Histone deacetylase 6 promotes growth of glioblastoma through inhibition of SMAD2 signaling

Abstract Histone deacetylases (HDACs) play a role in the tumorigenesis of glioblastoma multiforme (GBM), whereas the underlying mechanism has not been eluci- dated. Here, we reported significantly higher HDAC6 levels in GBM from the patients. GBM cell growth was significantly inhibited by ACY-1215, a specific HDAC6 inhibitor. Further analyses show that HDAC6 may pro- mote growth of GBM cells through inhibition of SMAD2 phosphorylation to downregulate p21. Thus, our data demonstrate a previously unrecognized regulation path- way in that HDAC6 increases GBM growth through attenuating transforming growth factor β (TGFβ) recep- tor signaling.

Keywords : HDACs . GBM . SMAD2 . p21

Introduction

Histone acetylation and deacetylation are epigenetic events that govern many physiological and pathological cellular processes [1–3]. Histone acetylation is mediated by histone acetyltransferases (HATs), which generally ac- tivates gene transcription. Conversely, histone deacetylation is catalyzed by histone deacetylases (HDACs), which more likely induces gene repression. Histone acetylation is a reversible, dynamic, and highly regulated process for gene regulation [1–3].

Recently, increasing evidence has suggested regulation of many non-histone proteins through reversible acetylation by HATs and HDACs [1–3]. Alterations in this dynamic equilib- rium, such as those caused by the aberrant expression or functional activation of HATs and HDACs, can disturb cell homeostasis and result in pathological states, including onco- genesis and tumor progression [1–3]. In line with these no- tions, histone deacetylase inhibitors (HDACi), a structurally diverse group of small molecules, are capable of antagonizing the effects of HDACs in cancer therapy [1–3].

Glioblastoma multiforme (GBM) is one of the most malig- nant tumors with a dismal prognosis despite the best standard therapy including surgery and radio- and chemotherapy [4–6]. The GBM cells are highly anaplastic and heterogeneous, with an ability to migrate along fiber tracts and blood vessels to disseminate in the brain [4–6]. The tumorigenesis of GBM has been extensively studied in the past decade, with new evi- dence highlighting HDACs as a promising therapeutic target for GBM [7–11]. Since aberrantly activated HDACs appear to play a critical role in the pathogenesis of GBM, HDACi may have anti-neoplastic effects in GBM cells, inducing growth inhibition, cell differentiation, and apoptosis [7–13]. Although great efforts have been made to understand the role of HDACs in regulating tumorigenesis of GBM, the underlying molecu- lar basis has not been elucidated.

Transforming growth factor β (TGFβ) receptor signal- ing pathway plays complicated roles in various biological events [14–19] and specifically affects the tumorigenesis of GBM [20–24]. When a ligand binds to a type II TGFβ receptor, it catalyzes the phosphorylation of a type I TGFβ receptor, which triggers phosphorylation of two intracellular proteins, SMAD2 and SMAD3, to form heteromeric complexes with SMAD4. The activated SMAD complexes then translocate to the nucleus, where they regulate the transcription of target genes [14–19]. SMAD7 has an inhibitory role on SMAD2/3 signaling. SMAD7 has been shown to be a direct substrate for HDACs [25–31], suggesting that HDACs may affect TGFβ receptor signaling. Nevertheless, the relationship between HDACs and TGFβ receptor signaling pathway, especially in GBM, has not been previously documented.

Here, we reported significantly increased HDAC6 levels in GBM from the patients. GBM cell growth was significantly inhibited by ACY-1215, a specific HDAC6 inhibitor. Further analyses show that HDAC6 may pro- mote growth of GBM cells through inhibition of SMAD2 phosphorylation to downregulate p21. Thus, our data demonstrate a previously unrecognized regulation path- way in that HDAC6 increases GBM growth through at- tenuating TGFβ signaling.

Materials and methods

Cell line culture and transfection and reagents

A-172 and U87MG are human glioblastoma cell lines purchased from ATCC. These cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 20 % fetal bovine serum (Invitrogen, Carlsbad, CA, USA). Recombinant TGFβ1 (10 μmol/l) and ACY-1215 (10 nmol/ l) were purchased from Sigma (Carlsbad, CA, USA) and Selleckchem (Carlsbad, CA, USA), respectively, and applied according to the manufacturer’s instruction. ACY-1215 is a specific HDAC6 inhibitor [32].

