|Home | About | Journals | Submit | Contact Us | Français|
Atypical teratoid/rhabdoid tumor (ATRT) is a highly malignant central nervous system neoplasm that primarily occurs in children less than 3 years of age. Because of poor outcomes with intense and toxic multimodality treatment, new therapies are urgently needed. Histone deacetylase inhibitors (HDIs) have been evaluated as novel agents for multiple malignancies and have been shown to function as radiosensitizers. They act as epigenetic modifiers and lead to re-expression of inappropriately repressed genes, proteins, and cellular functions. Because of the underlying chromatin remodeling gene mutation in ATRT, HDIs are ideal candidates for therapeutic evaluation. To evaluate the role of HDIs against ATRT in vitro, we assessed the effect of drug treatment on proliferation, apoptosis, and gene expression. In addition, we examined HDI pretreatment as a radiosensitization strategy for ATRT. 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium with phenazine methosulfate electron coupling reagent (MTS) and clonogenic assays demonstrated that HDI treatment significantly reduces the proliferative capacity of BT-12 and BT-16 ATRT cells. In addition, the HDI SNDX-275 was able to induce apoptosis in both cell lines and induced p21Waf1/Cip1 protein expression as measured by Western blot. Evaluation of differential gene expression by microarray and pathway analysis after HDI treatment demonstrated alterations of several key ATRT cellular functions. Finally, we showed that HDI pretreatment effectively potentiates the effect of ionizing radiation on ATRT cells as measured by clonogenic assay. Our findings suggest that the addition of HDIs to ATRT therapy may prove to be beneficial, especially when administered in combination with current treatment modalities, such as radiation.
Atypical teratoid/rhabdoid tumor (ATRT) is a malignant central nervous system (CNS) tumor that was initially distinguished from other pediatric embryonal brain tumors in the mid-1980s.1,2 It most commonly affects young children and historically carries a poor outcome.3–5 Therapy for this tumor is frequently multimodality in nature, involving surgery, radiation, and chemotherapy.5 In the past 10 years, therapy has either focused on high-dose chemotherapy with autologous stem cell rescue or, more recently, use of a regimen based on sarcoma therapy.6,7 Despite an increase in overall survival with these approaches, therapy related toxicity frequently remains a critical problem in this young age group. ATRT is frequently associated with a mutation or deletion of the hSNF5/INI1/SMARCB1 gene found on chromosome 22q.8,9 SMARCB1 functions as a tumor suppressor, and its absence has been shown to be associated with increased markers of cell cycle progression.10 Because SMARCB1 is a member of the SWI/SNF chromatin remodeling complex, therapeutic agents, such as histone deacetylase inhibitors (HDIs), that affect chromatin structure may prove to be effective in ATRT.
In recent years, the mechanisms of HDIs have been evaluated in multiple models of malignancies and, because of in vitro and in vivo efficacy, have begun to emerge as promising therapies.11–16 Thus far, suberoylanilide hydroxamic acid (SAHA; vorinostat) and romidepsin (depsipeptide) have been Food and Drug Administration approved for the treatment of cutaneous T-cell lymphoma; however, multiple HDIs are currently in development.17,18
HDIs function by increasing acetylation of histones at lysine residues, thus blocking the physiologic role of histone deacetylase (HDAC). Increased acetylation is also seen with nonhistone proteins.19 Histone acetylation permits a more open chromatin configuration, thus allowing more access for transcription factors and transcriptional molecules. Several antiproliferative cellular functions are often among those derepressed; this includes processes such as growth arrest, apoptosis, and cell cycle inhibition. In addition, HDI treatment is associated with decreased metastatic potential and angiogenesis.16
Because of the role of HDIs as promoters of cell cycle inhibition,16 ATRT is a particularly interesting model for evaluation of these agents because of its association with increased cyclin D1 expression.10 In addition, because ATRT is a tumor that frequently requires radiation as part of its therapy, HDIs are of particular interest, as previous literature has discussed their role in radiosensitization of other types of malignancies.20–23
Previous studies24–27 have analyzed some aspects of rhabdoid tumor treatment with HDIs, but to date, none have examined the capacity of HDIs to alter the effect of radiation in ATRT. In this study, we evaluated the HDI-induced alteration of in vitro proliferative capacity of ATRT cell lines and a primary patient sample. In addition, we examined the effect of HDI treatment on apoptosis, tumor suppressor protein expression, and differential gene expression. Finally, the ability of HDI treatment to potentiate the effect of radiation on ATRT cell lines was studied.
