A Novel Histone Deacetylase Inhibitor Exhibits Antitumor Activity via Apoptosis Induction, F-Actin Disruption and Gene Acetylation in Lung Cancer

Background Lung cancer is the leading cause of cancer mortality worldwide, yet the therapeutic strategy for advanced non-small cell lung cancer (NSCLC) is limitedly effective. In addition, validated histone deacetylase (HDAC) inhibitors for the treatment of solid tumors remain to be developed. Here, we propose a novel HDAC inhibitor, OSU-HDAC-44, as a chemotherapeutic drug for NSCLC. Methodology/Principal Findings The cytotoxicity effect of OSU-HDAC-44 was examined in three human NSCLC cell lines including A549 (p53 wild-type), H1299 (p53 null), and CL1-1 (p53 mutant). The antiproliferatative mechanisms of OSU-HDAC-44 were investigated by flow cytometric cell cycle analysis, apoptosis assays and genome-wide chromatin-immunoprecipitation-on-chip (ChIP-on-chip) analysis. Mice with established A549 tumor xenograft were treated with OSU-HDAC-44 or vehicle control and were used to evaluate effects on tumor growth, cytokinesis inhibition and apoptosis. OSU-HDAC-44 was a pan-HDAC inhibitor and exhibits 3–4 times more effectiveness than suberoylanilide hydroxamic acid (SAHA) in suppressing cell viability in various NSCLC cell lines. Upon OSU-HDAC-44 treatment, cytokinesis was inhibited and subsequently led to mitochondria-mediated apoptosis. The cytokinesis inhibition resulted from OSU-HDAC-44-mediated degradation of mitosis and cytokinesis regulators Auroroa B and survivin. The deregulation of F-actin dynamics induced by OSU-HDAC-44 was associated with reduction in RhoA activity resulting from srGAP1 induction. ChIP-on-chip analysis revealed that OSU-HDAC-44 induced chromatin loosening and facilitated transcription of genes involved in crucial signaling pathways such as apoptosis, axon guidance and protein ubiquitination. Finally, OSU-HDAC-44 efficiently inhibited A549 xenograft tumor growth and induced acetylation of histone and non-histone proteins and apoptosis in vivo. Conclusions/Significance OSU-HDAC-44 significantly suppresses tumor growth via induction of cytokinesis defect and intrinsic apoptosis in preclinical models of NSCLC. Our data provide compelling evidence that OSU-HDAC-44 is a potent HDAC targeted inhibitor and can be tested for NSCLC chemotherapy.


Introduction
Lung cancer is the leading cause of cancer mortality worldwide. The 5-year overall survival of non-small cell lung cancer (NSCLC) is less than 15% in many countries [1,2]. The standard therapeutic strategy for advanced NSCLC is platinum-based double-agent chemotherapy which, however, has reached a plateau of potency in improving survival of patients [3,4]. Only a few ''target agents'' have showed benefits when used in combination with platinumbased double-agent for NSCLC chemotherapy, such as bevacizumab, erlotinib and gefitinib, in a subset of patients [5,6]. Therefore, the development of novel molecular targeted drugs with more general effectiveness for lung cancer patients is an imperative task.
The epigenetic changes as well as genetic alterations are associated with tumorigenesis [7]. A recent report identifies that the epigenetic changes involving modifications of histones H2A and H3 in NSCLC patients influence the overall survival and disease-free survival, providing the prognostic value of histone modifications [8]. It also reveals the rationale for the use of drugs against histone modification as a therapeutic strategy for NSCLC.
Histone deacetylases (HDACs) are the enzymes that catalyze the deacetylation of histones and epigenetically regulate chromatin architecture and gene expression. It has been demonstrated that inhibition of HDACs reverses aberrant epigenetic status and exhibits potent antitumor activities by inducing cell cycle arrest, differentiation and/or apoptosis in diverse cancer cells [9,10]. HDAC inhibitors are classified into six groups according to their chemical structures and at least 12 of them have progressed to clinical trials [9,11]. To date, the U.S. Food and Drug Administration approves two HDAC inhibitors, vorinostat (SAHA, suberoylanilide hydroxamic acid, ZolinzaH) and romidepsin (FK228, depsipeptide, IstodaxH), for the treatment of cutaneous manifestations of cutaneous T-cell lymphoma (CTCL) [12]. However, some adverse events occur in patients treated with vorinostat or other HDAC inhibitors, which may result from the high concentrations of dose used during the treatment for solid tumors in clinical trials [11,13].
In the present study, we propose a novel class of potent phenylbutyrate-based HDAC inhibitor, OSU-HDAC-44 [4-(2,2dimethyl-4-phenyl-butyrylamino)-N-hydroxy-benzamide], a derivative of known HDAC inhibitor, N-Hydroxy-4-(4-phenylbutyryl-amino)benzamide (HTPB) [14]. The antitumor activities and mechanisms of OSU-HDAC-44 were studied in NSCLC cell and mice xenograft models. We found that OSU-HDAC-44 was a pan-HDAC inhibitor and exhibited 3-4 times more effectiveness in suppressing cell proliferation in vitro and tumor growth in vivo compared to SAHA or trichostatin A (TSA). In addition, OSU-HDAC-44 induced mitosis and cytokinesis defect followed by mitochondria-mediated apoptosis in both cell and animal models. Chromatin-immunoprecipitation-on-chip analysis revealed the genome-wide target genes which were induced by OSU-HDAC-44-mediated hyperacetylation of chromatin. Our data suggest that OSU-HDAC-44 was an HDAC inhibitor and could be applied as targeted anticancer drug for NSCLC chemotherapy.

