Identification and functional characterization of glioma ... - Springer Link

6 downloads 0 Views 579KB Size Report
Feb 24, 2011 - reaction screening, MAGE-A3 and SSX4 were found to be expressed in a tumor-specific manner. SSX4 gene promoter activity was high in ...
J Neurooncol (2011) 104:497–507 DOI 10.1007/s11060-010-0522-0

LABORATORY INVESTIGATION - HUMAN/ANIMAL TISSUE

Identification and functional characterization of glioma-specific promoters and their application in suicide gene therapy Toshio Yawata • Yusuke Maeda • Makiko Okiku Eri Ishida • Kazuhiro Ikenaka • Keiji Shimizu



Received: 31 May 2010 / Accepted: 22 December 2010 / Published online: 24 February 2011 Ó Springer Science+Business Media, LLC. 2011

Abstract Suicide gene therapy has been shown to be effective in inducing tumor regression. In this study, a human brain tumor-specific promoter was identified and used to develop transcriptionally targeted gene therapy. We searched for genes with brain tumor-specific expression. By in silico and reverse-transcription polymerase chain reaction screening, MAGE-A3 and SSX4 were found to be expressed in a tumor-specific manner. SSX4 gene promoter activity was high in human brain tumor cells but not in normal human astrocyte cells, whereas the MAGE-A3 promoter showed activity in both tumor and normal cells. A retrovirus vector carrying a suicide gene, the herpes simplex virus thymidine kinase gene controlled by the SSX4 promoter, was constructed to evaluate the efficacy of the promoter in tumor-specific gene therapy. Glioma and human telomerase catalytic subunit-immortalized fibroblast BJ-5ta cell lines transduced with retrovirus vectors were assayed for killing activity by ganciclovir. Glioma cell lines were effectively killed by ganciclovir in a concentration-dependent manner, whereas BJ-5ta cells were not. By contrast, MAGE-A3 promoter failed to induce cytotoxicity in a brain tumor-specific manner. In addition,

Electronic supplementary material The online version of this article (doi:10.1007/s11060-010-0522-0) contains supplementary material, which is available to authorized users. T. Yawata  Y. Maeda  M. Okiku  E. Ishida  K. Shimizu (&) Department of Neurosurgery, Kochi Medical School, Nankoku, Kochi, Japan e-mail: [email protected] K. Ikenaka Division of Neurobiology and Bioinformatics, National Institutes of Natural Sciences National Institute for Physiological Sciences, Okazaki, Japan

mouse glioma RSV-M cells transduced with retrovirus vector also showed suppressed tumor formation activity in syngeneic mice in response to ganciclovir administration. Therefore, the SSX4 promoter is a candidate for brain tumor-specific gene therapy and supports the efficacy and safety of suicide gene therapy for malignant brain tumors. Keywords Brain tumor  Suicide gene therapy  Tumor-specific promoter  Cancer-testis antigen  SSX4

Introduction Glioblastoma multiforme (GBM) is the most common primary brain tumor and has infiltrating edges that are difficult to completely remove with surgery. GBM also shows low sensitivity to modern therapies, despite developments in chemo- and radiotherapy, and the prognosis has not improved greatly over the years [1–3]. Gene therapy may provide an alternative strategy for treatment of malignant tumors. Expression of suicide genes delivered to GBM by various vectors has been shown to be very effective in killing tumor cells [4–7]. However, the toxicity of suicide genes in normal cells and tissues has not been well characterized. Theoretically, the restriction of suicide gene expression to tumor cells would increase the safety of gene therapy [8]. The myelin basic protein (MBP) gene has been used as a tissue-specific promoter for controlled expression of suicide genes in glioma models [9]. Human telomerase catalytic subunit (hTERT) promoters have also demonstrated potential as tumor-specific promoters for restricting suicide gene expression [10, 11]. Given that telomerase activity is observed in more than 80% of all malignant tumors, gene therapy using the telomerase catalytic subunit promoter has broad potential.

