Single Amino Acid Differences between Closely Related Reovirus T3D Lab Strains Alter Oncolytic Potency In Vitro and In Vivo

Abstract

Little is known about how genetic variations in viruses affect their success as therapeutic agents. The type 3 Dearing strain of Mammalian orthoreovirus (T3D) is undergoing clinical trials as an oncolytic virotherapy. Worldwide, studies on reovirus oncolysis use T3D stocks propagated in different laboratories. Here, we report that genetic diversification among T3D stocks from various sources extensively impacts oncolytic activity. The T3D strain from the Patrick Lee laboratory strain (TD3PL) showed significantly stronger oncolytic activities in a murine model of melanoma than the strain from the Terence Dermody laboratory (T3DTD). Overall in vitro replication and cytolytic properties of T3D laboratory strains were assessed by measuring virus plaque size on a panel of human and mouse tumor cells, and results were found to correlate with in vivo oncolytic potency in a melanoma model. T3DPL produced larger plaques than T3DTD and than the T3D strain from the ATCC (T3DATCC) and from the Kevin Coombs laboratory (T3DKC). Reassortant and reverse genetics analyses were used to decipher key genes and polymorphisms that govern enhanced plaque size of T3DPL. Five single amino acid changes in the S4, M1, and L3 genome segments of reovirus were each partially correlated with plaque size and when combined were able to fully account for differences between T3DPL and T3DTD. Moreover, polymorphisms were discovered in T3DTD that promoted virus replication and spread in tumors, and a new T3DPL/T3DTD hybrid was generated with enhanced plaque size compared to that of T3DPL. Altogether, single amino acid changes acquired during laboratory virus propagation can have a large impact on reovirus therapeutic potency and warrant consideration as possible confounding variables between studies. IMPORTANCE The reovirus serotype 3 Dearing (T3D) strain is in clinical trials for cancer therapy. We find that closely related laboratory strains of T3D exhibit large differences in their abilities to replicate in cancer cells in vitro, which correlates with oncolytic activity in a in a murine model of melanoma. The study reveals that five single amino acid changes among three reovirus genes strongly impact reovirus therapeutic potency. In general, the findings suggest that attention should be given to genomic divergence of virus strains during research and optimization for cancer therapy.

Closely related reovirus lab strains induce opposite expression of RIG-I/IFN-dependent versus -independent host genes, via mechanisms of slow replication versus polymorphisms in dsRNA binding σ3 respectively

“…Over the course of propagation in different laboratories, these strains have accumulated 20 amino acid polymorphisms. Surprisingly, despite only small genomic divergence, the T3D lab strains exhibit drastic differences in their ability to replicate in tumorigenic cells in vitro, and exhibit oncolytic activities in vivo [19,20]. Classical and reverse genetics approaches previously revealed that five polymorphisms dispersed among S4, M1, and L3 reovirus genes account for the oncolytic differences between T3D strains.…”

“…Together the data supports the following model: T3D TD induces more IFN signalling than T3D PL because of slower replication kinetics, rather than that IFN signalling being the cause of reduced T3D TD replication kinetics. In other words, T3D PL inherently replicates better than T3D TD through mechanisms independent of IFN; some of these mechanisms were previously described and include inherently better transcription activity by T3D PL particles [19,20]. However, since IFNs can suppress cell-cell spread, the induction of IFNs by T3D TD further suppresses dissemination of this T3D stain, producing an even-larger differential between T3D TD and T3D PL in IFN-proficient cells.…”

“…kinetics of T3D TD [19,20] is evident by quantifying the number of reovirus-antigen-positive cells by flow cytometric analysis at 12hpi (intermediate) versus 24hpi (late) stages of replication; T3D TD required 24 hours to achieve the level of infection seen by T3D PL at 12hpi ( Fig 1C). Virus protein expression and IFN signalling were assessed at 12hpi, an intermediate timepoint of virus replication when the potentially confounding effects of cell-cell virus spread are minimal.…”

“…As described in the introduction, despite only few polymorphisms between reovirus T3D lab strains, the viruses exhibit gross differences in their oncolytic activity in vivo and single-step growth kinetics in a panel of cancer cells [19,20]. Previous comparisons revealed that T3D lab strains possess inherent cell-independent differences, for example in the rate of transcription in a cell-free system.…”

“…We previously found that polymorphisms in S4, M1 and L3 genes were dominantly responsible for the increased replication of T3D PL in tumorigenic cells [19]. We wondered if S4, M1, and/or L3 were determinants of differential IFN production and response between the T3D strains.…”