OSU-HDAC-44 inhibits cell proliferation and shows synergistic effects with cisplatin regardless of p53 status
The structure of OSU-HDAC-44 and SAHA are shown in Fig. 1A. Docking analysis demonstrated that OSU-HDAC-44 interacted with the catalytic domain of HDAC 8, suggesting the direct function of OSU-HDAC-44 in targeting HDACs (Fig. 1B).

OSU-HDAC-44 induces cytokinesis inhibition and apoptosis
To investigate the underlying mechanism of cell growth repression by OSU-HDAC-44, the effects of OSU-HDAC-44 on cell cycle progression were assessed by flow cytometry. Treatment with 2.5 mM OSU-HDAC-44 for 24 hours caused A549 and H1299 cells to accumulate in G2/M phase (4N cells), and subsequently led to apoptosis (sub-G1 cells) at 48 hours treatment, while exposure to higher concentration (5 mM) of SAHA for 48 hours had similar effect ( Fig. 2A), indicating that OSU-HDAC-44 exerted a more potent cell cycle deregulation effect than did SAHA. To examine the cellular consequences of OSU-HDAC-44mediated accumulation of 4N cells, time-lapse microscopic analyses were performed. As shown in Fig. S1A, OSU-HDAC-44 caused the appearance of the defective cleavage furrow structure and the two daughter cells were fused back together, while untreated cells passed normally through cell division. Concordantly, about 20% cells treated with OSU-HDAC-44 were accumulated as bi-nucleated cells, compared with less than 5% of control cells ( Fig. 2B and Fig. S1B). OSU-HDAC-44 also caused micronuclei formation and disrupted the normal structure of F-actin of A549 and H1299 cells (Fig. 2B). Hence, these results suggested that OSU-HDAC-44 may cause aberrant cytokinesis and subsequently led to apoptosis in lung cancer cells.
To identify the molecular mechanism involved in OSU-HDAC-44 induced cytokinesis inhibition, the cell cycle-regulatory proteins were examined. The oscillation of mitotic inhibitor Weel and mitotic markers phosphorylated histone H3 and cyclin B expression indicated that OSU-HDAC-44-treated cells were in M phase after 12 hours treatment and subsequently exited M phase (Fig. S1C), accompanied with cytokinesis defect. Moreover, OSU-HDAC-44 caused decreases in protein levels of Aurora B and survivin ( Fig. 2C; upper), which are essential for the progression of mitosis and cytokinesis [15,16]. Notably, OSU-HDAC-44 induced ubiquitination of Aurora B and survivin, and cotreatment with proteosome inhibitor MG132 prevented the OSU-HDAC-44-induced degradation of Aurora B and survivin ( Fig. 2C; middle and lower). Next, we used nocodazole to synchronize cells at pre-metaphase and to further confirm that OSU-HDAC-44 indeed triggered abnormal degradation of Aurora B and survivin at mitotic phase. As shown in Fig. S1D and E, treatment with nocodazole for 24 hours caused accumulation in Aurora B and survivin proteins, whereas combination of OSU-HDAC-44 and nocodazole resulted in decreases Aurora B and survivin protein levels upon 24 hours post-treatment. These results suggested that the OSU-HDAC-44-mediated failure of cytokinesis may partly result from the downregulation of Aurora B and survivin proteins via 26S proteasome pathway.