123

498

However, 25% of GBM tissues show no telomerase activity, indicating an alternative mechanism for telomere lengthening [12]. In addition, the efficacy of hTERT promoter gene therapy limits further application, because its activity is not high [13]. Therefore, identification of novel promoters with tumor-specific and strong activity remains a goal for the development of tumor-specific gene therapy. Cancer-testis antigen (CTA) genes are expressed in restricted cell types (testis in normal tissue) and various tumors and are well-known candidate molecules for vaccine therapy of malignant tumors owing to their tumor specificity [14]. The CTA genes are epigenetically repressed in normal tissue, with the exception of the testis, and derepressed during malignant transformation. Many reports suggest that methylation of the CTA gene promoter region plays a crucial role in its regulated expression. However, the number of studies characterizing the CTA gene promoter according to differences of promoter activity in normal and malignant cells is very limited. The mechanism(s) underlying testisand tumor-specific expression of this gene family is not well understood, and the CTA gene promoter has not been evaluated for tumor-specific expression of suicide genes. In this study, we attempted to identify tumor-specific promoters driving CTA gene expression for the development of tumor-specific gene therapy. By in silico and reverse-transcription polymerase chain reaction (RT-PCR) screening, melanoma-associated antigen A3 (MAGE-A3) and synovial sarcoma, X breakpoint 4 (SSX4) genes were found to be expressed in all brain tumor cell lines tested but not in normal human astrocytes. The activity of the MAGE-A3 promoter did not reflect the expression pattern, but the SSX4 promoter showed tumor-specific activity. Therefore, we evaluated the efficacy of the SSX4 promoter in tumor-specific gene therapy against brain tumors.

Materials and methods Cell culture Human glioblastoma cell lines U87MG, SNB19, T98G, U373, and UW18, the medulloblastoma cell line UW228, and the telomerase-immortalized human fibroblast cell line BJ-5ta were purchased from the American Type Culture Collection (Manassas, VA). Other primary cell lines were established from patients with glioblastoma (ONS-12, ONS23), astrocytoma (ONS-65, grade II), anaplastic astrocytoma (ONS-75, grade III), and medulloblastoma (ONS-76). Rous sarcoma virus-induced malignant mouse glioma (RSV-M) cells were provided by Kumanishi et al. [15]. All tumor cell lines were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal calf serum (FCS) and 1% penicillin/streptomycin. BJ-5ta cells

123

J Neurooncol (2011) 104:497–507

were cultured in DMEM:199 (4:1) containing 10% FCS and 1% penicillin/streptomycin. Normal human astrocyte (NHA) cells were purchased and maintained with an Astrocyte Medium Bullet kit (CAMBREX, Baltimore, MD). Written informed consent was obtained from patients prior to tumor specimen collection. The Ethics Committee of Kochi Medical School approved this study. Reverse-transcription polymerase chain reaction analysis A total of 1 lg extracted RNA was reverse-transcribed with a SuperScript II complementary DNA (cDNA) synthesis kit (Invitrogen, San Diego, CA) according to the manufacturer’s instructions. cDNA was amplified with AmpliTaq Gold (Applied Biosystems, Foster City, CA), and PCR products were electrophoresed on 2% agarose gels and visualized by ethidium bromide staining. For quantification of MAGE-A3 and SSX4 mRNAs, real-time PCR was performed with initial denaturation at 94°C for 10 min followed by 45 cycles of 20 s at 94°C, 20 s at 57°C, 20 s at 72°C, and 84°C at 20 s using a LightCycler system (Roche, Indianapolis, IN) and a Quantitect SYBR Green PCR kit (Qiagen, Santa Clarita, CA). Primer pairs and cycling conditions for each gene are listed in Electronic Supplemental Table 1. Plasmid construction MAGE-A3, MAGE-D4, and SSX4 promoters were amplified with Phusion DNA polymerase (Finnzymes Oy, Espoo, Finland) from normal lymphocyte genomic DNA by PCR with specific primers (Electronic Supplemental Table 1) and cloned into pGL3 Basic (Promega, Madison, WI). For construction of unidirectional deleted fragments of SSX4 promoter, amplified fragments from 2,901-bp SSX4 promoter by PCR using SSX4-1931, -1162, -885, -485, -255, -107, -60, -20, and SSX4pR2 primers were digested with NheI and BglII and ligated with pGL3 Basic digested with the same enzymes. For construction of retroviral vectors, the cytomegalovirus (CMV) promoter in pRetroQDsRedN1 (Clontech, Kyoto, Japan) was deleted by PCR with Phusion DNA polymerase and primers pRetroDelF and pRetroDelR. The PCR product was digested with NheI and self-ligated (pRDCMVDs). The suicide gene herpes simplex virus thymidine kinase (HSVtk) was amplified from MBP/pIP250? [9] using HSVtkF and HSVtkR primers. The amplified fragment was digested with BamHI and NotI and ligated with pRDCMVDs digested with the same enzymes (pRDCMVHTK). Fragments of the SSX4 promoter (-255 to ?33 and -485 to ?33) were amplified from pGL3-SSX4 constructs by PCR using SSX4-485 or SSX4-255 and SSX4pro2R primers, digested with SalI and NheI, and