Patient tissue specimens

A total of 12 resected specimens from GBM patients were collected for this study. All specimens had been histolog- ically and clinically diagnosed at the Department of Neurosurgery of the First Affiliated Hospital of Jinan University from 2007 to 2013. The GBM and adjacent normal brain tissues (NBT) are from the same patient. For the use of these clinical materials for research pur- poses, prior patient’s consents and approval from the Institutional Research Ethics Committee were obtained.

RT-qPCR

RNAwas extracted from resected GBM or adjacent NBT with TRIzol (Invitrogen, Carlsbad, CA, USA) and used for cDNA synthesis. Quantitative PCR (RT-qPCR) was performed in duplicate with QuantiTect SYBR Green PCR Kit (Qiagen, Hilden, Germany). All primers were purchased from Qiagen. NBTwas used as individual sample. Values of genes were first normalized against α-tubulin and then compared to controls (=1).

Cell proliferation assay

For assay of cell proliferation, cells were seeded into a 96-well plate at 5000 cells per well and subjected to a cell proliferation kit (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro- mide (MTT), Roche, Carlsbad, CA, USA), according to the instruction from the manufacturer.

Western blot

The protein was extracted from the cultured cells, which were homogenized in RIPA lysis buffer (1 % NP40, 0.1 % sodium dodecyl sulfate (SDS), 100 μg/ml phenylmethylsulfonyl fluo- ride, 0.5 % sodium deoxycholate, in PBS) on ice. The super- natants were collected after centrifugation at 12,000×g at 4 °C for 20 min. Protein concentration was determined using a BCA protein assay kit (Bio-Rad, Shanghai, China), and whole lysates were mixed with 4× SDS loading buffer (125 mmol/l Tris-HCl, 4 % SDS, 20 % glycerol, 100 mmol/l DTT, and 0.2 % bromophenol blue) at a ratio of 1:3. Samples were heated at 100 °C for 5 min and were separated on SDS- polyacrylamide gels. The separated proteins were then trans- ferred to a PVDF membrane. The membrane blots were first probed with a primary antibody. After incubation with horse- radish peroxidase-conjugated second antibody, autoradio- grams were prepared using an enhanced chemiluminescent system to visualize the protein antigen. The signals were recorded using X-ray film. Primary antibodies were anti- SMAD2, anti-phosphorylated SMAD2 (phos-SMAD2), anti- HDAC6, anti-p21, and α-tubulin (all purchased from Cell Signaling, Carlsbad, CA, USA). α-Tubulin was used as a protein loading control.

Statistical analysis

All statistical analyses were carried out using the SPSS 16.0 statistical software package. All values are depicted as mean± standard deviation from five individuals and are considered significant if p <0.05. All data were statistically analyzed using one-way ANOVA with a Bonferroni correction.

Fig. 1 Significantly higher HDAC6 levels were detected in GBM from the patients. HDAC levels were examined in GBM and compared with those in the adjacent normal brain tissues (NBT), showing a significant increase in HDAC6 in GBM.

Results

Significantly higher HDAC6 levels were detected in GBM from the patients

We examined the HDAC levels in GBM and compared them with those from the adjacent NBT. We specifically detected a significant increase (p<0.05, one-way ANOVA) in HDAC6 in GBM (6.5±0.8-fold) (Fig. 1). Thus, here, we focused on the study of HDAC6 in GBM.

HDAC6 inhibition significantly decreased GBM cell growth

We used two human glioblastoma cell lines, A-172 and U87MG, for our study. Both lines expressed significant levels of HDAC6 (p<0.05, one-way ANOVA). We thus used ACY-1215 to inhibit HDAC6 and then examined its effect on the growth of GBM cells in a MTT assay. We found that HDAC6 inhibition significantly decreased GBM cell growth (p<0.05, one-way ANOVA) (Fig. 2a, b), suggesting that HDAC6 pro- motes GBM cell growth. The inhibition of HDAC6 by ACY- 1216 was assured by Western blot in both lines (Fig. 2c).