OSU-HDAC-44 activates the intrinsic apoptotic pathway
To further elucidate the OSU-HDAC-44-induced apoptosis, we performed phosphatidylserine (PS) staining analyses to detect the early process of apoptosis. As shown in Fig. S2, OSU-HDAC-44 treatment for 24 hours increased the intensity of PS staining in contrast to low staining intensity upon DMSO treatment in A549 and H1299 cells. In addition, OSU-HDAC-44 treatment significantly stimulated caspase-3 and caspase-9 (an indicator of the intrinsic mitochondrial pathway) activities after 24 hours treatment whereas the activity of caspase-8 (an indicator of the extrinsic membrane receptor pathway) remained unaffected in A549 and H1299 cells (Fig. 2D, left). Moreover, treatment with 2.5 mM OSU-HDAC-44 for 12 hours caused a decrease in anti-apoptotic protein Bcl-x L , while it increased the pro-apoptotic protein, Bad, within 6-12 hours treatment in A549 and H1299 cells (Fig. 2D, right). The release of cytochrome c into the cytosol accompanied by the cleavage of procaspase-9 was also seen after OSU-HDAC-44 treatment for 24-48 hours. These results further confirmed that OSU-HDAC-44 could induce the intrinsic apoptotic pathway in lung cancer cells.

OSU-HDAC-44 induces protein acetylation with its ability to target numerous HDACs
The biomarkers of HDAC inhibition are acetylation of histone and non-histone proteins, and induction p21 Cip1 expression in a p53-independent manner [17,18]. Exposure to OSU-HDAC-44 induced acetylation of histone H3, histone H4 and p53 in a dosedependent manner (Fig. 3A) and time-dependent manner (Fig. 3B), while it did not affect the HDAC1 and HDAC6 protein levels ( Fig. 3B and Fig. S3). Notably, such effects were greater compared to that of SAHA. Despite the p53 status, OSU-HDAC-44 induced the expression of p21 Cip1 mRNA and protein in A549 and H1299 cells (Fig. S4). To examine the target specificity of OSU-HDAC-44 on class I, II, and IV HDACs, in vitro HDAC inhibition assay was performed. As shown in Fig. 3C, the deacetylase activities of different HDAC isotypes including class I (HDAC1 and HDAC8), class II (HDAC4 and HDAC6), and class IV (HDAC11) were significantly inhibited by OSU-HDAC-44. Such effects were greater compared to that of SAHA, a known pan-HDAC inhibitor. These results suggested that OSU-HDAC-44 induced protein acetylation by exerting broad inhibitory activity upon numerous HDACs.

OSU-HDAC-44 increases gene expression by loosening the chromatin structure
To determine the direct effects of OSU-HDAC-44 on chromatin structure and gene expression, the chromatin-immunoprecipitation (ChIP)-on-chip analysis was performed using the antibody against the loose chromatin mark, acetylated lysines 9 and 14 of histone H3 (H3K9K14Ac), after 2.5 mM OSU-HDAC-44 treatment for 2 hours in A549 and H1299 cells. Induction of histone acetylation in 33 common gene loci of A549 and H1299 were identified after OSU-HDAC-44 treatment (Table S1). Several of these 33 genes had been demonstrated to play important roles in certain signaling pathways, such as apoptosis, oxidative stress response, axon guidance and protein ubiquitination pathways (Table 1). To confirm microarray data, we validated the chromatin structure of some of the gene loci including srGAP1, NR4A1 and FOXO4 by ChIP-PCR using the antibody against H3K9K14Ac. As shown in Fig. 4A, treatment with 2.5 mM OSU-HDAC-44 for 2 hours increased the amount of srGAP1, NR4A1 and FOXO4 promoter DNA with loose chromatin structure compared to untreated cells. Concordantly, the mRNA levels of srGAP1, NR4A1 and FOXO4 were increased after OSU-HDAC-44 treatment for 24 hours (Fig. 4B).