J Neurooncol (2011) 104:497–507

introduced into the same restriction enzyme sites of pRDCMVHTK (pRpSSX255HTK and pRpSSX485HTK). For construction of the retroviral vector carrying HSVtk under the control of the CMV promoter, pRetroQDsRed-N1 was digested with SalI and NotI and ligated with the HSVtk fragment digested with the same enzymes (pRpCMVHTK). All constructs were confirmed by automated sequencing. Luciferase assay Cells (5 9 104) suspended in 1 ml antibiotic-free medium were seeded into each well of a 12-well plate and cotransfected the next day with 0.5 lg reporter vector and 0.05 lg Renilla reniformis luciferase vector using FuGENE6 transfection reagent (Roche). Two days after transfection, cells were harvested and luciferase activity was measured with a Dual Luciferase kit (Promega). The level of firefly luciferase activity was normalized to that of R. reniformis luciferase activity for each transfection. For comparison among cell lines with different transfection efficiencies, the pGL3-Control plasmid (Promega), which contains the firefly luciferase gene under the transcriptional control of the SV40 enhancer/promoter, was transfected into each cell line and used for normalization. RNA-ligase-mediated rapid amplification of 50 cDNA ends analysis of capped RNA Total RNA was isolated from glioma cells using an RNA purification kit (Qiagen). Human testis total RNA was purchased from Clontech. RNA-ligase-mediated rapid amplification of 50 cDNA ends (50 -RLM-RACE) was performed using GeneRacer kit (Invitrogen) according to the manufacturer’s instructions. A total of 3 mg total RNA was used for each sample. In brief, total RNA was treated with calf intestinal phosphatase to remove 50 phosphates from truncated messenger RNA (mRNA). Dephosphorylated RNA was treated with tobacco acid pyrophosphatase to remove the 50 cap from full-length mRNA, leaving a 50 phosphate. The GeneRacer RNA oligomer was then ligated to the 50 end of the mRNA with T7 RNA ligase. Firststrand cDNA was produced with avian myeloblastosis virus (AMV) reverse-transcriptase and gene-specific primers. The regions corresponding to the legitimate 50 ends of the capped RNA species were PCR-amplified from cDNA templates with the GeneRacer (GR) 50 primer and MAGEA3GSP1 or SSX4GSP1 primers. Nested PCR was performed with the GR 50 nested primer and the MAGEA3GSP2 or SSX4GSP2 primers. For the amplification of SSX4 fragments, additional seminested PCR was performed using GR 50 nested primer and SSX4GSP3 primers.

499

The PCR products of the 50 -RLM-RACE RT-PCR reactions were cloned into pGEM-T vector (Promega). Clones were sequenced with an automated sequencer and M13-21 or M13rev primers. Production of retroviral vector and gene transfer We previously established the amphotropic retrovirus packaging cell line PAM51 yielding a high-titer retrovirus vector [16]. PAMP51 cells (2 9 105) were seeded in gelatin-coated, six-well plates and incubated at 37°C and 5% CO2. Twenty-four hours later, PAMP51 cells were transfected with 2 lg retroviral vector DNA containing the HSVtk gene under the control of no promoter or the SSX4 or CMV promoter with FuGENE6 and then incubated at 37°C. The medium was replaced the next day, and transfected cells were incubated at 32°C for 4 days to produce retroviruses. Viral supernatant was collected and filtered with a 0.45-lm pore membrane. After the addition of polybren (final concentration 8 lg/ml) to culture medium containing the viral supernatant, tumor or BJ-5ta cells (2–4 9 105/35-mm dish) were transduced with a multiplicity of infection (MOI) of 1–3. After 4 h of incubation at 37°C, 2 ml fresh medium was added to each dish, and the cells were allowed to incubate for 2 days. The cells were then plated in 10-cm dishes and cultured in complete medium containing 500 lg/ml G418 or 1 lg/ ml puromycin for 1–2 weeks at 37°C to generate clones of transduced cells. More than 50 independent, drugresistant colonies were identified and expanded as a mixed culture. MTT assay 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays were carried out according to the manufacturer’s instructions (Millipore, Bedford, MA). Cells were suspended in medium at concentration of 2 9 104 cells/ml and plated (100 ll/well) in 96-well plates. After addition of ganciclovir (GCV), the cells were incubated in a humidified CO2 incubator at 37°C for 5 days, and cytotoxicity was determined by MTT assay. Absorbance was recorded on a microplate reader at wavelength of 570 nm and reference wavelength of 630 nm. Data are presented as means of five separate experiments. In vivo subcutaneous xenografts Five female C3H/HeN mice (6–8 weeks old) per group were implanted subcutaneously with 5 9 106 cells. Tumor volume (mm3) was calculated by the following formula: volume = L(W2)/2.