BT-12 and BT-16 ATRT cells were kindly provided as a gift from Dr. Peter Houghton's laboratory (St. Jude Children's Research Hospital). Cells were cultured in RPMI-1640 (Invitrogen) supplemented with 10% fetal bovine serum according to the supplier's recommendations. Primary, short-term cell culture (UPN737) was derived from an immediate postsurgical specimen under a protocol approved by the institutional review board at the University of Colorado. Primary cell culture was generated as previously described.28 Cultures were maintained at low passage number (p2-p4) in RPMI-1640 supplemented with 20% fetal bovine serum. Cell lines were propagated at 37°C in 5% CO2 with humidity.
The HDIs trichostatin A (TsA), SAHA, and SNDX-275 (Sigma) were diluted in DMSO to appropriate concentrations.
The effect of HDIs on BT-12, BT-16, and UPN737 cells was analyzed with Cell Titer AQueous (Promega) 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium with phenazine methosulfate electron coupling reagent (MTS) assay as previously described.29 In brief, cells were plated in triplicate in 96-well plates for 24 h. Subsequently, TsA, SAHA, or SNDX-275 was added in 2-fold serial dilutions at 4 nM–1 μM (TsA) and 125 nM–128 μM (SAHA and SNDX-275). DMSO was added to control wells. After 72 h, 20 μL of MTS reagent was added to each well and the absorbance measured at 490 nm using a microplate reader (BioTek). Relative cell number was calculated by normalizing the absorbance of HDI-treated samples to controls; 50% inhibitory concentrations (IC50) and 25% IC (IC25) were calculated using GraphPad Prism software.
BT-12 and BT-16 cells were seeded on a 10 cm2 culture plate (300 000 cells/plate) for 24 h. SNDX-275 was then added to separate plates at concentrations equivalent to IC50 (per MTS assay analysis), IC25, and 2-fold serial dilutions downward from IC25. Controls were treated with DMSO. SNDX-275 or DMSO was removed after 48 h, and the cells were collected. Viable cells were counted by trypan blue exclusion. Cells were then seeded in triplicate per treatment condition into 6-well plates (400 cells/well) for clonogenic assay and allowed to grow in culture medium. After sufficient colony formation was demonstrated in control wells, cells were washed with phosphate-buffered saline (PBS) and then stained with crystal violet. Colonies were defined as those with more than 50 cells and were then counted. Analysis was performed with GraphPad Prism software.
Western blotting was performed per standard methods. In brief, BT-12 and BT-16 cells were treated with 1 μM SNDX-275 for 0, 3, 6, 24, and 48 h. Drug was then removed, and 2 further time points at 6 and 24 h after removal were obtained. At each time point, cells were collected, washed with PBS, resuspended in chilled RIPA buffer and Protease Inhibitor Cocktail (Roche), incubated on ice for 15 min, and centrifuged for 30 min. Protein quantification was performed using the Bradford assay. Incubation of protein lysates with Laemmli sample buffer (Bio-Rad) was performed at 95°C for 5 min. Samples were then subjected to SDS-PAGE (15% Tris-HCl precast gel; Bio-Rad) and transferred onto a polyvinylidene fluoride (PVDF) membrane. Immunoblotting was performed using the Snap I.D. system (Millipore) with primary [p21Waf1/Cip1 (BD Biosciences), actin (Chemicon)] and HRP-conjugated secondary (Jackson Laboratories) antibodies. Chemiluminescence was elicited with SuperSignal West pico and femto substrates (Fermo Scientific) and then exposed to film.
