Genome-Wide Expression and Anti-Proliferative Effects of Electric Field Therapy on Pediatric and Adult Brain Tumors

Abstract

The lack of treatment options for high-grade brain tumors has led to searches for alternative therapeutic modalities. Electrical field therapy is one such area. The Optune™ system is an FDA-approved novel device that delivers continuous alternating electric fields (tumor treating fields-TTFields) to the patient for the treatment of primary and recurrent Glioblastoma multiforme (GBM). Various mechanisms have been proposed to explain the effects of TTFields and other electrical therapies. Here, we present the first study of genome-wide expression of electrotherapy (delivered via TTFields or Deep Brain Stimulation (DBS)) on brain tumor cell lines. The effects of electric fields were assessed through gene expression arrays and combinational effects with chemotherapies. We observed that both DBS and TTFields significantly affected brain tumor cell line viability, with DBS promoting G0-phase accumulation and TTFields promoting G2-phase accumulation. Both treatments may be used to augment the efficacy of chemotherapy in vitro. Genome-wide expression assessment demonstrated significant overlap between the different electrical treatments, suggesting novel interactions with mitochondrial functioning and promoting endoplasmic reticulum stress. We demonstrate the in vitro efficacy of electric fields against adult and pediatric high-grade brain tumors and elucidate potential mechanisms of action for future study.

Keywords: Deep Brain Stimulation; ependymoma; glioma; tumor treating fields.

Figure 1

The impact of manipulations of frequency (kHz) on the viability of GBM, medulloblastoma and ependymoma cell lines. (A) Cell lines were treated for 72 h over a range of frequencies with the TTFields, with metabolic measurements being taken at the 72 h endpoints. Each cell line demonstrates variable efficacy in TTFields treatment, with no clear pattern of a most optimal frequency for each tumor type tested. (B) Cell lines were treated for 72 h at their determined most efficacious frequency (Figure 1A) with the Inovitro, with cell counts being taken at 72 h time-points. Each cell line was significantly affected by TTFields, with variability in efficacy being present between cell lines and tumor types. (C) Human neural stem cells and astrocyte cell lines were treated for 72 h at 200 kHz and 400 kHz TTFields with the Inovitro, with cell counts being taken at the 72 h endpoint. The actively dividing neural stem cells were significantly affected by the electric fields, while the non-dividing astrocytes were not affected by either treatment. **** = p value < 0.01.

Figure 2

The impact of manipulations of intensity (V) and frequency (Hz) on the viability of commercial and primary GBM cell lines. (A) U87-MG, KNS42, GIN-5 and GIN-31 cell lines were treated for 7 days with 130 Hz/450 µs electric fields over a range of intensities, with metabolic measurements being taken at Day 0, 1, 5 and 7 time-points. A clear positive correlation between intensity and duration of treatment is present across all the cell lines tested. (B) U87-MG and KNS42 cell lines were treated for 7 days with either 5 V/450 µs electric fields over a range 60–190 Hz or 1 V/450 and 500–1000 Hz frequencies, with metabolic measurements being taken at Day 0, 1, 5 and 7 time-points. Manipulations of frequency between 60–190 Hz demonstrated influence over metabolic activity, while frequencies of 500–1000 Hz had no effect at the tested intensities. (C) Human neural stem cells and astrocyte cell lines were treated for 7 days with 10 V/130 Hz/450 µs electric fields, with cell counts being taken at the Day 7 endpoint. The actively dividing neural stem cells were significantly affected by the electric fields, while the non-dividing astrocytes were not affected by the treatment.