123

500

Immunohistochemistry Tissue specimens were postfixed in 4% paraformaldehyde overnight and stored at 4°C in 30% sucrose prior to cryosectioning. Five-micron sections were cut and examined for HSVtk expression using immunohistochemistry. HSVtk was stained with rabbit polyclonal anti-HSVtk antibody (dilution, 1:200; Santa Cruz Biotechnology Inc., Santa Cruz, CA). Bound antibodies were detected using VECTASTAIN Elite ABC reagent (VECTOR, Burlingame, CA).

Results Identification of tumor-specific genes expressed in brain tumors at high frequency To identify promoters showing high activity in brain tumor cells, we planned to select genes expressed in brain tumor cells but not in normal cells. The CTA genes are well known to be expressed in tumor cells but not in normal cells, with the exception of the testis [17]. The MAGE family includes many CTA genes and is a candidate for molecular targeted therapy of malignant tumors. Therefore, we first surveyed the expression frequency of CTA and MAGE family genes in brain tumors by in silico screening of the NCBI-CGAP EST and SAGE database (http://cgap.nci.nih.gov/). Thirty-five of 104 CTA and MAGE family genes were selected as candidate genes frequently expressed in tumor tissues compared with normal brain. The expression of these 35 genes was evaluated in the testis, NHA cells, six glioma cell lines, and one medulloblastoma cell line by RT-PCR (Fig. 1a). As expected, all CTA genes tested showed expression in the testis. Contrary to our expectation, 18 of the 35 CTA genes were also expressed in NHA cells. Among genes not expressed in NHA cells, SSX4 and MAGE-A3 genes were expressed in all tumor cells. We then added four glioma cell lines and one medulloblastoma cell line to the analysis and performed semiquantitative RT-PCR and quantitative RT-PCR (qRT-PCR) for the expression of SSX4 and MAGE-A3 genes (Fig. 1b). Greater than 100-fold differences in expression level were observed among tumor cell lines. However, the expression levels in NHA cells were below the sensitivity of this analysis, suggesting that the expression of SSX4 and MAGE-A3 is strictly regulated in the testis and tumor cells. Thus, SSX4 and MAGE-A3 were not only typical CTA genes, but also frequently expressed genes in human brain tumors.

123

J Neurooncol (2011) 104:497–507

Determination of transcription initiation sites and promoter activity of the SSX4 and MAGE-A3 genes Transcription start sites of the MAGE-A3 and SSX4 genes were determined by 50 -RLM-RACE analysis, which has proved to be a very sensitive and accurate method for obtaining full-length 50 cDNA ends and ensuring that truncated messages are eliminated from amplification reactions [18, 19]. Nucleotide sequencing of the 50 ends of the amplified products for the testis and SNB19 cells confirmed the match to reference sequences NM_005362 and NM_005636 in the NCBI nucleotide database (http:// www.ncbi.nlm.nih.gov/) for initiation of MAGE-A3 and SSX4 transcripts, respectively. Discrepancies between exogenous promoter activity and expression level are often observed in epigenetically regulated genes. To examine whether SSX4 and MAGE-A3 promoter activities reflected their expression status, both promoter regions were cloned into the luciferase reporter vector pGL3 (pGL3-SSX4 and pGL3-MAGE-A3) (Fig. 2). Promoter activities were measured in NHA, U87MG, T98G, ONS-12, and ONS-23 cells. pGL3-SSX4 was significantly more active (*17.3-fold) in glioma cell lines than in NHA cells. However, high activity of pGL3MAGE-A3 was observed in all cells, including NHA cells, indicating nonspecific activity of the MAGE-A3 promoter. This result suggests that the tumor-specific activity of the SSX4 promoter is regulated by trans-acting factor(s) and that it may be suitable for transgene experiments. Determination of the minimally active region of the SSX4 promoter Because the size of the DNA fragment inserted into retrovirus vectors is limited, and large constructs often fail to produce viral solutions with high titer, the minimal promoter region should be used for construction. To determine the minimal promoter region of the SSX4 gene, we created a series of luciferase constructs containing unidirectionally deleted fragments from pGL3-SSX4 and used these in luciferase assays (Fig. 3). Compared with the activity of the pGL3-SSX4 construct, the promoter activity of a site 255 bp upstream of the transcriptional initiation site showed maximal activity. Further deletion of the promoter notably decreased activity. Thus, cis-element(s) responsible for maximal promoter activity recruit within a 149-bp region from position -255 to -107. The lack of enhancement of promoter activity in NHA and BJ-5ta cells by any SSX4 construct could reflect the absence of negative regulatory element(s). Therefore, we identified the 255 bp

