV5F35-mediated cell transduction, but a negative effect which did not involve the cell attachment or endocytic step, but the intracellular trafficking and nuclear import of the FX-HAdV5F35 complex. By cellular imaging, HAdV5F35 particles were observed to accumulate in the late endosomal compartment, and were released in significant amounts into the extracellular medium via exocytosis. We showed that the stability of serotype 5 hexon:FX interaction was higher at low pH compared to neutral pH, which could account for the retention of FX-HAdV5F35 complexes in the late endosomes. Our results suggested that, despite the high affinity interaction of hexon capsomeres to FX and cell surface HSPG, the adenoviral fiber acted as the dominant determinant of the internalization and trafficking pathway of HAdV5-based vectors.

Directed adenovirus evolution using engineered mutator viral polymerases

Adenoviruses (Ads) are the most frequently used viruses for oncolytic and gene therapy purposes. Most Ad-based vectors have been generated through rational design. Although this led to significant vector improvements, it is often hampered by an insufficient understanding of Ad’s intricate functions and interactions. Here, to evade this issue, we adopted a novel, mutator Ad polymerase-based, ‘accelerated-evolution’ approach that can serve as general method to generate or optimize adenoviral vectors. First, we site specifically substituted Ad polymerase residues located in either the nucleotide binding pocket or the exonuclease domain. This yielded several polymerase mutants that, while fully supportive of viral replication, increased Ad’s intrinsic mutation rate. Mutator activities of these mutants were revealed by performing deep sequencing on pools of replicated viruses. The strongest identified mutators carried replacements of residues implicated in ssDNA binding at the exonuclease active site. Next, we exploited these mutators to generate the genetic diversity required for directed Ad evolution. Using this new forward genetics approach, we isolated viral mutants with improved cytolytic activity. These mutants revealed a common mutation in a splice acceptor site preceding the gene for the adenovirus death protein (ADP). Accordingly, the isolated viruses showed high and untimely expression of ADP, correlating with a severe deregulation of E3 transcript splicing.

Enhanced transduction of CAR-negative cells by protein IX-gene deleted adenovirus 5 vectors

In human adenoviruses (HAdV), 240 copies of the 14.3-kDa minor capsid protein IX stabilize the capsid. Three N-terminal domains of protein IX form triskelions between hexon capsomers. The C-terminal domains of four protein IX monomers associate near the facet periphery. The precise biological role of protein IX remains enigmatic. Here we show that deletion of the protein IX gene from a HAdV-5 vector enhanced the reporter gene delivery 5 to 25-fold, specifically to Coxsackie and Adenovirus Receptor (CAR)-negative cell lines. Deletion of the protein IX gene also resulted in enhanced activation of peripheral blood mononuclear cells. The mechanism for the enhanced transduction is obscure. No differences in fiber loading, integrin-dependency of transduction, or factor-X binding could be established between protein IX-containing and protein IX-deficient particles. Our data suggest that protein IX can affect the cell tropism of HAdV-5, and may function to dampen the innate immune responses against HAdV particles.