Figure 3

The impact of electric fields on the cell cycle of brain tumor cell lines. (A) U87-MG, KNS42 and SF188 cell lines were treated for 5 days with 10 V/130 Hz/450 µs DBS electric fields, with flow cytometry being performed with PI staining to assess cell cycling. The electric field treatments caused significant accumulation of cells in the G0-phase, depletion of cells in the S-phase and minimal accumulation of cells in the subG0-phase. (B) Cell lines were treated for 72 h at the determined optimal frequency TTFields (Inovitro), with flow cytometry being performed with PI staining to assess cell cycling. The effects of TTFields treatments were different between cell lines, but the most common response of treatment was significant accumulation of cells in the G2-phase and S-phase and minimal accumulation of cells in the subG0-phase. * = p < 0.05 ** = p < 0.01 *** = p < 0.005 **** = p < 0.001.

Figure 4

The impact of the combination of DBS electric fields and mitotic inhibitors on the viability and cell cycling of GBM cell lines. (A) U87-MG, KNS42 and SF188 cell lines were treated for 5 days with 10 V/130 Hz/450 µs electric fields in combination with a higher and lower dose of paclitaxel, with metabolic measurements being taken at the endpoint. An increase in efficacy of the higher and lower doses of paclitaxel was achieved with the addition of electric fields. (B) U87-MG, KNS42 and SF188 cell lines were treated for 5 days with a higher and lower dose paclitaxel, with flow cytometry being performed with PI staining to assess cell cycling. The combination with higher dose paclitaxel caused significant accumulation of cells in the subG0-phase, while the combination with lower dose paclitaxel did not cause differential accumulations of cells relative to electric fields alone. ** = p < 0.01 *** = p < 0.005 **** = p < 0.001.

Figure 5

The impact of the combination of DBS electric fields and TTFields with mitotic inhibitors on the viability of brain tumor cell lines. Y axis indicates relative viability compared to untreated cells (0 Hz red column). (A) SF188, UW228-3 and BXD-1425EPN cell lines were treated for 3 days with TTFields in combination with a higher (5 nM) and lower (0.5 nM) dose of paclitaxel, with metabolic measurements being taken at the endpoint. (B) U87-MG, KNS42, SF188 and GIN-31 cell lines were treated for 5 days with 10 V/130 Hz/450 µs electric fields and a higher (1.25 μM) and lower (0.125 μM) dose mebendazole, with metabolic measurements being taken at the endpoint. (C) SF188, UW228-3 and BXD-1425EPN cell lines were treated for 3 days with TTFields in combination with a higher (1.25 μM) and lower (0.25 μM) dose of mebendazole, with metabolic measurements being taken at the endpoint. (D) U87-MG, GIN-28, GIN-31 and GCE-77 cell lines were treated for 5 days with 10 V/130 Hz/450 µs electric fields and a higher (5 μM) and lower (0.5 μM) dose TMZ, with metabolic measurements being taken at the endpoint. ** = p < 0.01 *** = p < 0.005 **** = p < 0.001.

Figure 6

The impact of electric fields on gene expression. (A) KNS42 and GIN-31 cell lines were treated for 5 days with 10 V/130 Hz/450 µs electric fields, and gene expression changes were assessed via Clariom™ S Assay arrays (significance criteria: FDR < 0.05 and fold-change <−2 or >2). (B) KNS42 and GIN-31 cell lines were treated for 72 h at the determined optimal frequency TTFields (Inovitro), and gene expression changes were assessed via Clariom™ S Assay arrays (significance criteria: FDR < 0.05 and fold-change <−2 or >2). (C) Western blot of MT-ND5, BTNL9 and GAPDH as the control. Target genes were among the top 50 most significantly differentially expressed genes for both treatments. The positive controls were breast carcinoma MCF-7 cells and the housekeeping gene chosen was GAPDH. Protein was extracted from samples following 5 days of electrical treatment and 5 days of continuous growth for the untreated samples. Western blot validates downregulation of MT-ND5 and CTSB and the upregulation of BTNL9 following electrical treatment. KEY: G0 = Gin-31 0 Hz sham treated; G10 = GIN-31 130 Hz treated; G200 = GIN-31 200 kHz treated; K0 = KNS42 0 Hz sham treated; K10 = KNS42 130 Hz treated; K200 = KNS42 200 kHz treated.