Both ATRT cell lines were seeded in triplicate into 6-well plates for 24 h. Cells were treated with IC50 of SNDX-275 (per MTS assay analysis) versus DMSO control for 24, 48, or 72 h. At the specified time points, media (inclusive of de-adherent cells) and adherent cells were collected, centrifuged, and counted by ViaCount assay on a Guava EasyCyte flow cytometer (Millipore). For each time point, 30 000 cells per cell line were incubated with Nexin reagent (Annexin V-PE and 7-AAD, Millipore) for 20 min. Cell fluorescence was measured and analyzed on the Guava EasyCyte flow cytometer. Numerical differences in apoptotic fractions were analyzed using GraphPad Prism software.
BT-12 and BT-16 cells were treated with IC25 (as measured by MTS assay) of SNDX-275 or DMSO (control) for 24 h. Total RNA was collected using the RNeasy kit (Qiagen). RNA was hybridized to HG-U133 Plus 2.0 GeneChips (Affymetrix). Results of the gene array were normalized with Gene-Chip robust multiarray average (gcRMA) software, resulting in log2 expression values. Genes with either an increase or decrease by 1.75-fold change (0.8 log2) between treated and control in both cell lines were selected for further evaluation by Ingenuity Pathway Analysis (IPA; Ingenuity Systems). Gene lists with mean fold changes were subjected to IPA, and differentially altered networks and biologic function lists were generated with corresponding scores or P value ranges, respectively. The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus (GEO)30 and are accessible through GEO Series accession number GSE31122 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE31122).
cDNA was generated according to the manufacturer's instructions using reverse transcription with mRNA isolated in the gene array experiment and the Mastercycler personal thermal cycler (Eppendorf). Real-time quantitative polymerase chain reaction (PCR) was performed using the StepOne Plus detection system (Applied Biosystems). Taqman reagents (Applied Biosystems) were used according to the manufacturer's recommendations. Gene expression was determined by the ΔCt method. All assays were performed in triplicate.
BT-12 and BT-16 cells were seeded into 6-well plates and subsequently treated with SNDX-275 at 1 and 2 μM concentrations and DMSO (control) for 24 h. Cells were then collected, and live cells were quantified using trypan blue exclusion. For each concentration, 500 cells were seeded per well in triplicate on 6-well plates. Four hours after drug removal, cells were subjected to X-irradiation at doses of 0, 0.5, 1, 2, 2.5, 3, 4, or 5 Gy. Colony formation proceeded until sufficient colony growth was demonstrated in control wells. Media was then removed; cells were washed with PBS and then were stained with crystal violet. Colonies were defined as more than 50 cells and were then counted. Statistical analysis was performed using GraphPad Prism software.
Data obtained from these experiments were processed using Excel (Microsoft), Prism (GraphPad), and gcRMA software, for gene array normalization. IC50 and IC25 were determined using nonlinear regression. Sensitizer enhancement ratios were calculated by dividing the radiation dose of DMSO-treated controls at a specified survival fraction by the radiation dose of the SNDX-275-treated cells at the same survival fraction. Radiation doses at the specified survival fractions were calculated using nonlinear regression analysis of survival curves. All experiments, with the exception of the microarray analysis, were performed, at minimum, independently in triplicate to allow for appropriate statistical analysis. Error bars, unless otherwise indicated, represent the standard error of the mean.
To measure the effect of HDAC inhibition on cell proliferation, BT-12, BT-16, and UPN737 ATRT cells were treated with TsA, SAHA, and SNDX-275. Addition of TsA, SAHA, and SNDX-275 to BT-12 and BT-16 cells and SNDX-275 to the primary patient sample UPN737 resulted in a decrease of metabolically active cells 72 h after drug treatment, as measured by MTS assay (Fig. 1A–C). Incubation with TsA resulted in the lowest IC50 (BT-12: 253 nM; BT-16: 207 nM). IC50 for SAHA (BT-12: 3.8 μM; BT-16: 0.82 μM) and SNDX-275 (BT-12: 2.21 μM; BT-16: 1.61 μM; UPN737: 2.49 μM) treatments were higher but remained in the low micromolar range. Of note, the patient sample UPN737 exhibited an IC50 for SNDX-275 that was comparable to that of the BT-12 and BT-16 cell lines (Fig. 1D).
