Thymosin-β4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

available at www.sciencedirect.com

ScienceDirect www.elsevier.com/locate/molonc

Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma Belamy B. Cheunga,*, Owen Tana, Jessica Koacha, Bing Liua, Michael S.Y. Shumb, Daniel R. Cartera, Selina Suttona, Sela T. Po’uhaa, Louis Cheslerc, Michelle Habera, Murray D. Norrisa, Maria Kavallarisa, Tao Liua, Geraldine M. O’Neillb,d, Glenn M. Marshalla,e,** a

Children’s Cancer Institute Australia, University of New South Wales, Sydney, Australia Kids Research Institute, Children’s Hospital at Westmead, Sydney, Australia c Division of Clinical Studies, Institute of Cancer Research, Sutton, Surrey, UK d Discipline of Paediatrics and Child Health, University of Sydney, Australia e Kids Cancer Centre, Sydney Children’s Hospital, Sydney, Australia b

A R T I C L E

I N F O

A B S T R A C T

Article history:

Retinoids are an important component of neuroblastoma therapy at the stage of minimal re-

Received 26 November 2014

sidual disease, yet 40e50% of patients treated with 13-cis-retinoic acid (13-cis-RA) still

Received in revised form

relapse, indicating the need for more effective retinoid therapy. Vorinostat, or Suberoylani-

17 April 2015

lide hydroxamic acid (SAHA), is a potent inhibitor of histone deacetylase (HDAC) classes I & II

Accepted 17 April 2015

and has antitumor activity in vitro and in vivo. Fenretinide (4-HPR) is a synthetic retinoid

Available online -

which acts on cancer cells through both nuclear retinoid receptor and non-receptor mechanisms. In this study, we found that the combination of 4-HPR þ SAHA exhibited potent cyto-

Keywords:

toxic effects on neuroblastoma cells, much more effective than 13-cis-RA þ SAHA. The 4-

Thymosin-b4 (Tb4)

HPR þ SAHA combination induced caspase-dependent apoptosis through activation of cas-

4-HPR

pase 3, reduced colony formation and cell migration in vitro, and tumorigenicity in vivo. The 4-

SAHA

HPR and SAHA combination significantly increased mRNA expression of thymosin-beta-4

Neuroblastoma

(Tb4) and decreased mRNA expression of retinoic acid receptor a (RARa). Importantly, the

Histone deacetylase inhibitors and

up-regulation of Tb4 and down-regulation of RARa were both necessary for the 4-

retinoic acid receptor alpha (RARa)

HPR þ SAHA cytotoxic effect on neuroblastoma cells. Moreover, Tb4 knockdown in neuroblastoma cells increased cell migration and blocked the effect of 4-HPR þ SAHA on cell migration and focal adhesion formation. In primary human neuroblastoma tumor tissues, low expression of Tb4 was associated with metastatic disease and predicted poor patient prognosis. Our findings demonstrate that Tb4 is a novel therapeutic target in neuroblastoma, and that 4-HPR þ SAHA is a potential therapy for the disease. ª 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

* Corresponding author. Children’s Cancer Institute, Lowy Cancer Research Centre, UNSW, PO Box 81, Randwick 2031, NSW, Australia. Tel.: þ61 2 9385 2450; fax: þ61 2 9662 6584. ** Corresponding author. Kids Cancer Centre, Sydney Children’s Hospital, Level 1, South Wing, High Street, Randwick 2031, NSW, Australia. Tel.: þ61 2 93821721; fax: þ61 2 93821789. E-mail addresses: [email protected] (B.B. Cheung), [email protected] (G.M. Marshall). http://dx.doi.org/10.1016/j.molonc.2015.04.005 1574-7891/ª 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

2

1.

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

Introduction

Neuroblastoma is the most common cancer in infants and accounts for 7% of all cancers in children (Barone et al., 2013). Neuroblastoma begins in embryonal neural crest cells which later give rise to the sympathetic nervous system, and is caused in part by factors which arrest differentiation (Calao and Sekyere, 2013). In the past decade, improved treatment of minimal residual disease by the addition of ch14.18 (a monoclonal antibody against the tumor-associated disialoganglioside GD2), GM-CSF and interleukin-2 to standard isotretinoin (13-cis-RA) therapy has increased survival rates (Yu et al., 2010). However, a considerable proportion of patients still experience relapse and are refractory to conventional treatment approaches. Retinoids are vitamin A analogs required for normal morphogenesis and maintenance of diverse embryologic and adult tissues, which act on cells by binding nuclear receptors (Germain et al., 2002). Retinoids such as all-trans-retinoic acid (aRA), 13-cis-RA, and 4-HPR at low doses are capable of inducing differentiation and apoptosis in cancer cells (Schroeder et al., 2007). Neuroblastoma cells differentiate in response to retinoic acid in vitro, an observation that has led to clinical trials using the 13-cis-RA (Matthay et al., 2009). 4HPR is a synthetic retinoid which works on cancer cells through nuclear receptor-dependent and -independent signaling mechanisms (Sogno et al., 2009). Non-receptor mechanisms of 4-HPR sensitivity include production of ceramide and reactive oxygen species (Maurer et al., 1999). 4HPR has been used as a single agent in early phase clinical trials for neuroblastoma and ovarian cancer (Colombo et al., 2006; Villablanca et al., 2006). Clinical trials have revealed that 4-HPR is a highly active therapeutic and chemopreventive agent with minimal side-effects in neuroblastoma patients (Formelli et al., 2008). A phase I/II trial of oral 4-HPR in children with high-risk, relapsed solid tumors demonstrated minimal 4-HPR toxicity, but only stable disease as the best clinical response. However, newer oral 4-HPR preparations have shown markedly increased bio-availability (Villablanca et al., 2011). In addition, 4-HPR can act synergistically with other anti-cancer compounds (Fang et al., 2011; Shibina et al., 2013). Increased histone deacetylase (HDAC) activity is a common causal factor in human cancer that leads to transcriptional silencing of tumor suppressor genes (Steele et al., 2009). HDAC inhibitors prevent these deacetylases removing acetyl groups from histone tails, thereby promoting gene transcription (Johnstone, 2002). Several HDAC inhibitors have entered early phase trials, and, Vorinostat (SAHA) has been approved for use in adult cutaneous T cell lymphoma (Duvic et al., 2007). The HDAC inhibitor side-effect profile is low when compared with cytotoxic chemotherapy (Hainsworth et al., 2011). However, early phase trials and pre-clinical data suggest that HDAC inhibitors will be most effective when used in combination with other anticancer agents (Liu et al., 2007). Synergistic anti-tumor activity between HDAC inhibitors and retinoids has been observed in a variety of preclinical models (Kato et al., 2007; Spiller et al., 2008). SAHA combined with 13-cis-RA, was well-tolerated in a Phase I/II pediatric trial,

but the best response for relapsed solid tumor patients was stable disease (Fouladi et al., 2010). Two unbiased preclinical screens identified retinoid signal activation as the most effective method of augmenting the HDAC inhibitor anti-cancer signal (Epping et al., 2009, 2007). In this study, we used human neuroblastoma cell lines, flow cytometry, gene-expression analyses, siRNA knockdown, gene expression profiling and a xenograft tumor model to evaluate the 4-HPR þ SAHA combination treatment of neuroblastoma for therapeutic effects and identification of genes which mediated the treatment response. We showed the 4HPR þ SAHA combination has significant cytotoxicity against neuroblastoma cells and identified an actin regulatory protein, thymosin-beta-4 (Tb4), as a critical molecular target of the combination therapy effects on neuroblastoma cell survival, cell migration and focal adhesion formation.

2.

Materials and methods

2.1.

Cell culture and reagents

Neuroblastoma cells, BE(2)-C (MYCN-amplified), SH-SY5Y (MYCN-non-amplified) and NBL-S (MYCN-non-amplified) were purchased from ATCC. They were cultured in DMEM medium (Gibco, Life Technologies) supplemented with 10% fetal calf serum. MRC-5 fibroblast cells were purchased from ATCC. These cells were cultured in alpha-MEM medium (Gibco, Life technologies) supplemented with 10% fetal calf serum. All cell lines were authenticated by CellBank Australia (NSW, Australia). 4HPR, SAHA, and 13-cis-RA were all purchased from Sigma (SigmaeAldrich). Antibody for procaspase 3 and activated-caspase 3 were purchased from Cell Signalling Technology (USA).

2.2.