OSU-HDAC-44 down-regulates F-actin dynamics via srGAP1 induction
OSU-HDAC-44 treatment induced F-actin aggregation (Fig. 2B). Previous study has indicated that srGAP1 binds to the active forms of RhoA and Cdc42 and inhibits their activities in regulating actin polymerization in neuron cells [19]. However, the biological function of srGAP1 binding to RhoA remains unclear in other cells. Using immunoprecipitation (IP)-Western, we showed that OSU-HDAC-44 increased the interaction between srGAP1 and RhoA in A549 lung cancer cells (Fig. 4C). Interestingly, knockdown of srGAP1 not only abolished the OSU-HDAC-44-mediated decrease in RhoA-GTP level (Fig. 4D, upper), but also restored the dynamics of F-actin after OSU-HDAC-44 treatment (Fig. 4D, lower). These results indicated that OSU-HDAC-44 down-regulated RhoA activity partly via srGAP1 induction, leading to destruction of normal F-actin fibers.

OSU-HDAC-44 inhibits lung tumor xenograft growth in vivo
To further evaluate the antitumor activity of OSU-HDAC-44, Bulb/c null mice bearing A549 lung tumor xenograft were injected intraperitoneally with 7.5-30 mg/kg of OSU-HDAC-44, 3 days/week for three weeks. TSA of 0.5 mg/kg, which has been demonstrated to exhibit anti-tumor growth effects in xenograft of breast and bladder cancer cells [20,21], was used as a positive control drug. As shown in Fig. 5A, treatment with 7.5, 15 and 30 mg/kg OSU-HDAC-44 significantly inhibited tumor growth by 62%, 78% and 90%, respectively, on day 33 post-treatment compared with vehicle control. Treatment with OSU-HDAC-44 did not adversely affect body weight and caused no detectable plotted in the histogram. (B) The bi-nucleated cells and dysregulation of F-actin induced by OSU-HSAC-44. Cells were treated with 2.5 mM OSU-HDAC-44 for 48 h, and then fixed and stained with DAPI (DNA) and phalloidin (F-actin). Asterisk pointed to the bi-nucleus. Scale bars: 30 mm. (C) OSU-HDAC-44 induced degradation of Aurora B and survivin via 26S proteasome pathway. Upper, time-dependent decreases in Aurora B and survivin protein levels after 2.5 mM OSU-HDAC-44 treatment. Middle, A549 cells were treated with 2.5 mM OSU-HDAC-44 in the presence or absence of MG132 for 24 h. Lower, A549 cells were treated with 2.5 mM OSU-HDAC-44 for 24 h and cell lysate was subjected to IP assay using anti-Aurora B or antisurvivin specific antibodies and blotted with anti-ubiquitination antibody (Anti-Ub). (D) Caspase activity assay (left) and Western blot analyses (Right) confirmed that OSU-HDAC-44 induced intrinsic apoptosis pathway. Cells were treated with 2.5 mM OSU-HDAC-44 for indicated times and the subjected to caspase activity assay and Western blot analyses. Data represent mean 6 SEM from three independent experiments. * P,0.05; ** P,0.01. doi:10.1371/journal.pone.0012417.g002 toxicity as examined by hematoxylin and eosin staining of major organs (Fig. 6A, B). Hematological biochemistry examinations were in the normal ranges for OSU-HDAC-44 treated animals (Fig. 6C).

OSU-HDAC-44 induces protein acetylation, apoptosis and cytokinesis inhibition in vivo
To confirm that OSU-HDAC-44 suppressed xenograft tumor growth via targeting the HDACs and inducing apoptosis in vivo, mice bearing established A549 tumors were treated with a single dose of OSU-HDAC-44. After treatment for indicated time, tumors were dissected and subjected to Western blot, immunohistochemistry or fluorescence immunohistochemistry analysis ( Fig. 5B-D). Acetylation of histone H3, histone H4 and p53 were profoundly increased after 2 hours treatment. The protein levels of Bcl-x L and survivin started to decrease after 2 hours treatment, while the level of Bad protein was increased after 4 hours treatment (Fig. 5B). Activated caspase-3 was also detected in both the cytosol and nucleus after 8 hours treatment and was further enhanced after 24 hours treatment (Fig. 5C). Furthermore, OSU-HDAC-44 decreased Aurora B levels and interrupted its association with metaphase chromosome in comparison with DMSO control cells (Fig. 5D). These results demonstrated that OSU-HDAC-44 could induce apoptosis and down-regulate mitotic and cytokinesis regulators, Aurora B and survivin, in vivo. In addition, increase of HDAC inhibition biomarkers such as acetylation of histone H3, histone H4 and p53 was evident in tumors of treated mice.

