Starvation-Induced Differential Virotherapy Using an Oncolytic Measles Vaccine and Vaccinia Virus

Abstract

Colorectal carcinoma (CRC) is today’s third most common cause of cancer related death worldwide. Even though considerable advances in prevention, early detection and treatment options have been obtained in the last decades, metastatic CRC still implies very poor prognosis.

Starvation (Fasting) has been shown to sensitize tumor cells to chemotherapy whilst protecting normal cells at the same time [Differential Stress Resistance, (DSR)]. Upregulation of pro-survival and proliferation pathways due to various mutations prevents cancer cells from responding to external growth factors. Consequently, malignant cells, unable to adapt to extreme environmental conditions such as absence of nutrients, are becoming more vulnerable to stress. Normal cells, in contrast, enter a standby mode in response to starvation and are getting more protected.

The ability of oncolytic virotherapeutics (OVs) to selectively infect, replicate in and lyse cancer cells outlines a promising approach in cancer therapy. However, combinatorial concepts seem to be needed to achieve sustained anti-cancer effects, such as combination of OVs with chemotherapy or new immunomodulatory drugs.

We hypothesized that starvation would increase the oncolytic potential of OVs in CRC cell lines and protect normal colon cells against virus-mediated cell lysis.

Three different human colon carcinoma cell lines (HT-29, HCT-15 and HCT-116) as well as two human normal colon cell lines (CCD-18 Co and CCD-841 CoN) were subjected to various starvation regimes in glucose and/or serum restricted cell culture medium and infected with two state-of-the-art OVs [i.e., measles vaccine virus (MeV) and vaccinia virus (GLV-1h68)].

Fasting regimes applied were either short-term starvation (24 h pre-infection) or long-term starvation (24 h pre- and 96 h post-infection).

We used cell viability assays to determine the cell killing capabilities of i) virotherapy, ii) starvation, and iii) the combination of these two. Virus growth curves were generated to assess the replication of MeV in starved and non-starved HT29 cells.

As a result, starvation retarded cell growth in all cell lines in a time and concentration dependent manner. Infection of starved cancer cells exhibited additional oncolytic potential of virotherapy plus starvation for most combinations, indicating that virus-mediated oncolysis is sufficiently working under starvation conditions.

Remarkably, long-term standard glucose, low-serum starvation potentiated the efficacy of MeV-mediated cell killing in HT-29 cancer cells, whereas it was decreased in normal colon cells CCD-18 Co and CCD-841 CoN. Interestingly, viral replication of MeV in HT-29 was decreased in long-term starved cells, but was increased after short-term low-glucose, low-serum starvation.

We speculate that particular nutrient signaling pathways such as the PI3K/ Akt/ mTOR pathway are modified upon fasting depending on specific mutations in cancer cells resulting in a differential response of distinct CRC cells to OVs.

In conclusion, starvation based virotherapy could enhance the oncolytic effect on CRC in future anti-cancer therapy while protecting normal tissues from side effects.