In vitro combination studies

BE(2)-C, SH-SY5Y, NBL-S and MRC-5 cells were selected for these studies. 4-HPR was combined with SAHA at a fixed ratio of concentrations 6:1, respectively. Cell viability was measured using Alamar blue assay and the fluorescence output was measured on Victor 3 plate reader (Perkin Elmer). The combination index (CI) was calculated based on the ChoueTalalay equation, which takes into account both potency (Dm or IC50) and the shape of the doseeeffect curve.(Chou and Talalay, 1984) CI < 1, CI ¼ 1, CI > 1 indicate synergism, additive effect and antagonism, respectively. CalcuSyn software (Biosoft, Ferguson, MO, USA) was used for the ChoueTalalay combination index analysis.

2.3.

Flow cytometry

Neuroblastoma cell lines; BE(2)-C & SH-SY5Y and normal lung fibroblast; MRC5 were treated with 2uM 4-HPR and 0.33uM SAHA for 48 h then fixed with 80% ethanol. Propidium iodide (PI, 10 ug/ml) (SigmaeAldrich) and RNAse (5 ug/ml) (Roche Applied Science) were added to each sample. Cell cycle and uptake of PI was analyzed on the FACS Calibur (BD Biosciences) and CellQuest software. Measurement of early stages

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

of apoptosis was done using Annexin V-FITC conjugate (Molecular Probes, Life technologies) and 7AAD apoptosis detection kit (BD Pharmingen) following manufacture protocols.

2.4.

Caspase-3 activity assay

BE(2)-C cells were treated with DMSO (control), 2uM 4-HPR, 0.33uM SAHA and combination of SAHAþ 4HPR for 48 h. Cytosolic lysates were extracted to measure caspase-3 activity using the Caspase-3 Colorimetric Assay Kit (Abcam). In brief, 150e200ug proteins were mix with 2x sample buffer and 200uM DEVD-pNA substrate. The pNA light emission was read at 405 nm on the Victor 3 plate reader (Perkin Elmer) at regular one hour time intervals to detect the optimum light emission. Duplicate samples were also performed with the addition of 40uM caspase-3 inhibitor, Z-DEVD-FMK (R&D Systems), to the cytosolic lysate.

2.5.

Colony forming assay

BE(2)-C cells were plated at 50 cells/well in Costar 6 Well ClearTC treated plate a day prior to drug treatment to allow attachment. 2 mM 4-HPR or 0.33 mM SAHA, or both reagents in combination were added to each well with growth medium, and incubated at 37  C, 5% CO2 for 12 days in dark. Growth medium was aspirated at the end of incubation and wash with PBS once. Cells were fixed and stained with 2e3 ml of a mixture of 6.0% glutaraldehyde and 0.5% crystal violet for 30 min. The staining was washed by immersing the plate in tap water and drying at room temperature overnight. Colonies with minimum of 20 cells were counted.

(RMA). In three comparisons, (4-HPR vs. control, SAHA vs. control and 4-HPR þ SAHA vs. control), the gene expression greater than 2 fold was considered to be differentially expressed. The pathway analysis was conducted with the differentially expressed genes against KEGG using DAVID (Huang da et al., 2009) and WebGestalt (Zhang et al., 2005).

2.8.

RNA interference

Cells were reverse transfected with 40 nM siRNA duplexes against Tb4 and control siRNA (Dharmacon, Thermo Scientific) using LipofectamineTM 2000 (Invitrogen, Life Technologies) and Opti-MEM (Gibco, Life Technologies) for 24 and 48 h. Cells were then used in cell viability assays or their RNA was isolated for real-time PCR analysis.

2.9.

Real-time PCR

RNA from cell lines was synthesized into cDNA using SuperScript III reverse transcriptase (Invitrogen, Life Technologies) and Oligo (dT)18 (Bioline) according to manufacture protocols. Real-time PCR was performed using Power SYBR Green mix (Applied Biosystems, Life Technologies) and PCR was carried out on the ABI 7500 thermo cycler (Applied Biosystems). Relative gene expression data was analyzed using the 2-DDCT method. Forward sequences and reverse sequences: Tb4: 50 CAGAGACGCAAGAGAAAAATCCA 30 and 50 CGATTCGCCTGCTTGCTT 30 . b2M: 50 ACTGGTCTTTCTATCTCTTGTACTACACTGA 30 and 50 TGATGCTGCTTACATGTCTCGAT 30 .

2.10. 2.6.

3

Chromatin immuno-precipitation (ChIP) assays

In vivo tumorigenic assay

The 4-HPR and SAHA anti-tumor efficacy studies were performed by GenScript (Piscataway, NJ 08854, USA) in a xenograft tumor mice model. 5  106 of BE(2)-C cells were subcutaneously injected into the right flank of a cohort of 32 female BALB/c nude mice. When the tumor size reached a volume of 54 mm3, the tumor-bearing nude mice were randomly assigned to 4 treatment groups: vehicle control, 4-HPR (1.45 mg/kg i.v), SAHA (35 mg/kg i.p) or 4-HPR þ SAHA combination for 18 days (Hill et al., 2009; Reddy et al., 2004). Tumor volume was measured every 2 days with a caliper, and the volume was expressed in mm3 using the formula: Volume ¼ ½  A  B2 where A and B are the long and short diameters of the tumor, respectively.

Standard ChIP assays were performed as described (Iraci et al., 2011). Antibodies used in this study were as follows: IgG (sc2027, Santa Cruz Biotechnology, Santa Cruz, CA), RARa (Santa Cruz Biotechnology). The primers used were as follows: Tb4-A, 50 -ttagttattatgcactggtccccaa-30 (forward) and 50 -ccctccaacacttcataacacattt-30 (reverse); Tb4-B, 50 -tcgaaataaaagaatgcctgtcacc-30 (forward) and 50 -cttggggaccagtgcataataacta-3 (reverse); Tb4-C, 50 -aaatcacttaagggtgctttacgtc-30 (forward) and 50 -tacgattcctgtgactcagctttc-30 (reverse); Tb4-intron A, 50 -cactgcgcagaccagactt30 (forward) and 50 -ttagggaatgccttcacagg-30 (reverse); Tb4-intron B, 50 - cagaccagacttcgctcgta-30 (forward) and 50 - ttagggaatgccttcacagg-30 (reverse); Control (3000 bp), 50 - tagccaacttctccatggttagaaa-30 (forward) and 50 - ctggaaagcaaaataaaactgcgtc-30 (reverse).

2.11. 2.7. Differential gene expression and pathway analysis on microarray data Microarray data was profiled using Affymetrix HuGene 1.0 st v1. BE(2)-C cells were treated with solvent control, 2 mM of 4HPR, 0.33 mM of SAHA or 4HPR þ SAHA for 24 h. Differential gene analysis was conducted in R (http://www.rproject.org/). The raw gene expression data was loaded into R followed by analysis with packages from Bioconductor (http://www.bioconductor.org/). Using affy package (Gautier et al., 2004), the expression data were subjected to background correction and normalization with robust multi-array average measure

In vitro wound healing assay

BE(2)-C cells were plated in culture insert (Ibidi, Martinsried, Germany) and allowed to reach 90% confluency overnight. Treatment drugs (4-HPR, SAHA, or both) were added to the culture medium and incubated for 24 h. The culture insert was then removed, leaving a uniform gap for cells to migrate toward. Cells were monitored and images were taken every hour with Zeiss Axiovert 200M live cell imager up to 13 h. The wound was analyzed with AxioVision 4.8 software. Migration index (%) was computed as following, (“wound area at beginning” -wound area at 13 h)/(wound area at beginning) 100%.

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

4

2.12.

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

Invasion assay

Cell invasion assays were performed as previously described (Lees et al., 2011). Following gel polymerization, the media containing drug was added to the wells (0.75 mM 4-HPR and 0.125 mM SAHA alone or in combination). After incubation for 72 h, the cultures were fixed with 4% paraformaldehyde and then stained with fluorescently-tagged phalloidin and nuclei counterstained with DAPI. Images of stained cells were captured using a Leica SP5 confocal scanning laser microscope (Leica Microsystems, Germany) with a 10 X air objective. Optical sections were captured in 3 mm steps beginning with the bottom layer of cells underneath the collagen gel. Maximum projections of z-stacks were generated using the Leica software, final micrograph images and grey level adjustments were prepared in Adobe Photoshop.

2.13.