501

Fig. 1 Screening of CTA genes frequently expressed in brain tumor cell lines. a Summary of the expression of 35 CTA genes in brain tumor cell lines by RT-PCR analysis. Closed and open rectangles indicate expression and no expression, respectively. Glyceraldehyde-3phosphate dehydrogenase (G3PDH) was used as a control in this analysis. b Representative expression of SSX4 and MAGE-A3 by RTPCR analysis. c Ubiquitous expression of SSX4 and MAGE-A3 in brain tumor cell lines. Relative SSX4 and MAGE-A3 mRNA levels were quantified by real-time PCR normalized with beta-actin (ACTB) as a control. The error bars represent the standard deviation

H2 O Te sti s N H A O N SO 12 N SO 23 N S O -65 N SO 75 N SU 76 87 M T9 G 8G Ex pr es s /to ed i ta n t l um or

J Neurooncol (2011) 104:497–507

A

Brain tumor cell line

XAGE-3a MAGEB1 BORIS MAGEB4 SSX3 SYCP1 TPX1 MAGEB2 SPANXC TRAG-3 MMA-1a LAGE-1 NY-ESO-1 BAGE SSX1 MAGEA3 SSX4 HAGE MAGEL2 MAGED3 NY-CO-45 TEX15 CSAGE NDN MAGEH1 IL13RA1 MAGED4 MAGEF1 MAGEG1 NY-SAR-35 SPA17 SPANXA BAGE4 MAGED2 MAGED1 G3PDH

-

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + + + + + + + +

0/7 0/7 0/7 0/7 0/7 0/7 0/7 3/7 3/7 4/7 4/7 4/7 4/7 4/7 6/7 7/7 7/7 0/8 3/7 5/7 5/7 5/7 6/7 6/7 7/7 7/7 7/7 7/7 7/7 7/7 7/7 7/7 7/7 7/7 7/7 7/7

H

2

O Te st is N H A U W 18 SN B1 9 U 87 M G T9 8G U 37 3 O N S1 O 2 N S2 O 3 N S6 O 5 N S7 K 5 N -1 U W 22 O 8 N S76

B

SSX4 MAGEA3 G3PDH

upstream of the transcriptional initiation site as the minimal promoter region of SSX4. Tumor-specific killing activity of the pSSX4-HSVtk construct Preservation of normal tissues is important in suicide gene therapy. Hence, the tumor-specific activity of the SSX4 promoter is an attractive feature for the control of the

expression of suicide genes. Self-inactivating (SIN) vectors are useful for avoiding promoter interference and exercising precise control of gene expression by the inserted promoter [20]. We constructed an SIN retrovirus vector harboring the suicide gene HSVtk under the control of the SSX4 or MAGE-A3 or MAGE-D4 promoter (Fig. 4a) and transduced glioma and telomerase-immortalized human fibroblast BJ-5ta cells with this vector. Cytotoxicity of transduced and nontransduced cells was assayed by the

123

502

C

Relative ratio ( SSX4/ACTB)

Fig. 1 continued

J Neurooncol (2011) 104:497–507 0.01

0.001

0.0001

0.00001

0.000001

S7

22

6

8

-1

O

U

N

W

K

S7 N

N

5

5

3

S6 N

O

O

O

O

N

S2

2

3

S1

37

N

U

9

M G T9 8G

87

B1

U

SN

18

A

W

H

U

N

Te st is

0.0000001

Relative ratio ( MAGE-A3/ACTB)