Discussion
Since HDACs are promising targets for cancer therapy, a number of HDAC inhibitors are in clinical trials as single therapy and/or in combination with other anticancer drugs [9]. However, effective HDAC inhibitors for treatment of solid tumors remain to be developed. In this study, we provide compelling evidence from cell and animal studies that OSU-HDAC-44, a phenylbutyrate-based compound, is a potential HDAC inhibitor for NSCLC treatment. OSU-HDAC-44 targeted numerous members within three classes of HDACs in vitro and efficiently stimulated protein acetylation in cell and animal models ( Fig. 3 and 5). OSU-HDAC-44 repressed cell viability and induced apoptosis in various NSCLC cell lines with 3-4 times greater potency than SAHA ( Fig. 1C and 2A). In addition, submicromolar concentration of OSU-HDAC-44 exhibited prominently synergistic effects in combination with cisplatin on suppressing proliferation of NSCLC cell lines (Fig. 1D). The xenograft experiments further confirmed that OSU-HDAC-44 induced cell apoptosis and thereby inhibited tumor growth in vivo (Fig. 5) without adversely affected body weight, major organs and hematological parameters (Fig. 6). Collectively, these results suggested that OSU-HDAC-44 is a promising candidate HDAC inhibitor for NSCLC treatment.
It has been shown that several kinases and regulatory proteins, such as Aurora B, suvivin as well as small GTPase RhoA are required to complete cytokinesis [22]. Inhibition of Aurora B or depletion of survivin can prevent the late steps of cytokinesis, leading to formation of multi-nucleated cells [15,16]. In the current study, we provided evidence that OSU-HDAC-44 induced proteolysis of Aurora B and survivin both in vitro and in vivo ( Fig. 2C and Fig. 5B, D), which was associated with OSU-HDAC-44mediated cytokinesis inhibition, resulting in the accumulation of bi-nucleated cells ( Fig. 2B and Fig. S1A-B). In addition, combination of a pre-metaphase inducer nocodazole and OSU-HDAC-44 resulted in decrease of Aurora B and survivin protein levels upon 24 h post-treatment (Fig. S1E). These data suggested that OSU-HDAC-44-mediated cytokinesis defect was due to abnormal degradation of Aurora B and survivin in mitotic phase. It has been reported that overexpression of Aurora B correlates with survivin expression in the nucleus, lymph node invasion, and poor prognosis in NSCLC patients [23]. Thus, the clinical efficacy of OSU-HDAC-44 in relation to down-regulated Aurora B and surivin in treatment of NSCLC patients is worthy of further investigation.
In this study, we performed a ChIP-on-chip analysis to investigate the genome-wide target genes induced by OSU-HDAC-44-mediated hyperacetylation of chromatin after 2 hours exposure, and found that histone acetylation were stimulated in 33 common genes in the cell lines examined, including eight tumor suppressor genes (TSGs) or TSG-like genes (Table S1). Several genes play essential roles in apoptosis, oxidative stress response, axon guidance and protein ubiquitination pathways ( Table 1). The srGAP1 gene, which encodes a GTPase activating protein known to regulate axon guidance [19], was confirmed to be in the open chromatin structure and increased in expression level (Fig. 4A, B). Interestingly, we found that OSU-HDAC-44 decreased the activity of a small GTPase RhoA via induction of srGAP1 and contributed to dysregulation of F-actin dynamics (Fig. 4C, D). These results indicated that OSU-HDAC-44 may interrupt mitosis and cytokinesis resulting from alteration of several additional pathways, such as srGAP1/RhoA/F-actin control. Moreover, two apoptosis-related genes, NR4A1/Nur77 and FOXO4, were validated from the ChIP-on-chip data and their mRNA expressions were indeed increased by OSU-HDAC-44 (Fig. 4A, B). NR4A1/Nur77 and FOXO4 have been shown to trigger intrinsic apoptosis through induction of mitochondrial cytochrome c release and down-regulation of Bcl-x L expression, respectively [24][25][26]. Such NR4A1/Nur77-mediated apoptosis has been demonstrated to be induced by an HDAC inhibitor, LBH589, in CTCL cells [27]. Our results from cell and animal models showed that the OSU-HDAC-44-induced cell death was possibly through the intrinsic apoptotic pathway (Fig. 2D and 5B). Therefore, the transcriptional up-regulation of NR4A1/Nur77 and FOXO4 may contribute to OSU-HDAC-44-mediated intrinsic apoptosis.
Similar to our finding of selective chromatin change of a fraction of gene loci in ChIP-on-chip, recent studies using cDNA microarrays indicate that several HDAC inhibitors such as TSA, SAHA, MS-275 and depsipeptide alter only 7-20% gene expressions in various cancer cell lines [28][29][30]. Specific recruitment of corepressor complexes containing HDACs by transcription factors and/or transcription regulators is believed to play an essential role in transcriptional repression [31][32][33], however, the selective action of HDAC inhibitors on specific genes remains unclear. Thus, it is worthy to investigate whether there may be common and critical transcription-regulatory complexes containing HDACs that determine the acetylation levels of chromatin of these genes validated from ChIP-on-chip data.
In conclusion, our findings shows that OSU-HDAC-44 is a novel pan-HDAC inhibitor that exhibits a broad spectrum of antitumor activities in NSCLC cell and xenograft models, which involves not only histone acetylation-dependent activation of gene transcription, but also activation of intrinsic apoptotic pathways and post-translational down-regulation of mitotic regulators, Aurora B and survivin. In addition, RhoA/F-actin motility control was inhibited by srGAP1 and several apoptosis induction proteins were activated by OSU-HDAC-44 (Fig. 7). Collectively, our data provide compelling evidence that OSU-HDAC-44 is an HDAC targeted inhibitor and has the potential to be tested for NSCLC treatment and combination chemotherapy.