Immunofluorescence for focal adhesions

Immunofluorescence was carried out as described previously (Bell et al., 2013). Cells were grown on 8-well chamber slides prior to fixing with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and stained with anti-paxillin (Abcam, Cambridge, UK) to visualize the focal adhesions. Secondary antibody staining with Alexa Fluor 555-conjugated antimouse antibody (Molecular Probes; Invitrogen) was applied. Phalloidin conjugated with Alexa Fluor 488 (Life Technologies) was applied to visualize the actin stress fibers. Vectashield mounting medium with DAPI (Vector Laboratories, Burlingame, CA, USA) was used to visualize nuclear material. Confocal microscopy images were acquired using an Olympus FluoView FV1000 confocal microscope (Olympus, Tokyo, Japan), 40 objective lens, NA 1.30. A Z series of images of 0.44 mm/section was collected using sequential scanning. Images were processed using the Olympus FluoView software (Olympus).

2.14.

Statistical analysis

All statistical tests were performed in GraphPad Prism5 (GraphPad Software Inc.) All experiments were performed in minimum of triplicates. Two-way ANOVA with Bonferroni post-test and the One-way ANOVA with Tukey’s post-test were performed for the statistical analysis. The two-sided student t-tests were used in the tumor volume analysis.

3.

HPR þ SAHA showed slight to moderate synergism in BE(2)C and NBL-S neuroblastoma cell lines, with a CI < 1 for three of the four drug concentrations tested. CI was 0.94 for BE(2)C cells and 0.95 for NBL-S cells at 2 mM 4-HPR þ 0.33 mM SAHA after 48 h treatment. CI was 0.85 for BE(2)-C cells and 0.82 for NBL-S cells at 3 mM 4-HPR þ 0.5 mM SAHA. CI was 0.47 for BE(2)-C cells and 0.86 for NBL-S cells at 4.5 mM 4HPR þ 0.75 mM SAHA for 48 h 4-HPR þ SAHA showed an additive effect on treated SH-SY5Y cells at two concentrations after 48 h treatment (CI ¼ 1 for 3 mM 4-HPR þ 0.5 mM SAHA and 4.5 mM 4-HPR þ 0.75 mM SAHA) (Chou, 2006). SAHA and 4HPR as single agents, or in combination, were minimally toxic to treated MRC-5 normal fibroblasts at concentrations up to 4.5 mM for 4-HPR and 0.75 mM for SAHA (Figure 1D). In contrast, the combination of 2 mM 13-cis-RA and 0.33 mM SAHA exhibited no significant synergy or therapeutic toxicity in treated BE(2)C cells (Figure 1E).

Results

3.1. 4-HPR þ SAHA combination therapy has significant cytotoxicity to neuroblastoma cells We first examined the cytotoxic effect of 4-HPR þ SAHA on three human neuroblastoma cell lines (BE(2)-C, SH-SY5Y and NBL-S) by Alamar blue assay at a range of different concentrations (4-HPR: 1.33 mMe4.5 mM, and SAHA: 0.22 mMe0.75 mM) using a 6:1 molar ratio of each compound (Figure 1AeC). To assess drug synergy the Combination Index (CI) values were calculated by CalcuSyn at the fixed ratio of the drug concentrations, which tested for the cytotoxicity assays. 4-

3.2. 4-HPR þ SAHA combination therapy induced caspase-dependent apoptosis through activation of caspase 3 Since both 4-HPR þ SAHA can individually induce apoptosis in neuroblastoma cell lines (Armstrong et al., 2012; De los Santos et al., 2007), we examined the apoptotic effects of the combination in two neuroblastoma cell lines. We observed significantly higher levels of the sub-G1 apoptotic cell population in BE(2)-C and SH-SY5Y neuroblastoma cells when the two drugs were combined, in comparison with either drug alone (Figure 2A). The combination did not induce apoptosis in MRC-5 normal fibroblast cells using propidium iodide staining and flow cytometric analysis (Figure 2A). We next examined the early and late stages of apoptosis, by treating neuroblastoma cells with either 2 mM 4HPR or 0.33 mM SAHA alone, 4-HPR þ SAHA, or 13-cis-RA þ SAHA for 48 h, then stained with Annexin V-Fluorescein isothiocyanate (FITC) and 7-Amino Actinomycin D (7AAD). The 4-HPR þ SAHA combination induced a marked increase in Annexin V positive neuroblastoma cells when compared with either drug alone, or a combination of 13-cis-RA þ SAHA in both BE(2)-C and SH-SY5Y cells (Figure 2B and C).The percentage of early (Annexin V positive and 7AAD negative) and late apoptotic cells (both Annexin V and 7AAD positive) was significantly higher for the 4-HPR þ SAHA combination (Supplementary Figure 1 and Supplementary Figure 2). To investigate the mechanism of the 4-HPR þ SAHA-induced apoptosis, BE(2)-C and SH-SY5Y cells were treated with either 2 mM 4HPR or 0.33 mM SAHA alone, or both reagents for 48 h and then protein lysates were immunoblotted for the protein level of activated caspase 3. Treatment with 4-HPR þ SAHA caused increased activated caspase-3 protein levels at 48 h in both BE(2)-C and SH-SY5Y (Figure 2D and E). In addition, BE(2)-C cells were treated with DMSO control, 2 mM 4-HPR, 0.33 mM SAHA and combination of SAHA þ 4-HPR for 48 h. Cytosolic lysates were extracted to measure the caspase-3 enzymatic activity using the Caspase-3 Colorimetric Assay Kit. The combination of 4-HPR and SAHA treatment also increased the level of caspase 3 enzymatic activity in comparison with the treatment of the two single agents, and this caspase 3 enzymatic activity was blocked by a caspase 3 inhibitor in BE(2)-C cells (Figure 2F).

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

5

Figure 1 e The 4-HPR D SAHA combination demonstrates strong cytotoxicity on neuroblastoma cells with limited toxicity to normal MRC-5 fibroblasts. (AeD) BE(2)-C, SH-SY5Y, NBL-S neuroblastoma and MRC5 lung fibroblast cells were treated with either 4-HPR, or SAHA, or the combination of both, at a range of concentrations for 48 h, with the cell viability measured by Alamar blue assay. Statistical significance: Two-way ANOVA with Bonferroni post-test was performed. *p < 0.05, **p < 0.01, ***p < 0.001. (E) BE(2)-C cells were treated with 2mM13-cis-RA, or 0.33 mM SAHA, or the combination of both for 48 h, before performing Alamar blue assay. N.S [ not statistically significant.

3.3. 4-HPR þ SAHA combination therapy reduces colony formation in vitro and tumorigenicity in vivo We next performed clonogenicity assays on BE(2)-C cells treated with 4-HPR þ SAHA. Colonies in vitro were defined as cluster of 20 cells or more after 12 days (Franken et al., 2006). While colony formation was significantly reduced by 4-HPR or SAHA alone, combination therapy completely abrogated colony formation (Figure 3A). To determine whether 4HPR þ SAHA had a synergistic activity in vivo, we selected

BE(2)-C, a cell line which exhibited strong synergy in vitro. While 4-HPR or SAHA alone reduced tumor volume by 30% of control at day 18 of treatment, combination therapy had a significantly greater effect, reducing tumor volume by 50% at day 18 (The control vs SAHAþ4-HPR: P ¼ 0.0023, SAHA vs SAHAþ4-HPR: P ¼ 0.0238 and 4-HPR vs SAHAþ4-HPR: P ¼ 0.0486). (Figure 3B and Supplementary Table 1). During the experimental period, the mice in all treated groups showed no decrease in body weights (Supplementary Figure 3).

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

6

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

Figure 2 e The combination of 4-HPR and SAHA induced caspase-dependent apoptosis through activation of Caspase 3. (A) BE(2)-C, SH-SY5Y and MRC5 cells were treated with either 2 mM 4HPR, or 0.33 mM SAHA alone, or both reagents for 48 h. The cells were stained with propidium iodide (PI), and the percentage of cell population in sub-G1, G1, S and G2 phases were analyzed on the FACS Calibur. (BeC) BE(2)-C cells and SH-SY5Y cells were treated with either 2 mM 4HPR or 0.33 mM SAHA alone, 4HPR D SAHA or 2 mM of 13-cis-RAD 0.33 mM of SAHA combinations for 48 h then stained with annexin V-Fluorescein isothiocyanate (FITC) and 7-Amino Actinomycin D (7AAD), and the proportion of Annexin V positive cells were analyzed on the FACS Calibur. Statistical significance: One-way ANOVA with Tukey’s post-test. *p < 0.05, **p < 0.005, ***p < 0.0001. The “n.s” for no statistical significance. (D) The BE(2)-C and SH-SY5Y cells were treated with either 2 mM 4HPR or 0.33 mM SAHA alone, or both reagents for 48 h. The level of activated Caspase 3 protein was determined from cytosolic protein lysates by Western blot. Pro-caspase 3 and activated caspase 3 antibodies were used to detect changes in caspase 3 protein expression. (E) Quantification of western blotting with an anti-Caspase 3 antibody by densitometry. (F) The BE(2)-C cells were treated with either 2 mM 4HPR or 0.33 mM SAHA alone, or both reagents for 48 h in the absence or presence of a caspase 3 inhibitor, Z-VAD-fmk (40 mM). The caspase 3 activity was measured by a colorimetric caspase 3 activity assay. The One-way ANOVA with Tukey’s post-test was used for the statistical analysis.

