Mutant Isocitrate Dehydrogenase 1 Expression Enhances Response of Gliomas to the Histone Deacetylase Inhibitor Belinostat

Abstract

Histone deacetylase inhibitors (HDACis) are drugs that target the epigenetic state of cells by modifying the compaction of chromatin through effects on histone acetylation. Gliomas often harbor a mutation of isocitrate dehydrogenase (IDH) 1 or 2 that leads to changes in their epigenetic state presenting a hypermethylator phenotype. We postulated that glioma cells with IDH mutation, due to the presence of epigenetic changes, will show increased sensitivity to HDACis. This hypothesis was tested by expressing mutant IDH1 with a point alteration-converting arginine 132 to histidine-within glioma cell lines that contain wild-type IDH1. Glioma cells engineered to express mutant IDH1 produced D-2-hydroxyglutarate as expected. When assessed for response to the pan-HDACi drug belinostat, mutant IDH1-expressing glioma cells were subjected to more potent inhibition of growth than the corresponding control cells. Increased sensitivity to belinostat correlated with the increased induction of apoptosis. Finally, a phase I trial assessing the addition of belinostat to standard-of-care therapy for newly diagnosed glioblastoma patients included one patient with a mutant IDH1 tumor. This mutant IDH1 tumor appeared to display greater sensitivity to the addition of belinostat than the other cases with wild-type IDH tumors based on both standard magnetic resonance imaging (MRI) and advanced spectroscopic MRI criteria. These data together suggest that IDH mutation status within gliomas may serve as a biomarker of response to HDACis.

Keywords: HDAC; IDH; glioblastoma; sMRI.

Figure 1

Immunoblots of lysates from the NHA and LN229 cell-line pairs (vector control (C) and mtIDH-expressing (M)) are shown. Upper blots were probed for the IDH1 R132H mutant, while the lower blots show detection of eIF5 as a loading control.

Figure 2

Immunoblots of lysates from the (A) NHA and (B) L229 cell-line pairs are shown. For each cell line, treatment was without the drug (0 μM) or increasing levels of belinostat (0.1, 0.3, 1.0, 3.0, and 10 μM) indicated above the lanes, with an ascending triangle signifying the progressively higher dose levels. Each group was probed for acetylated α-tubulin (Ac-α-tubulin) (upper blots) and acetylated Histone H4 (Ac-Hist H4) (middle blots). The lower blots show detection of GAPDH as a loading control.

Figure 3

Assessments of cell growth after belinostat treatment were performed using either direct cell counting (A,B) or MTT assay (C,D). Results for control (solid lines) and mtIDH1 (dotted lines) cells were obtained in duplicate, with average values normalized as described in Methods. Error bars are ± one standard deviation for each data point. Significant p-values based on Student’s t-test and percent reduction ranges of the values of specific data points of mtIDH1-expressing versus control cells are shown on the graphs.

Figure 4

Assessment of apoptosis after belinostat treatment was performed by measuring either caspase activity (A,B) or fraction annexin V staining (C). Results for control (light bars) and mtIDH1 (dark bars) were obtained in duplicate, with average values normalized as described in Methods. Error bars are ± one standard deviation for each value. Significant p-values based on Student’s t-test for specific data points of mtIDH1-expressing versus control cells are shown on the graphs.

Figure 5

Assessment of sMRI response in GBM patients treated on our belinostat trial: (A) Waterfall plot for percent change in residual tumor between the pre-RT and 4-week post-RT sMRI scans, defined as volume of CNI values greater than 2. Nine patients had assessable sMRIs at both time points, comprising eight wild-type IDH GBMs (dark bars) and one mutant IDH1 GBM (light bar). Numbers above bars indicate the value of the pre-RT residual tumor volume. (B) The pre-RT and 4-week post-RT CNI maps for indicated slices on the 3D renderings for volumes of CNI values greater than 2 are shown for two wild-type IDH GBMs and one mutant IDH1 GBM. Arrows on the graph indicate the individual cases that are shown in this section.

Figure 6

Serial standard MRIs (T1 post and FLAIR sequences) are shown for the same three patients illustrated in Figure 5B. Time points are as indicated, with the last time point being the scan where progression was officially called. One time point (upper panel, pre-RT) shows a T2 image as an alternative to a FLAIR image because a FLAIR sequence was not obtained at that time point.