10

1

0.01

0.001

123

NS

-7

6

8 22

O

N-

1 UW

5 -7

-6

NS O

NS O

K

5

3 -2

2 -1

NS O

NS

U3

73 O

G

8G T9

7M

B1

9

U8

W U

addition of increasing concentrations of GCV (Fig. 4b). Transduced and nontransduced cells with promoter-deleted constructed were not affected by GCV addition. All transduced cells with the CMV promoter-driven HSVtk expression vector were effectively killed with increasing concentrations of GCV. Transduced glioma cells with the SSX4 promoter-driven HSVtk expression vector were also effectively killed. However, transduced BJ-5ta cells with the SSX4 promoter-driven HSVtk expression vector were not affected. In addition, retrovirus vector harboring HSVtk under the control of MAGEA3 or MAGE-D4 promoter showed nonspecific or limited cytotoxicity in these cells. This result demonstrates that the SSX4 promoter-driven HSVtk construct induces cytotoxicity in a tumor-specific manner and may be effective for preventing damage of normal cells.

SN

18

A H N

Te s

tis

0.0001

Suppression of subcutaneously transplanted mouse glioma cells Because RSV-M consistently produces tumor masses in syngeneic C3H/HeN mice, we used this mouse cell line for an in vivo sensitivity assay to GCV (Fig. 4c). Results of in vivo GCV administration showed that tumor cells transduced with SSX4 promoter constructs completely stopped proliferation, similar to cells transduced with CMV promoter constructs, whereas wild-type cells and cells transduced with promoter-deleted constructs continued to proliferate. In mice without GCV administration, tumor cells transduced with SSX4 promoter constructs also formed tumor masses. In addition, expression of HSVtk was observed in tumor masses formed by the transplantation of transduced cells with CMV or SSX4 promoter construct (Fig. 4d). Therefore, the

Relative ratio (Firefly luc./Renilla luc.)

J Neurooncol (2011) 104:497–507

503

14 NHA U87MG T98G SNB19 ONS-12 ONS-23

12 10 8 6 4 2 0 SSX4

MAGE-A3

Fig. 2 Promoter activity of SSX4 and MAGE-A3 genes in NHA and glioma cell lines. A fragment of the SSX4 and MAGE-A3 genes extending from ?32 to -2,832 and ?52 to -2,980, respectively, was cloned into a luciferase reporter vector (pGL3 basic). The transcription-directed SSX4 and MAGE-A3 promoters were then assessed by measuring the amount of luciferase (luc.) activity in the transfectants. Means of at least three independent experiments are shown for each cell; bars indicate standard deviation

suppressive effects of SSX4 promoter constructs were dependent on GCV administration and HSVtk expression. This result demonstrates that SSX4-driven HSVtk expression can be used to suppress tumor formation.

Discussion In the present study, we found that the SSX4 and MAGEA3 genes are frequently expressed in brain tumor cell lines.

Fig. 3 Luciferase assay to identify the minimal promoter region of the SSX4 gene. A series of SSX4-luciferase constructs was generated by 50 -end unidirectional truncation. Each construct is identified regarding the sequence number at the 50 end relative to the transcriptional start site. Each bar represents the mean of triplicate relative luciferase activities in NHA, telomeraseimmortalized fibroblast (BJ-5ta), and glioma cells

SSX4 belongs to the family of highly homologous synovial sarcoma X (SSX) breakpoint proteins and is reported to be involved in the t(X;18) translocation characteristic of all synovial sarcomas [21]. SSX1 and SSX2 contain the KRAB motif and function as a transcription factor [22, 23]. SSX4 also contains the KRAB motif and is predicted to function as a transcriptional repressor on the basis of sequence homology (http://www-bimas.cit.nih.gov/cards//). However, the function of SSX4 has not been investigated. MAGE-A3 is a member of the MAGE gene family, which encodes proteins with 50–80% sequence identity to each other. This gene downregulates p53 expression via histone acetylation and methylation modifications. The role of MAGE-A3 in brain tumors has also not yet been investigated. Thus, both CTA genes expressed in brain tumors may be involved in transcriptional regulation and tumor malignancy. It has also been reported that SSX genes are involved in the migration of mesenchymal stem cells [24]. Highfrequency SSX4 expression may affect the invasiveness of brain tumor cells. It has been suggested that the expression of many CTA genes is regulated by epigenetic changes, that is, DNA methylation and histone modification of their promoter regions [25]. Methylated DNA and hypoacetylated histone of CTA gene promoters are observed in normal cells not expressing these genes. In contrast to this observation, these epigenetic modifications are reversed in the testis and tumor cells. In the present study, the nonspecific promoter activity of MAGE-A3 was revealed by promoter assays using transient transfection. We previously