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

7

Figure 2 e (continued).

3.4. Actin binding protein, Tb4, expression is induced by 4-HPR þ SAHA treatment To identify factors which regulated the sensitivity of neuroblastoma cells to 4-HPR þ SAHA, we performed an Affymetrix microarray of transcriptomes in BE(2)-C cells treated with solvent control, 2 mM 4-HPR, 0.33 mM SAHA or the combination for 24 h. Pathway analysis of differentially expressed genes against KEGG using DAVID (Huang da et al., 2009) and

WebGestalt,(Zhang et al., 2005) showed a number of signaling pathways which were significantly altered by the combination treatment in comparison with untreated control and the single agents alone, such as, focal adhesion formation, cell adhesion molecules and regulation of actin cytoskeleton pathways from the KEGG database (Figure 4A and B). One of the most significantly up-regulated genes by the combination was Tb4 (TMSB4X). On the microarray Tb4 was induced 2.9-fold by 4-HPR, 4.6-fold by SAHA, and 19.39-fold

Figure 3 e The 4-HPR and SAHA combination blocked colony formation and reduced tumor growth of mouse xenograft. (A) The clonogenicity of BE(2)-C cells was determined using soft agar. The number of colonies were compared between untreated control, 2 mM 4HPR or 0.33 mM SAHA, or both reagents treated BE(2)-C cells for 14 days. Statistical significance: One-way ANOVA with Tukey’s post-test. *p < 0.001, ***p < 0.0001. (B) The tumor volume of BE(2)-C cells injected into Balb/c nude mice and treated with respective drugs for 18 days. Each value shown in the figure represented data from 8 mice. Student T-test was performed to analyze tumor volume at day 18: vehicle control vs SAHAD4-HPR: P [ 0.0023; SAHA vs SAHAD4-HPR: P [ 0.0238 and 4-HPR vs SAHAD4-HPR: P [ 0.0486. Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

8

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

Figure 4 e The Tb4 gene expression is modulated by the combination treatment and the high level of Tb4 expression is associated with good prognosis in neuroblastoma patients. (A) A heatmap of top 60 differentially expressed genes comparing 4-HPR D SAHA combination treatment with untreated control are showed. The expressions of these genes in 4-HPR and SAHA as single agents are also showed in the heatmap. A number of genes up-regulated in the 4-HPR D SAHA combination treatment which associated with the pathways were indicated by arrows. (B) The top 15 pathways enriched by GSEA. GSEA revealed that pathways including focal adhesion, cell adhesion molecules, and regulation of action cytoskeletion are enriched by comparing the 4-HPR D SAHA combination versus solvent treated cells. The C2 collection of the Molecular Signatures Database v3.0 was queried for the enrichment. The table lists gene sets enrichment score (ES), normalized enrichment score (NES) and the relevant nominal p-value (NOM p-val), with false discovery rate (FDR q-val) less than 0.25, and familywise-error rate (FWER p-val) less than 0.05. (C) The fold changes of Tb4 gene expression in BE(2)-C cells with 4-HPR alone, SAHA alone or 4-HPR D SAHA for 24 h in comparison with untreated control in the microarray study. (D) The mRNA expression of Tb4 was analyzed by RT-qPCR following BE(2)-C cells treated with

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

9

Figure 4 e (continued). by the combination treatment, compared to untreated control (Figure 4C). Real-time RT-PCR analysis confirmed similar levels of Tb4 induction (Figure 4D). Tb4 is a polypeptide involved in regulation of the actin cytoskeleton and cell migration, which can demonstrate the characteristics of an oncogene or a tumor suppressor gene, dependent on the

cellular context (Caers et al., 2010a; Cha et al., 2003). Thus, we next analyzed publically available mRNA expression array datasets from primary human neuroblastoma tumor tissues for a relationship between Tb4 expression level and patient prognosis (Kocak et al., 2013). When patient cohorts were subdivided around the median Tb4 expression level, we found

respective drugs for 24 h. Statistical significance: ***p £ 0.009. (E) KaplaneMeier curve showed the probability of overall survival of patients according to the level of Tb4 expression in the 102 neuroblastoma patients in the Seeger dataset (Sourced from publically available Oncogenomics e Seeger database). (FeG) KaplaneMeier curves showed the probability of overall survival and Event-free survival of patients according to the level of Tb4 expression in the 477 neuroblastoma patients in the Kocak dataset (R2 microarray analysis and visualization platform, http:// r2.amc.nl). CI [ confidence interval; HR [ hazard ratio. P value was obtained from two-sided log-rank test. (H) expression of Tb4 was evaluated in tumors from 477 neuroblastoma patients and dichotomized by known clinical indicators: stage (INSS 1,2,4S vs Stage 3,4), MYCN amplification status (non-amplified vs amplified) and age (<18 months vs > 18 months). Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

10

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

high Tb4 expression strongly correlated with good patient prognosis in two separate cohorts of 102 and 477 neuroblastoma patients (Seeger dataset: p < 0.0001; Kocak dataset: HR ¼ 3.85, 95% CI ¼ 2.38e6.23, p ¼ 3.13ee09 for the overall survival and HR ¼ 2.01, 95% CI ¼ 1.45e2.79, p ¼ 2.01ee05 for the

event-free survival) (Figure 4E, F and G). Gene expression analysis for Tb4 revealed that higher expression correlated with known indicators of good prognosis in neuroblastoma, including early clinical stage, MYCN non-amplification and younger patient age (Figure 4H).

Figure 5 e Down-regulation of RARa enhances effect of 4-HPR and SAHA combination and up-regulates Tb4 mRNA expression. (A) The mRNA expression of RARa was analyzed by RT-qPCR following BE(2)-C and SH-SY5Y cells treated with respective drugs for 48 h. Statistical significance: **p < 0.005 and ***p < 0.0005. (B) The BE(2)-C cells were transfected with scrambled siRNA control or two RARa specific siRNAs and treated with respective drugs for 48 h, and the cell viability were analyzed by Alamar Blue assay. (C) The BE(2)-C cells were stably transfected with empty vector or MEP-RARa expression vector and treated with either 2 mM 4HPR, or 0.33 mM SAHA or both reagents for 48 h. The cell viability was analyzed by Alamar Blue assay. Statistical significance: **p < 0.01, ***p < 0.001. (D) The BE(2)-C cells were transfected with scrambled siRNA control or two RARa specific siRNAs for 96 h, and the mRNA expression of RARa and Tb4 were analyzed by RT-qPCR. The protein expression of RARa was analyzed by western blot. Statistical significance between RARa 1 and RARa 2 siRNAs knockdown compared to siRNA control: ***p < 0.001, **p < 0.01 and *p < 0.05. Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

The nuclear retinoid receptor and signaling molecule, retinoic acid receptor alpha (RARa), has been shown to be a resistance factor for SAHA effects on cancer cells (Epping et al., 2007). Indeed, in its unliganded state, RARa acts as a transcriptional trans-repressor. Thus, we examined the hypotheses that RARa was a resistance factor for 4-HPR þ SAHA in neuroblastoma cells and a repressor of Tb4 transcription. The expression of RARa was significantly repressed by 4HPR þ SAHA in treated BE(2)-C and SH-SY5Y cells compared with 4-HPR alone and with SAHA alone, whereas 13-cisRA þ SAHA had no effect on RARa (Figure 5A). Transient transfection of two RARa-specific siRNAs reduced cell viability in control, single agent or combination treated neuroblastoma cells (Figure 5B). In contrast, RARa overexpression blocked the effect of combination therapy on cell viability (Figure 5C). RARa protein expression was decreased by single agent and combination treatment (Supplementary Figure 4). Moreover, knockdown of RARa by two RARa-specific siRNAs significantly induced Tb4 expression and the knockdown of RARa mRNA and protein were confirmed by RT-PCR and western blot. (Figure 5D). However, we were unable to demonstrate direct binding of the RARa protein to the Tb4 gene promoter using ChIP assay in the absence and presence of the combination treatment, whereas RARa directly bound the RARb promoter region in the assay, as has been reported (Supplementary Figure 5 and data not shown) (Wang et al., 2010). These data suggested RARa indirectly repressed Tb4 expression in neuroblastoma cells and was itself a repression target of 4-HPR þ SAHA treatment.

