Lung cancer is the leading cause of cancer mortality in both men and women in the United States  and all cancer deaths worldwide . The most common form of lung cancer is NSCLC that includes squamous cell carcinoma, adenocarcinoma, and large cell carcinoma. Despite the tremendous efforts and progress in lung cancer research, treatment outcomes for non-localized NSCLC remain poor . New treatment strategies are urgently needed to improve survival for advanced NSCLC patients. In the current study, we uncovered a potent oncolytic activity of ReoT3D against a panel of human NSCLC cell lines, in particular, NSCLC cell lines of adenocarcinoma or large cell carcinoma origin. The susceptibility of cancer cells to ReoT3D-mediated cytolysis has been attributed to increased Ras activity [11, 12]. However, we did not observe any significant association between the ReoT3D-permissibility and the presence of Ras-activating gene mutations or activated Ras in human NSCLC cells. The lack of association has also been reported by others in human colon cancers . It is possible that in addition to the activation status of Ras-associated pathways [11, 12], there are other molecular determinants of ReoT3D-sensitivity, such as the cell surface density of putative ReoT3D receptors/coreceptors [27–29] and intracellular virion uncoating processes [30, 31], all of which can affect ReoT3D infection efficiency.
The combination effects of herpesvirus or adenovirus-based oncolytic viral vectors and chemotherapeutic agents have previously been evaluated against different human cancers [32–37]. Synergistic activity was reported in the majority of these studies. However, combination regimens selected for the previous studies were mostly limited in scope in terms of dose range and the number of chemotherapeutic agents investigated. In the current study, we demonstrated that the oncolytic activity of ReoT3D against NSCLC cells could be significantly potentiated by a number of chemotherapeutic agents used in the treatment of NSCLC, including paclitaxel, cisplatin, gemcitabine and vinblastine. Combination analysis based on the Chou-Talalay's method [16, 17] clearly showed significant levels of synergy between ReoT3D and each chemotherapeutic agent tested. Interestingly, we found that the drug sensitivity of each NSCLC cell line was an important determinant for the in vitro synergistic effect of ReoT3D-chemotherapeutic combination regimens, with the exception of ReoT3D-paclitaxel combination. It is conceivable that certain molecular changes conferring drug resistance can antagonize the process of ReoT3D-mediated cell killing. Our data, therefore, caution against the use of chemotherapeutic agents combined with ReoT3D for the treatment of NSCLC that have developed resistance to the agents. In contrast, the level of ReoT3D sensitivity did not appear to compromise the combination effects in NSCLC cells. Rather, the addition of chemotherapeutic agents may help accelerate ReoT3D-induced cell death process, which is otherwise slowed in NSCLC cells with low susceptibility to ReoT3D infection.
The most intriguing finding from our study was the synergistic effect of ReoT3D-paclitaxel combination consistently observed in all the NSCLC cell lines examined, regardless of the level of sensitivity to the compound. Because previous studies of oncolytic virus-chemotherapeutic combinations, in particular with paclitaxel, did not address the impact of drug resistance on the combination effects, we cannot ascertain whether our finding is unique to reovirus-containing combination therapy. Mammalian reoviruses are known to exploit microtubules for the formation of viral replication complexes (inclusion bodies) . Based on our initial findings that the addition of paclitaxel to ReoT3D significantly increased the level of progeny virion production from all the NSCLC cell lines tested, we speculated that microtubule-stabilizing paclitaxel might have enhanced reoviral replication, resulting in a more efficient and synergistic oncolytic effect. However, the increased progeny virion production was not necessarily a unique outcome of ReoT3D-paclitaxel combination, but was also observed with ReoT3D-vinblastine combination in vinblastine-resistant NCI-H322M cells, where the combination of ReoT3D and vinblastine was found to be strongly antagonistic. Moreover, the addition of gemcitabine to ReoT3D treatment was not associated with an increased progeny virion production, regardless of the combination effects (synergy or antagonism) attained. These data suggested that the synergistic effect of ReoT3D and chemotherapeutic agents was not the direct result of enhanced lytic cell death, but more likely the manifestation of accelerated programmed cell death, which was triggered before virion assembly and release.