Cell lines and culture conditions
Human normal lung cell line IMR90 and human NSCLC cell lines A549 and H1299 were obtained from the American Type Culture Collection (ATCC, Manassas, VA), and the human NSCLC cell line CL1-1 was kindly provided by Dr. P-C Yang (Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan) [34]. All cell lines were cultured in Dulbecco's Modified Eagle's Medium (GIBCO, Grand Island, NY) containing 10% fetal bovine serum (FBS) (BIOCHROM AG, Leonorenstr, Berlin, Germany) and 1% penicillin-streptomycin (GIBCO) and incubated at 37uC in 5% CO 2 atmosphere.

Preparation of OSU-HDAC-44
Isobutyric acid (1.4 mL) was added dropwise to a mixture of diisopropylamine (2.2 mL, 0.015 mol) and 54% sodium hydride in mineral oil (0.74 g, 0.0165 mol) in THF (40 mL), and refluxed for 15 min. When the solution was cooled to 0uC, a standard solution of n-butyllithium in heptane (1.45 mmol/mL; 9.4 mL) was added. After 20 min at 0uC, the mixture was heated to 30-35uC for 30 min and then cooled to 0uC when (2-bromoethyl)-benzene (2.8 mL, 15 mmol) was added to the reaction mixture over 20 min. The ice-bath was retained for 30 min, the mixture was then heated to 30-35uC for 1 h, and then 40 mL of water was added to the reaction mixture at a temperature below 15uC. The aqueous layer was separated, and the organic layer was washed Hexane was added to the resulting colorless oil to yield 1.1 g of white solid 2, 2,-Dimethyl-4-phenylbutyric acid compound. Oxalyl chloride (2 mmol) was added to the cooled solution of 2, 2,-Dimethyl-4-phenylbutyric acid (1 mmol) in dichloromethane (5 mL), and the reaction mixture was then brought to room temperature and stirred for 4 h. After the completion of the reaction, solvent was removed under vacuum. The residue was dissolved in dichloromethane (10 mL) and cooled to 0uC. Paraamino benzoic acid was then added to the reaction mixture followed by addition of triethyl amine. Resultant mixture was brought to room temperature and stirred overnight. Reaction mixture was then concentrated and purified by column chromatography to give 4-(2, 2-dimethyl-4-phenylbutanamido) benzoic acid compound. The cooled 4-(2, 2-dimethyl-4-phenylbutanamido) benzoic acid compound (1 mmol) in DMF (1 mL) was added triethyl amine (1.2 mmol) followed by PyBOP (1.2 mmol). Resultant mixture was stirred at room temperature for 4 h. After complete consumption of starting material as evidenced by TLC, reaction mixture was cooled to 0uC and hydroxylamine hydrochloride (1.2 mmol) was added to the reaction mixture followed by addition of triethyl amine (1.5 mmol). Resultant mixture was stirred at room temperature overnight and then quenched with water. Solid was filtered and purified by column chromatography to give OSU-HDAC-44.