neuroblastoma cells (Figure 6F and Supplementary Figure 7). Since Tb4 affects actin function and the actin cytoskeleton during focal adhesion formation (Caers et al., 2010a; Wozniak et al., 2004), we next investigated whether Tb4 plays a role in focal adhesion formation in the combination treatment. We showed that 4-HPR þ SAHA markedly reduced actin filaments in neuroblastoma cells stained with the highaffinity filamentous actin (F-actin) probe, phalloidin (Figure 6G). Focal adhesions are coupled to bundles of polymerized actin filaments or stress fibers. The disassembly of stress fibers is accompanied by a breakdown of the focal adhesion tethered at the end of the actin filament bundle. When we treated neuroblastoma cells with 4-HPR þ SAHA, we found focal adhesion formation was reduced 3-fold by SAHA alone or 4-HPR þ SAHA treatment, compared with 4-HPR or solvent control alone (Figure 6H). However, the cells treated with SAHA alone still maintained similar cell shapes as untreated cells, but we have observed some changes in cell shape in the combination treated cells, which may due to the effect of the combination treatment on the structure of actin filaments (Figure 6G). Most importantly, knockdown of Tb4 expression completely reversed the 4-HPR þ SAHA treatment effect on focal adhesion formation and was also accompanied by reversal of the cell shape effects in the 4-HPR þ SAHA treated cells (Figure 6H), indicating Tb4 expression was necessary for the effect of 4-HPR þ SAHA on focal adhesion formation.

4. 3.5. Up-regulation of Tb4 is necessary for the 4HPR þ SAHA effects on cell viability, cell migration and focal adhesion formation To determine whether induction of Tb4 expression in neuroblastoma cells enhanced the cytotoxic effect of 4HPR þ SAHA, we transiently transfected neuroblastoma cells with siRNAs targeting Tb4 for 24 h, then treated cells with single agents or the combination for 48 h. Neuroblastoma cells transfected with the two Tb4 siRNAs exhibited a marked resistance to the cytotoxic effects of the 4-HPR þ SAHA in both BE(2)-C and SH-SY5Y cells (Figure 6A and B). Since Tb4 regulates the cytoskeleton, we next assessed the role of combination therapy and Tb4 on neuroblastoma cell migration using in vitro scratch-wound assays on confluent monolayers (Marshall et al., 2010). BE(2)-C and SH-SY5Y cells treated with 4-HPR þ SAHA migrated slower than the single agents or untreated control at 12 h from the time of wounding (Figure 6C and Supplementary Figure 6). In contrast, 13-cisRA þ SAHA treated cells did not reduce cell migration in comparison with control or single agent (Figure 6C and Supplementary Figure 6). An in vitro collagen gel invasion assay showed that 4-HPR and SAHA alone were as effective as the combination in reducing invasion of BE(2)-C cells (Figure 6D and E), indicating 4-HPR þ SAHA has stronger effect in decreasing cell migration, but not invasion, in comparison with single agents. Furthermore, we found that knockdown of Tb4 increased migration of untreated and 4-HPR þ SAHA treated neuroblastoma cells, suggesting Tb4 expression is required for 4-HPR þ SAHA reduced cell migration in

11

Discussion

Preclinical studies suggested that combination of HDAC inhibitor and retinoid therapy would prove to be an effective anticancer treatment in the clinic, however, this promise has not been reflected in early phase trials. Here we provide evidence of a novel and potent combination between a synthetic retinoid, 4-HPR, and a HDAC inhibitor, SAHA, which was far more effective than the retinoid in current clinical use in neuroblastoma, either 13-cis-RA alone, or in combination with SAHA. The molecular mechanisms underlying the cytotoxic action studies has identified a novel anticancer treatment target, Tb4, a regulator of the actin cytoskeleton, and consequent inhibitory effects on cell migration and focal adhesion formation. Taken together, our data suggest a unique 4HPR þ SAHA treatment effect on up-regulation of Tb4, inhibition of focal adhesions and cell migration, with important implications for the prevention of metastasis in patients with minimal residual neuroblastoma (see the schematic overview in Figure 7). In the neuroblastoma cell lines and drug concentrations tested in this study, we found that 4-HPR and SAHA alone had a modest effect on cell viability and apoptosis in BE(2)-C cells, whereas 4-HPR and SAHA alone had no effect on apoptosis in SH-SY5Y cells. However, in both cell lines, the combination of the two drugs was significantly induced apoptosis and reduced cell viability. We observed that the combination of these two agents had minimal toxic effects on normal human fibroblasts in vitro. The combination of 4HPR þ SAHA increased activated caspase-3, indicating that the cytotoxic effects of this drug combination may act via a

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

12

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

Figure 6 e Up-regulation of Tb4 is necessary for 4-HPR and SAHA reduced cell viability and focal adhesion formation. (A) BE(2)-C and SHSY5Y cells were transfected with control siRNA or two Tb4 specific siRNAs for 24 h then treated with respective drugs for 48 h, and the cell viability was analyzed by Alamar Blue assay. Statistical significance difference between control siRNA and Tb4 siRNA: *p < 0.05, **p < 0.01, ***p < 0.001. (B) Knockdown of Tb4 was confirmed with RT-qPCR analysis for Tb4 mRNA expression BE(2)-C and SH-SY5Y cells. Statistical significance difference between control siRNA and Tb4 siRNA: *p < 0.05 and **p < 0.01. (C) The histogram displays the proportion of wound closure over 12 h from the time of wounding for BE(2)-C treated with respective drugs (solvent control, 0.75 mM of 4HPR D 0.125 mM of SAHA, 0.75 mM of 13-cis-RA or 0.75 mM of 13-cis-RA D 0.125 mM of SAHA combination) for 24 h. Migration index (%) [ (“starting wound area” e “ending wound area”) and divided by “starting wound area”. Statistical significance: *p < 0.05 and **p < 0.01. (D) Invasion of cells under the indicated conditions into 3D collagen gels. After invasion, cells were fixed and immunostained with phalloidin and DAPI. Cells were then analyzed by confocal microscopy, with images taken every 3 mm. Arrow (scale bar) indicates direction of invasion. (E) Percentage of cell invasion (positive pixels at 36 mm divided by total positive pixels at 0 mm). Bars show the average of 5 separate experiments and error bars show the SEM. *p < 0.05, One-way ANOVA with Tukey’s post-test. (F) The histogram displays the proportion of wound closure over 12 h from the time of wounding for BE(2)-C transfected with control siRNA or Tb4 specific siRNAs for 24 h then treated with respective drugs for 24 h (DMSO solvent control; and 0.75 mM of 4HPR D 0.125 mM of SAHA). P < 0.05. Statistical analysis: Two-way ANOVA, Dunnett post-test. , *p < 0.05, n [ 3. Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

13

Figure 6 e (continued).

caspase-dependent apoptotic pathway. However, 4-HPR has been reported to be cytotoxic agent and 13-cis-RA is a differentiating agent for neuroblastoma cell lines (Ponzoni et al., 1995). In this current study, we did not perform experiments to determine the role of 4-HPR in differentiation in neuroblastoma cells. In addition, the 13-cis-RA reduces cell proliferation after one week of treatment, therefore, therefore, more

studies are required for the comparison of in vitro and in vivo drug efficacy between 4-HPR þ SAHA and 13-cis-RA þ SAHA combinations, especially in the induction of cell differentiation in neuroblastoma. Tb4 forms a 1:1 complex with G (globular) actin, sequestering G-actin monomers from polymerizing to form F (filamentous) actin which comprise the cellular microfilaments