As a measure of long-term cellular proliferation, BT-12 and BT-16 cell lines were treated with SNDX-275 for 48 h and then evaluated by clonogenic assay. SNDX-275 treatment was associated with decreased colony formation in a dose-dependent manner for both cell lines (Fig. 2, Supplementary Fig. S1). Statistically significant differences (P< .05) were achieved at the 54 nM concentration for BT-16 and 704 nM concentration for BT-12, consistent with previous results from MTS assays demonstrating a higher IC50 for BT-12 cells.
The tumor suppressor protein p21Waf1/Cip1 is frequently implicated in growth arrest and is known to be induced by HDI treatment.16 To determine whether treatment with SNDX-275 increases p21Waf1/Cip1 expression in ATRT, Western blotting of protein samples from BT-12 and BT-16 cell lines was performed. Cells were treated with 1 μM SNDX-275 for 0, 3, 6, 24, or 48 h. Treatment with SNDX-275 resulted in increased p21Waf1/Cip1 expression, with maximum expression demonstrated after 24 h of drug exposure (Fig. 3). After SNDX-275 was removed from the culture medium, p21Waf1/Cip1 expression decreased but continued at 6 and 24 h after removal.
To evaluate the effect of SNDX-275 treatment on apoptosis, BT-12 and BT-16 cells were treated for 24, 48, and 72 h with concentrations of SNDX-275 corresponding to IC50 obtained from the MTS assay analysis. Control samples were treated with DMSO. Apoptosis was measured by Nexin assay (7-AAD and Annexin V-PE). In BT-12 cells, a nonstatistically significant apoptotic effect was seen starting at 24 h after SNDX-275 treatment (Fig. 4A, Supplementary Fig. S2). Early-mid apoptotic and late apoptotic/dead fractions represented 5.5% versus 14.9% (mean values, DMSO vs. SNDX-275) of gated cells and 1.4% versus 2.9%, respectively. At 48 h, statistically significant differences in alive and early-mid apoptotic cell fractions were demonstrated (alive: 91.8% vs. 67%, P< .001; early-mid apoptotic: 5.1% vs. 20.6%, P< .05). After 72 h of drug exposure, all fractions of gated cells were statistically significant (alive: 90.1% vs. 48.5%, P< .001; early-mid apoptotic: 6.1% vs. 25.9%, P< .001; late apoptotic/dead: 3.3% vs. 24.9%, P< .001).
SNDX-275 also induced apoptosis in BT-16 cells; however, the effect did not become evident until 48 h after drug treatment (Fig. 4B, Supplementary Fig. S2). At 48 h, statistically significant differences were seen in all fractions of gated BT-16 cells (DMSO vs. SNDX-275, alive: 91.5% vs. 76.1%, P< .001; early-mid apoptotic: 4.3% vs. 9.8%, P< .05; late apoptotic/dead: 4.1% vs. 13.8%, P< .001).
Because HDIs are known to alter several cellular functions, gene array analysis was performed on BT-12 and BT-16 ATRT cells after SNDX-275 treatment to identify key pathways affected by drug treatment. Total mRNA was collected after 24 h of SNDX-275 exposure, and gene array analysis was performed. After normalization using gcRMA software, the analysis produced a group of 396 differentially upregulated and 71 differentially downregulated genes shared by both cell lines (1.75 minimum fold-change from DMSO-treated controls). Gene lists were then analyzed using IPA. The top 10 differentially altered biologic networks are listed in Table 1, and the top 30 differentially altered biologic functions are listed in Supplementary Table S1. In IPA, networks are defined as collections of interconnected molecules assembled by a network algorithm. Of note, for this study, networks such as cellular growth and proliferation; DNA replication, recombination, and repair; and cell cycle were 3 of the networks identified as being altered by SNDX-275 treatment.