Caspase activity assay
Caspase activity was measured with the caspase luminescent assay kit (Promega, Madison, WI) according to the manufacturer's instructions. Briefly, cells were plated in a 96-well plates and treated with 2.5 mM OSU-HDAC-44 for 12 or 24 h, followed by incubating with various synthetic caspase substrates (Ac-DEVD-pNA, Ac-LETD-pNA, and Ac-LEHD-pNA) to measure the activity of caspases23, 28, and 29, respectively. After incubation for 1 h, luminescence was detected on a SpectraMaxH M5 microplate reader (Molecular Devices, Sunnyvale, CA).

Immunoprecipitation assay
Catch and Release Reversible Immuonprecipitation System kit (Upstate, Temecula, CA) was used for protein-protein interaction analysis. One mg cell protein lysates were incubated with the appropriate antibodies, including anti-srGAP1, anti-RhoA, anti-Aurora B, anti-survivin or normal mouse-IgG, and 10 ml affinity ligand, and immunoprecitation was then performed according to the manufacturer's protocol. After incubation at 4uC overnight, immune complexes were washed with wash buffer for three times. Proteins were eluted and then blotted with appropriate antibodies using the conditions described in the Table S2.

RhoA activation assay
The RhoA activation assay was performed by using active Rho pull-down and detection Kit (Pierce, Rockford, IL). Briefly, a glutathione S-transferase (GST) fusion protein containing the Rho binding domain (RBD) from Rhotekin was used. One mg protein lysates were incubated with 400 mg of purified GST-Rhotekin-RBD immobilized on agarose-glutathione beads for 1 hour at 4uC with constant agitation. The beads were washed three times with 1X Lysis/Wash buffer and bound proteins were eluted and subjected to Western blot analysis using RhoA antibody described in the Table S2.

Molecular docking analysis
In order to show the interaction between OSU-HDAC-44 and HDAC, molecular docking assay was conducted. The reference compound, SAHA, was included. We calculated the structure of OSU-HDAC-44 and SAHA and predicted the docking mode on catalytic domain of HDAC8 using the docking program GOLD 4.0.1 to confirm the accuracy of this prediction program. The three dimensional structure of OSU-HDAC-44, the binding affinity of OSU-HDAC-44 to HDAC8, and the angles of OSU-HDAC-44 and HDAC8 were calculated by this prediction program, with consideration of molecular interaction, such as hydrogen bound and van der Waals force.

HDAC inhibition Assay
Different HDAC isotypes were immunoprecipitated from nuclear extract using specific anti-HDAC-1, 24, 26, 28, and 211 antibodies. The HDAC activity assay was performed using a HDAC fluorescent activity assay kit (BIOMOL Inc, Plymouth Meeting, PA) according to the manufacturer's instructions. Briefly, the specific HDAC isotypes were added to the diluted OSU-HDAC-44 (1 mM) and SAHA (1 mM), and then the substrate was added. Samples were incubated for 10 min at 25uC, followed by adding developer to stop the reaction. After incubation for 10 min, luminescence was recorded on a SpectraMaxH M5 microplate reader (Molecular Devices, Sunnyvale, CA).