(G) BE(2)-C cells were treated with 4-HPR, SAHA, or combination of both drugs for 24hr. Cells were then fixed and stained with anti-Paxillin (top row). Merged images were shown at the bottom row, with Paxillin represented in red, phalloidin represented in green, and DAPI in blue. The numbers of focal adhesion observed per mm2 of cell area were quantitated using Axiovision (Carl Zeiss). (H) The quantification of the focal adhesion formation in BE(2)-C cells transfected with either control siRNA or Tb4 siRNA for 24 h and treated with respective drugs for an additional 24 h. Error bars show the SEM, **p < 0.01, *p < 0.05, One-way ANOVA with Tukey’s post-test. Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

14

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

Figure 7 e Schematic overview of the signaling events that induced by 4-HPR D SAHA in neuroblastoma cells. The mechanism of synergy was dependent on repression of RARa expression, and induction of caspase 3 and Tb4. Treated neuroblastoma cells exhibited reduced cell viability, cell migration and increased apoptosis, in a Tb4dependent manner.

needed for the normal cellular cytoskeleton and motility (Carlier et al., 1996; Shum et al., 2011). Different studies indicated a pivotal role of Tb4 in the metastatic process of solid tumors. Up-regulation of Tb4 was found in Ras-transformed fibroblasts and metastatic human melanoma cells (Nummela et al., 2006). Overexpression of the Tb4 gene was associated with increased invasion of SW480 colon carcinoma cells and the distant metastasis of human colorectal carcinoma (Wang et al., 2004). Our study suggests the role of Tb4 in motility of cancer cells is context dependent, since we found that low levels of Tb4 correlated with advanced clinical stage in vivo and migration in vitro of neuroblastoma cells. Cell migration is initiated by protrusion at the leading edge of the cell, the formation of peripheral focal adhesions, the exertion of force on these adhesions, and finally the release of the adhesions at the rear of the cell (Tilghman et al., 2005). Treatment with exogenous Tb4 resulted in the formation of smaller focal adhesions in human conjunctival cells (Sosne et al., 2002). We hypothesize that Tb4 may play a role in the inhibitory effects of 4-HPR þ SAHA on focal adhesion formation in neuroblastoma cells. Indeed, we found that Tb4 knockdown completely reversed the 4-HPR and SAHA treatment effect on focal adhesion formation, suggesting that 4-HPR and SAHA treatment alters the actin cytoskeleton in a Tb4dependent manner in neuroblastoma cells. While some studies have suggested that Tb4 has a role in metastasis (Wang et al., 2004) (Xiao et al., 2012), Caers et al. have shown a tumor suppressive function of Tb4 in myeloma development (Caers et al., 2010a). Tb4 is a poor prognostic factor in a number of different types of tumors and a good prognostic indicator in multiple myeloma (Caers et al., 2010a; Xiao et al., 2012; Caers et al., 2010b). Our finding indicated that the neuroblastoma patients with high expression of Tb4 not only have longer event-free survival, but other good prognostic indicators in neuroblastoma, such as early clinical stage, MYCN non-amplification and younger patient age were also associated with high Tb4 levels. The mechanistic explanation of this discrepancy is not clear, and the functional role of Tb4 in neuroblastoma is currently unknown. Our data showed that up-regulation of Tb4 was required for the 4HPR þ SAHA effects on cell viability, and down-regulation of

Tb4 significantly increased cell migration, suggesting that induction of Tb4 limits neuroblastoma cell migration and leads to death of neuroblastoma cells. The explanation for the different functions of Tb4 among cancers of different tissue origins may relate to tissue of origin-specific differences in the regulation of cytoskeleton, but will require further investigation. RARa is a nuclear hormone receptor and dual-function transcription factor, which is bound to co-repressor complexes in the unliganded state, but forms heterodimers with other retinoid receptors and can transactivate its target genes, upon retinoid ligand binding (Glass and Rosenfeld, 2000). RARa, and, preferentially expressed antigen of melanoma, both repressor proteins for the retinoid signal, were shown to mediate resistance to HDAC inhibitors (Epping et al., 2007). Furthermore, RARa-deficient cells showed enhanced sensitivity to HDAC inhibitors in vitro and in vivo (Epping et al., 2007). Our data indicates that RARa acted as a resistance factor for the anti-cancer effect of SAHA and 4HPR, and down-regulation of RARa potently reverses this inhibitory effect, allowing an increase in Tb4 and reducing cell viability. We have found a consensus retinoic acid response element in the promoter of Tb4 gene, which matches the preferred binding site of RARa (Popperl and Featherstone, 1993). However, we were unable to demonstrate that direct binding of RARa to the Tb4 gene promoter by chromatin immuno-precipitation. Future studies are required to determine whether RARa decreases Tb4 mRNA expression through post-transcriptional events involving mRNA, or non-coding RNA-mediated post-transcriptional control (Belasco, 2010). Our in vitro results show that 4-HPR þ SAHA therapy exhibited high levels of synergy at low concentrations, however, our single in vivo murine xenograft experiment using BE(2)C cells showed only an additive therapeutic effect of the two agents. Indeed, 13-cis-retinoic acid was rejected as a treatment in neuroblastoma patients, despite a wealth of in vitro evidence of its therapeutic potential, when in vivo experiments and an early phase single agent human trial showed little effect against large solid tumor masses (Reynolds and Reynolds, 2004). However, subsequent randomized clinical trials demonstrated that 13-cis-retinoic acid was effective in preventing disease relapse at the stage of minimal residual neuroblastoma (Matthay et al., 2009). Furthermore, another study found that retinoic-acid-resistant neuroblastoma cell lines show altered MYC regulation and high sensitivity to 4HPR, suggesting that therapy with 4-HPR may be especially effective if used against minimal residual neuroblastoma (Reynolds et al., 2000). Therefore, we hypothesize that 4HPR þ SAHA therapy may be also more effective against minimal residual neuroblastoma. However, to determine the effects of 4-HPR þ SAHA on minimal residual neuroblastoma and metastasis in vivo, future studies will need to be performed in the mouse models of minimal residual neuroblastoma and metastases. The concentrations of 4-HPR and SAHA used in this study were within the range achieved in children in clinical trials (Fouladi et al., 2010; Villablanca et al., 2011). Since both 4-HPR and SAHA are in clinical use and have relatively low toxicity profiles, this combination should be assessed in a clinical trial setting in the future.

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

Conflict of interest The authors declare no conflict of interest.

Acknowledgments This work was supported by Program Grants from the NHMRC Australia (grant number APP1016699); Cancer Institute NSW, Australia (grant number 10/TPG/1-03); and Cancer Council NSW, Australia (grant numbers PG-11-06 and RG12-06). Children’s Cancer Institute Australia is affiliated with the University of New South Wales and Sydney Children’s Hospital.

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.molonc.2015.04.005 R E F E R E N C E S

Armstrong, J.L., Martin, S., Illingworth, N.A., Jamieson, D., Neilson, A., Lovat, P.E., Redfern, C.P., Veal, G.J., 2012. The impact of retinoic acid treatment on the sensitivity of neuroblastoma cells to fenretinide. Oncol. Rep. 27, 293e298. Barone, G., Anderson, J., Pearson, A.D., Petrie, K., Chesler, L., 2013. New strategies in neuroblastoma: therapeutic targeting of MYCN and ALK. Clin. Cancer Res. 19, 5814e5821. Belasco, J.G., 2010. All things must pass: contrasts and commonalities in eukaryotic and bacterial mRNA decay. Nat. Rev. Mol. Cell Biol. 11, 467e478. Bell, J.L., Malyukova, A., Kavallaris, M., Marshall, G.M., Cheung, B.B., 2013. TRIM16 inhibits neuroblastoma cell proliferation through cell cycle regulation and dynamic nuclear localization. Cell Cycle 12, 889e898. Caers, J., Hose, D., Kuipers, I., Bos, T.J., Van Valckenborgh, E., Menu, E., De Bruyne, E., Goldschmidt, H., Van Camp, B., Klein, B., Vanderkerken, K., 2010a. Thymosin 4 has tumor suppressive effects and its decreased expression results in poor prognosis and decreased survival in multiple myeloma. Haematologica 95, 163e167. Caers, J., Otjacques, E., Hose, D., Klein, B., Vanderkerken, K., 2010b. Thymosin beta4 in multiple myeloma: friend or foe. Ann. N. Y. Acad. Sci. 1194, 125e129. Calao, M., Sekyere, E.O., 2013. Direct effects of Bmi1 on p53 protein stability inactivates oncoprotein stress responses in embryonal cancer precursor cells at tumor initiation. Oncogene 32, 3616e3626. Carlier, M.F., Didry, D., Erk, I., Lepault, J., Van Troys, M.L., Vandekerckhove, J., Perelroizen, I., Yin, H., Doi, Y., Pantaloni, D., 1996. Tbeta 4 is not a simple G-actin sequestering protein and interacts with F-actin at high concentration. J. Biol. Chem. 271, 9231e9239. Cha, H.J., Jeong, M.J., Kleinman, H.K., 2003. Role of thymosin beta4 in tumor metastasis and angiogenesis. J. Natl. Cancer Inst. 95, 1674e1680. Chou, T.C., 2006. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacol. Rev. 58, 621e681.