To validate gene expression from the microarray data, real-time quantitative PCR was performed from mRNA isolated during the gene array experiment. Two of the genes significantly upregulated by SNDX-275 treatment in both cell lines were CDKN2B (p15) and CDKN2D (p19), both of which function as inhibitors of cyclin-dependent kinase 4 (CDK4) and inhibit cell cycle progression. In concordance with the microarray data, both of these genes were found to be statistically significantly upregulated by real-time quantitative PCR analysis (Supplementary Fig. S3).
To test the hypothesis that HDAC inhibition in ATRT cells would potentiate the effect of radiation, BT-12 and BT-16 cells were treated with DMSO or SNDX-275 (1 or 2 µM) for 24 h prior to receiving varying doses of X-irradiation. Samples were then analyzed for cellular proliferation by colony formation assay. For both BT-12 and BT-16 cell lines, SNDX-275 pretreatment was found to decrease colony formation after irradiation, compared with the individual doses of radiation alone. Figure 5A and B illustrates colony formation versus radiation as a function of SNDX-275 dose. As shown in Fig. 5C, colony formation was statistically significantly decreased at the lowest radiation dose (0.5 Gy). As a standard measure of radiosensitization, sensitizer enhancement ratios (SERs) were then calculated at survival fractions of 10% (SF0.1) and 50% (SF0.5). SF0.1 and SF0.5 radiation doses were determined using nonlinear regression analysis of survival curves (Supplementary Fig. S4). For BT-12, SERs at SF0.1 were 1.57 and 1.55 for the 1 µM and 2 µM SNDX-275 doses, respectively, and SERs at SF0.5 were 1.68 (1 µM SNDX-275) and 1.69 (2 µM SNDX-275). The BT-16 cell line demonstrated increased sensitivity to the combination of SNDX-275 and X-irradiation. SERs for BT-16 at SF0.1 were 1.61 (1 µM SNDX-275) and 1.75 (2 µM SNDX-275) and, at SF0.5, were 1.72 (1 µM) and 1.89 (2 µM).
Despite some advancements6,7,31 in the treatment of ATRT, long-term outcomes remain suboptimal. In addition, current multimodality therapy for this tumor has proven to be toxic and can be associated with late effects, most notably from intensive radiation dosing delivered to young children. Clearly, ATRT is a malignancy that requires evaluation of newer therapies to improve survival and minimize toxicity.
We have demonstrated in this study that HDI treatment of ATRT cell lines in vitro results in reduced cellular proliferation as measured by MTS and clonogenic assays. In addition, we have shown that a primary ATRT patient sample responds to HDI treatment in a similar manner when compared with established cell lines. Furthermore, we found that the HDI SNDX-275 induces apoptosis in BT-12 and BT-16 cell lines when administered at an IC50 and results in increased p21Waf1/Cip1 tumor suppressor expression. These aspects of our study correlate to previous evaluations that examined the efficacy of HDAC inhibition on rhabdoid cell lines in vitro. In Furchert et al,26 HDAC inhibition by SNDX-275 decreased proliferation of the BT-12 and BT-16 cell lines, but at higher micromolar concentrations than described in our study. In addition, 7.5 µM SNDX-275 was able to elicit an apoptotic response in BT-12 cells after 48 h of incubation. However, in our study, we observed an apoptotic effect of SNDX-275 on BT-12 cells at a much lower concentration. Jaboin et al described the efficacy of SNDX-275 on the renal rhabdoid cell line G401.24 In their study, 1 µM SNDX-275 treatment resulted in cell cycle arrest of G401 cells at 24 and 48 h after drug treatment. In addition, they found that transcription of the p21 tumor suppressor gene (CDKN1A) was induced by SNDX-275. The HDI romidepsin has also shown in vitro activity against non-CNS rhabdoid tumor cell lines G401 and STM91-01.27 In that study, romidepsin was able to induce cell cycle arrest and restore CDKN1C tumor suppressor gene expression. Of interest, they also found that forced expression of SMARCB1 in rhabdoid tumor cell lines was associated with increased histone H3 and H4 acetylation at the CDKN1C promoter. In vivo data has also demonstrated that HDAC inhibition by the HDI depsipeptide results in slower ATRT tumor growth and control in a xenograft model using the BT-40 cell line.25