Target promoter chromatin immunoprecipitation (ChIP)-PCR Assay
Treated and untreated lung cancer cells were cross-linked with 1% formaldehyde for 15 min at 37uC. Chromatin was immunoprecipitated with anti-acetylated lysine 9 and lysine 14 of histone H3 (H3K9K14Ac) using the ChIP assay kit (Upstate) according to the manufacturer's instructions and the conditions were described in the Table S2. PCR analysis for protein-DNA complex was performed using the following primer pairs: Chromatin structure profiling assay: ChIP-on-chip assay The A549 and H1299 cells (4610 6 ) were treated with DMSO or OSU-HDAC-44 for 2 h, and then immunoprecipitated using antibody to H3K9K14Ac as the conditions described in the Table  S2. DNA was amplified and labeled by ligation-mediated PCR with Cy5 and Cy3 fluorescent dyes, respectively. Both pools of labeled DNA were hybridized to the NimbleGen human 385k RefSeq Promoter array (Roche NimbleGen Inc., Madison, WI). Images of fluorescence intensities were generated by scanning array using GenePix 4000B scanner (Axon Instruments, Union City, CA), and then data were extracted and ChIP signals were normalized using NimbleGen SignalMap software. The ChIP-onchip data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE20304 (http://www.ncbi.nlm. nih.gov/geo/query/acc.cgi?acc = GSE20304). Relative quantitation using the comparative Ct method with the data from ABI PRISM 7000 (version 1.1 software) was performed according to the manufacturer's protocol. Analysis of p21 gene expression and its primer sequence are described in the Supplementary Methods S1.

Immunofluorescence and confocal microscopic analysis
To stain for DNA and F-actin, the fixed cells were stained with DAPI and Phalloidin, respectively, for 1 hour and then the images were recorded by an OLYMPUS FV1000 confocal microscope (Olympus America Inc., Melville, NY). For examination of the degradation of Aurora B and survivin at mitotic phase by OSU-HDAC-44, nocodazole was used to synchronize cell and then cells were subjected to immunofluorescence and confocal microscopic analysis as described in the Supplementary Methods S1.

Xenograft studies
Athymic nu/nu female mice (BALB/c), 4-5 weeks of age, were obtained from the National Laboratory Animal Center (Republic of China, Taiwan) after being approved by Institutional Animal Care and Use Committee (IACUC), National Cheng Kung University (IACUC Approval No. 99131) and maintained in pathogen free conditions. Eight mice per group were used in the xenograft studies. The animals were implanted subcutaneously with 5610 6 A549 cells in 0.1 ml Hanks' balanced salt solution (HBSS) in one flank per mouse. The tumor size was measured according to the formula: (Length6Width 2 )/2. When tumors had attained a mass of ,50 mm 3 , animals were treated intraperitoneally with OSU-HDAC-44 (7.5 mg/kg, 15 mg/kg or 30 mg/kg), TSA (1.5 mg/kg) or DMSO as control on days 1, 3, and 5 for three weeks. Prior to being sacrificed, the animals were anesthetized and blood samples were collected by intracardiac puncture for the hematological biochemistry tests. Tumor samples and mice organ tissues were resected, fixed and embedded in paraffin for histologic examination. To examine the biological effects of HDAC inhibition in tumors, mice bearing established (about 100,200 mm 3 ) A549 tumors were treated intraperitoneally with a single dose of OSU-HDAC-44 at 60 mg/kg. After treatment for indicated time, tumors were harvested and subjected to Western blot or immunohistochemistry analyses.

Immunohistochemistry (IHC) and fluorescence IHC assays
Tumor tissues from mice exposed to OSU-HDAC-44 were analyzed using IHC assay to detect the expression levels of cleaved caspase-3 protein and were also used for immunofluorescence and confocal microscopic analysis of Aurora B where DAPI was used to stain the DNA. The conditions were as described in the Table  S2.

Statistical analysis
The SPSS program (SPSS Inc. Headquarters Chicago, Illinois) was used for all statistical analysis. Statistical analysis was performed using Student's t-test. Data shown were representatives of at least three independent experiments. Data represent mean 6 SEM. P,0.05 was considered to be statistically significant.  Figure S4 OSU-HDAC-44 increased p21 mRNA and protein levels in a p53-independent manner. (A) A549 (p53 wild-type) and H1299 (p53 null) cells were treated with or without 2.5 mM OSU-HDAC-44 for the indicated times and total RNA was extracted for the quantitative RT-PCR analyses using the specific primers for p21 and GAPDH. Data represent mean 6 SEM from three independent experiments. ***P,0.001. (B) Cells were treated with or without 2.5 mM OSU-HDAC-44 for the indicated times. Immunoblot analyses were performed using the indicated antibodies. The immunoblots shown are representatives of three independent experiments.