15

Chou, T.C., Talalay, P., 1984. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv. Enzyme Regul. 22, 27e55. Colombo, N., Formelli, F., Cantu, M.G., Parma, G., Gasco, M., Argusti, A., Santinelli, A., Montironi, R., Cavadini, E., Baglietto, L., Guerrieri-Gonzaga, A., Viale, G., Decensi, A., Colombo, N., Formelli, F., Cantu, M.G., Parma, G., Gasco, M., Argusti, A., Santinelli, A., Montironi, R., Cavadini, E., Baglietto, L., Guerrieri-Gonzaga, A., Viale, G., Decensi, A., 2006. A phase I-II preoperative biomarker trial of fenretinide in ascitic ovarian cancer. Cancer Epidemiol. Biomarkers Prev. 15, 1914e1919. De los Santos, M., Zambrano, A., Aranda, A., 2007. Combined effects of retinoic acid and histone deacetylase inhibitors on human neuroblastoma SH-SY5Y cells. Mol. Cancer Ther. 6, 1425e1432. Duvic, M., Talpur, R., Ni, X., Zhang, C., Hazarika, P., Kelly, C., Chiao, J.H., Reilly, J.F., Ricker, J.L., Richon, V.M., Frankel, S.R., 2007. Phase 2 trial of oral vorinostat (suberoylanilide hydroxamic acid, SAHA) for refractory cutaneous T-cell lymphoma (CTCL). Blood 109, 31e39. Epping, M.T., Meijer, L.A.T., Bos, J.L., Bernards, R., 2009. UNC45A confers resistance to histone deacetylase inhibitors and retinoic acid. Mol. Cancer Res. MCR 7, 1861e1870. Epping, M.T., Wang, L., Plumb, J.A., Lieb, M., Gronemeyer, H., Brown, R., Bernards, R., 2007. A functional genetic screen identifies retinoic acid signaling as a target of histone deacetylase inhibitors. Proc. Natl. Acad. Sci. USA 104, 17777e17782. Fang, H., Harned, T.M., Kalous, O., Maldonado, V., DeClerck, Y.A., Reynolds, C.P., 2011. Synergistic activity of fenretinide and the Bcl-2 family protein inhibitor ABT-737 against human neuroblastoma. Clin. Cancer Res. 17, 7093e7104. Formelli, F., Cavadini, E., Luksch, R., Garaventa, A., Villani, M.G., Appierto, V., Persiani, S., 2008. Pharmacokinetics of oral fenretinide in neuroblastoma patients: indications for optimal dose and dosing schedule also with respect to the active metabolite 4-oxo-fenretinide. Cancer Chemother. Pharmacol. 62, 655e665. Fouladi, M., Park, J.R., Stewart, C.F., Gilbertson, R.J., Schaiquevich, P., Sun, J., Reid, J.M., Ames, M.M., Speights, R., Ingle, A.M., Zwiebel, J., Blaney, S.M., Adamson, P.C., 2010. Pediatric phase I trial and pharmacokinetic study of vorinostat: a Children’s Oncology Group phase I consortium report. J. Clin. Oncol. 28, 3623e3629. Franken, N.A., Rodermond, H.M., Stap, J., Haveman, J., van Bree, C., 2006. Clonogenic assay of cells in vitro. Nat. 1, 2315e2319. Gautier, L., Cope, L., Bolstad, B.M., Irizarry, R.A., 2004. affyeanalysis of Affymetrix GeneChip data at the probe level. Bioinformatics 20, 307e315. Germain, P., Iyer, J., Zechel, C., Gronemeyer, H., Germain, P., Iyer, J., Zechel, C., Gronemeyer, H., 2002. Co-regulator recruitment and the mechanism of retinoic acid receptor synergy. Nature 415, 187e192. Glass, C.K., Rosenfeld, M.G., 2000. The coregulator exchange in transcriptional functions of nuclear receptors. Genes Dev. 14, 121e141. Hainsworth, J.D., Infante, J.R., Spigel, D.R., Arrowsmith, E.R., Boccia, R.V., Burris, H.A., 2011. A phase II trial of panobinostat, a histone deacetylase inhibitor, in the treatment of patients with refractory metastatic renal cell carcinoma. Cancer Invest 29, 451e455. Hill, D.S., Martin, S., Armstrong, J.L., Flockhart, R., Tonison, J.J., Simpson, D.G., Birch-Machin, M.A., Redfern, C.P., Lovat, P.E., Hill, D.S., Martin, S., Armstrong, J.L., Flockhart, R., Tonison, J.J., Simpson, D.G., Birch-Machin, M.A., Redfern, C.P.F., Lovat, P.E., 2009. Combining the endoplasmic reticulum stress-inducing

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

16

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

agents bortezomib and fenretinide as a novel therapeutic strategy for metastatic melanoma. Clin. Cancer Res. 15, 1192e1198. Huang da, W., Sherman, B.T., Lempicki, R.A., 2009. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat 4, 44e57. Iraci, N., Diolaiti, D., Papa, A., Porro, A., Valli, E., Gherardi, S., Herold, S., Eilers, M., Bernardoni, R., Della Valle, G., Perini, G., 2011. A SP1/MIZ1/MYCN repression complex recruits HDAC1 at the TRKA and p75NTR promoters and affects neuroblastoma malignancy by inhibiting the cell response to NGF. Cancer Res. 71, 404e412. Johnstone, R.W., 2002. Histone-deacetylase inhibitors: novel drugs for the treatment of cancer. Nat. Rev. Drug Discov. 1, 287e299. Kato, Y., Salumbides, B.C., Wang, X.-F., Qian, D.Z., Williams, S., Wei, Y., Sanni, T.B., Atadja, P., Pili, R., 2007. Antitumor effect of the histone deacetylase inhibitor LAQ824 in combination with 13-cis-retinoic acid in human malignant melanoma. Mol. Cancer Ther. 6, 70e81. Kocak, H., Ackermann, S., Hero, B., Kahlert, Y., Oberthuer, A., Juraeva, D., Roels, F., Theissen, J., Westermann, F., Deubzer, H., Ehemann, V., Brors, B., Odenthal, M., Berthold, F., Fischer, M., 2013. Hox-C9 activates the intrinsic pathway of apoptosis and is associated with spontaneous regression in neuroblastoma. Cell Death Dis. 4, e586. Lees, J.G., Bach, C.T., Bradbury, P., Paul, A., Gunning, P.W., O’Neill, G.M., 2011. The actin-associating protein Tm5NM1 blocks mesenchymal motility without transition to amoeboid motility. Oncogene 30, 1241e1251. Liu, T., Tee, A.E.L., Porro, A., Smith, S.A., Dwarte, T., Liu, P.Y., Iraci, N., Sekyere, E., Haber, M., Norris, M.D., Diolaiti, D., Della Valle, G., Perini, G., Marshall, G.M., 2007. Activation of tissue transglutaminase transcription by histone deacetylase inhibition as a therapeutic approach for Myc oncogenesis. Proc. Natl. Acad. Sci. USA 104, 18682e18687. Marshall, G.M., Bell, J.L., Koach, J., Tan, O., Kim, P., Malyukova, A., Thomas, W., Sekyere, E.O., Liu, T., Cunningham, A.M., Tobias, V., Norris, M.D., Haber, M., Kavallaris, M., Cheung, B.B., 2010. TRIM16 acts as a tumour suppressor by inhibitory effects on cytoplasmic vimentin and nuclear E2F1 in neuroblastoma cells. Oncogene 29, 6172e6183. Matthay, K.K., Reynolds, C.P., Seeger, R.C., Shimada, H., Adkins, E.S., Haas-Kogan, D., Gerbing, R.B., London, W.B., Villablanca, J.G., 2009. Long-term results for children with high-risk neuroblastoma treated on a randomized trial of myeloablative therapy followed by 13-cis-retinoic acid: a children’s oncology group study. J. Clin. Oncol. 27, 1007e1013. Maurer, B.J., Metelitsa, L.S., Seeger, R.C., Cabot, M.C., Reynolds, C.P., 1999. Increase of ceramide and induction of mixed apoptosis/necrosis by N-(4-hydroxyphenyl)retinamide in neuroblastoma cell lines. J. Natl. Cancer Inst. 91, 1138e1146. Nummela, P., Yin, M., Kielosto, M., Leaner, V., Birrer, M.J., Holtta, E., 2006. Thymosin beta4 is a determinant of the transformed phenotype and invasiveness of Sadenosylmethionine decarboxylase-transfected fibroblasts. Cancer Res. 66, 701e712. Ponzoni, M., Bocca, P., Chiesa, V., Decensi, A., Pistoia, V., Raffaghello, L., Rozzo, C., Montaldo, P.G., 1995. Differential effects of N-(4-hydroxyphenyl)retinamide and retinoic acid on neuroblastoma cells: apoptosis versus differentiation. Cancer Res. 55, 853e861. Popperl, H., Featherstone, M.S., 1993. Identification of a retinoic acid response element upstream of the murine Hox-4.2 gene. Mol. Cell Biol. 13, 257e265. Reddy, P., Maeda, Y., Hotary, K., Liu, C., Reznikov, L.L., Dinarello, C.A., Ferrara, J.L., Reddy, P., Maeda, Y., Hotary, K.,

