Sustained inhibition of PARP-1 activity delays glioblastoma recurrence by enhancing radiation-induced senescence

Highlights

• Residual Disease cells of GBM enter a transient senescent state.

• Primary, residual disease, and recurrent GBM show dynamic changes in PARP-1 activity.

• Reduced PARP-1 activity in residual disease cells facilitates their senescent state.

• PARP-1i with IR accelerates & extends senescence state and delays GBM recurrence.

• Olaparib treated senescent cells fail to form tumor in orthotopic GBM mouse model.

Abstract

Glioblastoma (GBM) is the most common primary brain tumor and is highly aggressive with a median survival of 15 months. We have previously shown that residual cells of GBM form multinucleated giant cells (MNGCs) showing a senescent phenotype, but eventually escape from therapy induced senescence (TIS), resulting in GBM recurrence. Here we demonstrate the role of PARP-1 in TIS and its recovery. We show that genetic and pharmacological inhibition of PARP-1 has an anti-proliferative effect on GBM cell lines and primary cultures derived from patient samples. Furthermore, the PARP-1 inhibitor olaparib, in combination with radiation increased MNGCs formation and senescence as assessed by β-galactosidase activity, and macroH2A1 levels in residual cells. Additionally, we found that reduced PARP-1 activity and not protein levels in residual cells was crucial for MNGCs formation and their maintenance in the senescent state. PARP-1 activity was restored to higher levels in recurrent cells that escaped from TIS. Importantly, olaparib + radiation treatment significantly delayed recurrence in vitro as well in vivo in orthotopic GBM mouse models with a significant increase in overall survival of mice. Overall, this study demonstrates that sustained inhibition of PARP-1 activity during radiation treatment significantly delays GBM recurrence.

Introduction

Glioblastoma (GBM) is the most lethal primary brain tumor with a median survival of 15 months. Despite multimodal treatment that includes surgery followed by concomitant treatment with ionizing radiation (IR) and temozolomide, GBM shows therapy resistance and a high rate of recurrence [[1], [2], [3], [4]]. We have earlier shown that radiation induced the formation of multinucleated giant cell (MNGC) from residual GBM cells, and that these MNGCs are the primary cause of relapse [[5], [6], [7]]. Furthermore, we also showed that MNGCs were non proliferative and senescent but were able to overcome senescence to contribute to relapse. Fundamentally, cellular senescence is a tumor suppressive phenomenon [[8], [9], [10]], which therefore, can be exploited for cancer therapeutics. However, as mentioned above, our previous studies on glioblastoma have shown that TIS is transient and residual cells eventually escape senescence giving rise to a highly aggressive mononucleated relapse population [5]. Similar studies on other cancer types like breast, non small cell lung, prostate, colon, B-cell leukemia, and melanomas also show that therapy induced senescent cells can re-enter the cell cycle leading to relapse [[11], [12], [13], [14], [15], [16]]. In fact, a few reports also show that senescent cells can promote a more aggressive tumor phenotype [13,17,18]. Altered mRNA splicing factors, inhibition of mTOR kinase, and overexpression of p21 in the background of p53 mutation, are a few mechanisms known to be involved in reversal from cellular senescence [[19], [20], [21]]. However, a detailed understanding of reversal from therapy induced senescence remains elusive limiting the utility of TIS as a viable cancer therapeutic option.

Poly(ADP-ribose) polymerase-1 (PARP-1) is a nuclear protein having pleiotropic role in various biological and patho-physiological processes, like genome maintenance, DNA repair, transcription, apoptosis, inflammation and cancer [22]. PARP-1 is found to be overexpressed in almost all cancers [23], making it an ideal target for cancer therapy. While currently, there exists one phase I clinical trial for the PARP-1 inhibitor, olaparib in combination with radio-chemo therapy for newly diagnosed GBM patients [24], multiple studies in other cancers also show that PARP-1 acts as a radio and chemo-sensitizer in vitro and in vivo [[25], [26], [27], [28], [29], [30], [31], [32], [33]]. However, none of these studies address the role of PARP-1 in TIS and recovery from senescence.

Here, we recapitulated the clinical setting of GBM resistance in a cellular model derived from GBM cell lines and primary patient samples, to capture residual resistant (RR) cells [[5], [6], [7],34] that are transiently senescent but enter cell cycle to cause aggressive relapse. We found that PARP-1 activity is required for residual cells to escape from TIS. We demonstrate that PARP-1 inhibition in residual GBM cells keeps them non proliferative and senescent for a significantly longer period of time in vitro and in vivo, increasing overall survival in an orthotopic GBM mouse model.