By evaluating differential gene expression altered by SNDX-275 treatment, we have shown that several key cellular processes are altered by HDI activity. To our knowledge, this is the first study to assess the genomic response to HDAC inhibition in rhabdoid tumor. Of note is the significant alteration of genes involved in cellular growth and proliferation, DNA replication, recombination, and repair and cell cycle. The alterations of these biologic functions correlate well to our data demonstrating the antiproliferative effects of HDAC inhibition on BT-12 and BT-16 cells. Of interest, we specifically found the known tumor suppressor genes CDKN2B and CDKN2D to be upregulated in response to SNDX-275 treatment. Other key biologic processes in malignancy, such as cellular development, cellular movement, and inflammatory response, were also significantly altered by HDAC inhibition. Further investigation will be required to determine these processes’ specific roles in the ATRT response to HDIs. In addition, evaluation of these pathways may prove to be valuable as biological markers of drug response in future experimentation because most current clinical studies focus on histone acetylation as the primary pharmacodynamic parameter.32–35
It has previously been shown that cells with loss of SMARCB1 are more sensitive to DNA damage,36 and we have demonstrated that ATRT is certainly a radiosensitive tumor. Because the effect of radiation can be increased in the context of preexisting DNA damage, it is particularly intriguing that pretreatment with HDIs could enhance the radioresponsiveness of ATRT. This suggests that HDIs lead to DNA damage, a hypothesis that is supported by findings of increased γH2AX foci after HDI treatment of glioma and prostate carcinoma cell lines.20 Our study has demonstrated that SNDX-275 sensitizes ATRT cells to radiation at relatively low drug concentrations and radiation doses. This is presumptively attributable to enhanced, HDI-induced DNA damage, but further studies will be required to elucidate the mechanisms behind this phenomenon. We did, however, observe that 1 of the top 10 biologic networks differentially altered by HDI treatment of BT-12 and BT-16 cells was DNA replication, recombination, and repair. In the clinical setting, the ability of HDAC inhibition to potentiate the effect of radiation in ATRT may allow for the administration of lower radiation doses without sacrificing efficacy. The lower doses could, in turn, aid in decreasing long-term effects seen in young children who receive radiation to the CNS.
HDIs are currently being evaluated in multiple adult clinical trials, including 20 currently active or completed for SNDX-275 administered alone or in combination with other agents.37 However, very few pediatric clinical studies with HDIs have been published to date. In a phase I study, depsipeptide was administered to children with recurrent or refractory solid tumors.38 Although there were no objective responses, the HDI was being used as monotherapy, and the primary goal of the trial was to evaluate dosing. Of note, no patients with ATRT were included in that study. Certainly, further investigation into combination therapy will be needed to evaluate HDIs in the pediatric population because of the in vitro evidence that HDIs frequently function optimally when administered with other treatments. Although HDIs likely have limited use as monotherapy in highly malignant tumors, we believe that our in vitro studies warrant further investigation of the use of HDIs in the context of ATRT, particularly in combination with current therapeutic modalities.
Conflict of interest statement. None declared.
This work was supported by the National Institutes of Health Ruth L. Kirschstein National Research Service Award (T32CA082086 to J.K.), National Institute of Neurological Disorders and Stroke (K08NS59790 to R.V.), and Morgan Adams Foundation to N.F.
We thank Deb DeRyckere, Lori Gardner, Peter Harris, and Meg Macy for their technical advice and the University of Colorado Denver Microarray Core for their assistance with the microarray data.
Presented in part: American Society of Pediatric Hematology/Oncology annual meeting, Montreal, Quebec, Canada, April 7–10, 2010 (Poster 294), and Children's Oncology Group fall meeting, Dallas, Texas, September 21–24 2010.