Liu, C., Reznikov, L.L., Dinarello, C.A., Ferrara, J.L.M., 2004. Histone deacetylase inhibitor suberoylanilide hydroxamic acid reduces acute graft-versus-host disease and preserves graft-versus-leukemia effect. Proc. Natl. Acad. Sci. USA 101, 3921e3926. Reynolds, C.P., Reynolds, C.P., 2004. Detection and treatment of minimal residual disease in high-risk neuroblastoma. Pediatr. Transpl. 8 (Suppl 5), 56e66. Reynolds, C.P., Wang, Y., Melton, L.J., Einhorn, P.A., Slamon, D.J., Maurer, B.J., 2000. Retinoic-acid-resistant neuroblastoma cell lines show altered MYC regulation and high sensitivity to fenretinide. Med. Pediatr. Oncol. 35, 597e602. Schroeder, S.E., Reddy, M.B., Schalinske, K.L., 2007. Retinoic acid modulates hepatic iron homeostasis in rats by attenuating the RNA-binding activity of iron regulatory proteins. J. Nutr. 137, 2686e2690. Shibina, A., Seidel, D., Somanchi, S.S., Lee, D.A., Stermann, A., Maurer, B.J., Lode, H.N., Reynolds, C.P., Huebener, N., 2013. Fenretinide sensitizes multidrugresistant human neuroblastoma cells to antibodyindependent and ch14.18-mediated NK cell cytotoxicity. J. Mol. Med. 91, 459e472. Shum, M.S., Pasquier, E., Po’uha, S.T., O’Neill, G.M., Chaponnier, C., Gunning, P.W., Kavallaris, M., 2011. -Actin regulates cell migration and modulates the ROCK signaling pathway. Faseb J. 25, 4423e4433. Sogno, I., Vene, R., Sapienza, C., Ferrari, N., Tosetti, F., Albini, A., Sogno, I., Vene, R., Sapienza, C., Ferrari, N., Tosetti, F., Albini, A., 2009. Anti-angiogenic properties of chemopreventive drugs: fenretinide as a prototype. Recent Results Cancer Res. 181, 71e76. Sosne, G., Hafeez, S., Greenberry 2nd, A.L., KurpakusWheater, M., 2002. Thymosin beta4 promotes human conjunctival epithelial cell migration. Curr. Eye Res. 24, 268e273. Spiller, S.E., Ditzler, S.H., Pullar, B.J., Olson, J.M., 2008. Response of preclinical medulloblastoma models to combination therapy with 13-cis retinoic acid and suberoylanilide hydroxamic acid (SAHA). J. Neuro-Oncol. 87, 133e141. Steele, N., Finn, P., Brown, R., Plumb, J.A., 2009. Combined inhibition of DNA methylation and histone acetylation enhances gene re-expression and drug sensitivity in vivo. Br. J. Cancer 100, 758e763. Tilghman, R.W., Slack-Davis, J.K., Sergina, N., Martin, K.H., Iwanicki, M., Hershey, E.D., Beggs, H.E., Reichardt, L.F., Parsons, J.T., 2005. Focal adhesion kinase is required for the spatial organization of the leading edge in migrating cells. J. Cell Sci. 118, 2613e2623. Villablanca, J.G., Krailo, M.D., Ames, M.M., Reid, J.M., Reaman, G.H., Reynolds, C.P., 2006. Phase I trial of oral fenretinide in children with high-risk solid tumors: a report from the Children’s Oncology Group (CCG 09709) [Erratum appears in J Clin Oncol. 2006 Sep 1;24(25):4223 Note: Reynolds, Patrick C [corrected to Reynolds, C Patrick]] J. Clin. Oncol. 24, 3423e3430. Villablanca, J.G., London, W.B., Naranjo, A., McGrady, P., Ames, M.M., Reid, J.M., McGovern, R.M., Buhrow, S.A., Jackson, H., Stranzinger, E., Kitchen, B.J., Sondel, P.M., Parisi, M.T., Shulkin, B., Yanik, G.A., Cohn, S.L., Reynolds, C.P., Villablanca, J.G., London, W.B., Naranjo, A., McGrady, P., Ames, M.M., Reid, J.M., McGovern, R.M., Buhrow, S.A., Jackson, H., Stranzinger, E., Kitchen, B.J., Sondel, P.M., Parisi, M.T., Shulkin, B., Yanik, G.A., Cohn, S.L., Reynolds, C.P., 2011. Phase II study of oral capsular 4hydroxyphenylretinamide (4-HPR/fenretinide) in pediatric patients with refractory or recurrent neuroblastoma: a report from the Children’s Oncology Group. Clin. Cancer Res. 17, 6858e6866.

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005

M O L E C U L A R O N C O L O G Y X X X ( 2 0 1 5 ) 1 e1 7

Wang, K., Wang, P., Shi, J., Zhu, X., He, M., Jia, X., Yang, X., Qiu, F., Jin, W., Qian, M., Fang, H., Mi, J., Yang, X., Xiao, H., Minden, M., Du, Y., Chen, Z., Zhang, J., 2010. PML/RARalpha targets promoter regions containing PU.1 consensus and RARE half sites in acute promyelocytic leukemia. Cancer Cell 17, 186e197. Wang, W.-S., Chen, P.-M., Hsiao, H.-L., Wang, H.-S., Liang, W.-Y., Su, Y., 2004. Overexpression of the thymosin beta-4 gene is associated with increased invasion of SW480 colon carcinoma cells and the distant metastasis of human colorectal carcinoma. Oncogene 23, 6666e6671. Wozniak, M.A., Modzelewska, K., Kwong, L., Keely, P.J., 2004. Focal adhesion regulation of cell behavior. Biochim. Biophys. Acta 1692, 103e119.

17

Xiao, Y., Chen, Y., Wen, J., Yan, W., Zhou, K., Cai, W., 2012. Thymosin beta4: a potential molecular target for tumor therapy. Crit. Rev. Eukaryot. Gene Expr. 22, 109e116. Yu, A.L., Gilman, A.L., Ozkaynak, M.F., London, W.B., Kreissman, S.G., Chen, H.X., Smith, M., Anderson, B., Villablanca, J.G., Matthay, K.K., Shimada, H., Grupp, S.A., Seeger, R., Reynolds, C.P., Buxton, A., Reisfeld, R.A., Gillies, S.D., Cohn, S.L., Maris, J.M., Sondel, P.M., Children’s Oncology, G, 2010. Anti-GD2 antibody with GM-CSF, interleukin-2, and isotretinoin for neuroblastoma. N. Engl. J. Med. 363, 1324e1334. Zhang, B., Kirov, S., Snoddy, J., 2005. WebGestalt: an integrated system for exploring gene sets in various biological contexts. Nucleic Acids Res. 33, W741eW748.

Please cite this article in press as: Cheung, B.B., et al., Thymosin-b4 is a determinant of drug sensitivity for Fenretinide and Vorinostat combination therapy in neuroblastoma, Molecular Oncology (2015), http://dx.doi.org/10.1016/j.molonc.2015.04.005