Reovirus has been shown to induce apoptotic cell death in a variety of cell types, including cancer cells [18, 39]. Indeed, increased caspase activity and apoptotic cleavage of PARP were readily detectable in ReoT3D-treated NSCLC cells within 24 hours in the current study, as has been shown in ReoT3A-exposed cancer cells . The combination treatment with ReoT3D and paclitaxel led to more robust caspase activation than ReoT3D alone with a significant leftward shift of the dose response curve in both paclitaxel-sensitive and -resistant NSCLC cell lines, suggesting that the enhanced apoptosis most likely constituted the synergistic cell killing by the combination, regardless of the level of paclitaxel sensitivity. To gain more insight into the mechanistic basis of accelerated apoptosis associated with the ReoT3D-paclitaxel combination, we examined the effects of ReoT3D and paclitaxel on caspase-3 activation in relation to cell cycle progression. While taxanes and other antimicrotubule agents are known to activate the spindle checkpoint and induce mitotic arrest , ReoT3D infection has been shown to effect cell cycle arrest at G1/S and G2/M [19, 20]. Because arrests in cell cycle progression induced by anticancer agents are commonly followed by apoptosis , we hypothesized that these two agents with differing effects on cell cycle progression may have synergistically activated apoptotic pathways in dually treated NSCLC cells.
We found that the proportion of cells expressing activated caspase-3 was significantly increased by the ReoT3D-paclitaxel combination as compared to either ReoT3D or paclitaxel single treatment in each NSCLC cell line tested. Interestingly, the activation of caspase-3 was found more prominent in post-G1 cell population with ≥ 4N DNA content. Escape from mitotic arrest induced by spindle poisons such as taxanes and other antimicrotubule agents is commonly observed in cancer cells with impaired (weakened) mitotic checkpoint [21, 41]. These cells that prematurely exit mitosis without proper cell division form large multinucleated cells with DNA content of 4N or greater, as demonstrated by EM in our study. After such mitotic slippage, some cells may undergo p53-dependent apoptosis, while others survive through senescence or continuing cell cycle (endoreduplication) , depending on the functions of p53, MAP kinase pathways, and p21-activated kinase [43, 44]. In the current study, we found that NSCLC cells could efficiently escape from mitotic arrest induced by paclitaxel at 0.1 ~1 μM and survive at least for the first 24 hours of exposure, with the exception of NCI-H23 cells that were more prone to apoptosis upon paclitaxel exposure than 3 other NSCLC cell lines examined. While paclitaxel-treated NCIH460 cells ultimately underwent dramatic cell death after 48 hours, EKVX and NCI-H322M demonstrated considerable levels of resistance to paclitaxel-induced CPE. Nonetheless, the combination of ReoT3D and paclitaxel consistently accelerated the apoptotic process in post-G1 cells, including in paclitaxelresistant EKVX and NCI-H322M cells. This enhanced apoptosis appeared to have resulted from prolonged mitotic arrest, as corroborated by the EM data demonstrating that treatment with the ReoT3D-paclitaxel combination resulted in increased numbers of both mitotically arrested and apoptotic cells while decreasing the number of multinucleated cells as compared to paclitaxel alone.
The molecular mechanism of prolonged mitotic arrest induced by the ReoT3D-paclitaxel combination has yet to be elucidated. It is possible that ReoT3D infection may enhance the mitotic checkpoint activity in cancer cells with weakened mitotic checkpoint, for example, by upregulating the expression of mitotic checkpoint proteins (such as Mad1, Mad2, BubR1/Mad3, Bub1 and Bub3) , Cdk1 and/or cyclin B , or suppressing the anaphase-promoting complex/cyclosome activity . Such ReoT3D-induced alterations in the mitotic regulatory network may reinforce the mitosis-arresting signal of taxanes, leading to prolonged mitotic arrest and apoptosis. Better understanding of the molecular consequences of ReoT3D infection on cell cycle checkpoint function and apoptotic signaling pathways in cancer cells will greatly enhance our ability to design rational combination therapies with proapoptotic oncolytic agent, ReoT3D, and various classes of anticancer agents. Concurrently, it is also of high importance to investigate potential consequences of ReoT3D-chemotherapeutic combinations on normal tissues in order to identify undesirable combination regimens that are associated not only with synergistic oncolytic activity, but also with enhanced toxicity in humans.