HDAC6 promoted growth of GBM cells through inhibition of SMAD2 phosphorylation

In order to figure out the underlying mechanism, we examined the SMAD2 and its phosphorylation. We found that ACY- 1215 significantly increased phosphorylation of SMAD2 and subsequently its nuclear target p21, a well-known cell cycle inhibitor (Fig. 2c). To examine whether the modulation of SMAD2 phosphorylation is responsible for the effect of HDAC6 on GBM cell growth, we gave both lines recombi- nant TGFβ1, a strong inducer of SMAD2 phosphorylation. We found that TGFβ1 significantly induced SMAD2 phos- phorylation and increased p21, without affecting HDAC6 levels (Fig. 2c). Moreover, the intervention with TGFβ1 resulted in decreases in cell growth (p < 0.05, one-way ANOVA) (Fig. 2a, b). These data thus suggest that HDAC6 may increase GBM growth through attenuating TGFβ signaling.
Discussion

Histone acetylation and deacetylation are dynamic and highly regulated processes to regulate gene expression [1–3]. Alterations in this dynamic equilibrium can disturb cell homeostasis and result in pathological states, including onco- genesis [1–3]. GBM is one of the most malignant tumors with an extremely poor prognosis [4–6]. Of note, accumulating evidence has highlighted HDACs as a promising therapeutic target for GBM [7–13].

Fig. 2 HDAC6 promoted growth of GBM cells through inhibition of SMAD2 phosphorylation. a A-172 cell growth was examined in a MTT assay. b U87MG cell growth was examined in a MTT assay. ACY-1215 is a specific HDAC6 inhibitor. TGFβ1 is a

Fig. 3 Schematic of the model. HDAC6 inhibits SMAD2 phosphoryla- tion, which is required for p21 activation to inhibit GBM growth. ACY- 1215 is a specific HDAC6 inhibitor.

It has been more and more recognized that many non- histone proteins are regulated through reversible acetylation by HATs and HDACs [1–3].
Although great efforts have been made to understand the regulation of tumorigenesis of GBM by HDACs, the under- lying molecular basis has not been elucidated. Similarly, al- though increasing evidence suggest that HDACs may affect TGFβ receptor signaling, the relationship between PDACs and TGFβ receptor signaling pathway, especially in GBM, has not been previously documented. Here, we showed sig- nificantly higher HDAC6 levels in GBM from the patients and further showed evidence supporting TGFβ receptor signaling pathway as a downstream target of HDAC6 in GBM, consis- tent with previous reports highlighting SMAD7 as a direct substrate for HDACs [25–31].

From all HDACs, we specifically detected the most signif- icant upregulation of HDAC6 in GBM. This specificity may be cell type- and tumor type-dependent. In the future, other assays such as migration and invasion may be included to strengthen our conclusion. Although we only examined tran- script levels of these HDACs, we did not expect presence of post-translational control, according to previous studies [7–11]. We have actually examined more GBM cell lines and reached essentially the same conclusion, which thus ex- cludes a cell line-dependent possibility (data not shown).

In order to figure out the underlying mechanism, we ex- amined the SMAD2 and its phosphorylation, which has been well characterized for its inhibitory role in tumor growth [33, 34]. SMAD2 phosphorylation and nuclear translocation for targeting p21 are well-known means of cell cycle control in various cell types, including tumor cells (Fig. 3) [14–19].

Here, we show, for the first time, that SMAD2 signaling is directly regulated by HDAC6. We did check SMAD3 but found that it was expressed in a much lower level, compared to SMAD2 in the current context. Thus, we think that SMAD2 signaling plays a predominant role here and we just focus on SMAD2. According to previous studies showing SMAD7 as a direct target for HDACs [25–31], the effect of HDAC6 on SMAD2 signaling may result from a direct enhancing role of HDAC6 on SMAD7, which in turn inhibits SMAD2 phos- phorylation. This hypothesis may be examined in future experiments.
Taken together, our data demonstrate a previously unrec- ognized regulation pathway in that HDAC6 increases GBM growth through attenuating TGFβ signaling.