+32

- 2832 - 1931

+32 - 1162

+32

- 885

+32

- 485 +32 - 255 +32 - 107 +32 - 60 +32

Luc Luc Luc Luc Luc Luc Luc NHA BJ-5ta T98G SNB19 U87MG

Luc

- 22 +32 Luc 0

0.1 0.2 0.3 Relative promoter activity

0.4

123

504

J Neurooncol (2011) 104:497–507

A Puro r

promoterless

LTR

HSVtk

pCMV

LTR

pCMV

HSVtk

Puro r

LTR

pSSX4-255 or 485

LTR

pSSX4

HSVtk

Puro r

LTR

pMAGE-A3 or pMAGE-D4

LTR

pMAGE

HSVtk

Puro r

LTR

LTR

B non-transduced

Survival rate

1.2

promoterless

1.2

1.0

1.0

1

0.8

0.8

0.8

0.6

0.6

0.6

0.4

0.4

0.4

0.2

0.2

0.2

0

0

0

2

1.2

4 6 8 10 12 0 pSSX4-255 1.2

2

pCMV

1.2

RSV-M U87MG BJ-5ta

0 4 6 8 10 12 0 pSSX4-485 1.2

2 4 6 8 10 12 pMAGE-A3 1.2

1.0

1.0

1.0

1

0.8

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0.2

0

0

2

4

6

8 10 12

0

0

2

4

6

8 10 12

0

0

2

4

6

8 10 12

0

pMAGE-D4

0

2

4

6

GCV concentration (µM)

C 1000

Tumor volume (mm3)

900 800 700

non-transduc edGCV non-transduced w/ pCMV w/ GCV pCMV promoterless w/ GCV promoterless pSSX4-255 w/ GCV pSSX4-255 pSSX4-485 w/ GCV pSSX4-485 pSSX4-255 w/ow/o GCV pSSX4-255 pSSX4-485 w/ow/o GCV pSSX4-485

600 500 400 300 200 100 0

D

123

0

5

10 15 Days after injection

20

25

8 10 12

J Neurooncol (2011) 104:497–507 b Fig. 4 In vitro and in vivo ganciclovir (GCV) sensitivity of glioma

and telomerase-immortalized fibroblast cell lines transduced with retrovirus vector harboring HSVtk. a Schema of the self-inactivating retroviral vectors used in this study. The expression of HSVtk was regulated by SSX4 promoter consisting of 255 or 485 bp upstream of transcriptional start site. MAGE-A3, MAGE-D4, and cytomegalovirus promoter (pCMV) were used for nonregulated expression of HSVtk. b The human glioma cell line U87MG, mouse glioma cell line RSV-M, and telomerase-immortalized human fibroblast cell line BJ-5ta were transduced with retroviral vector. Ganciclovir was applied at various concentrations, and cell toxicity was measured by MTT assay 5 days later. Data are expressed as the ratio of absorbance compared with conditions without GCV. c Tumor formation of subcutaneously administered RSV-M cells nontransduced or transduced with retroviral vector in syngeneic mice. Nontransduced or transduced RSV-M cells (5 9 106) were injected subcutaneously into the right flank region of C3H/HeN mice. Ganciclovir (25 mg/kg) or phosphate-buffered saline was injected intraperitoneally every day for 24 days. Tumor volume was calculated every other day. The error bars represent the standard deviation. d Immunohistochemical expression of HSVtk in subcutaneous tumor. HSVtk immunohistochemical staining of subcutaneous tumors 30 days after inoculation of RSV-M cells (left), transduced cells with CMV (middle) or SSX4 (right) promoter constructs (scale bars 100 lm). The inset shows a magnified image of HSVtk-expressing cells (scale bars 50 lm)

reported that the CpG island of the MAGE-A3 promoter region is heavily methylated in NHA cells and partially methylated in U87MG cells [26]. The expression of MAGE-A3 in NHA and brain tumor cells may be tightly regulated by DNA methylation, as has been reported in tumor cells derived from other tissues. Given these results, normal cells may retain the trans-acting factor(s) that activate the MAGE-A3 promoter but negatively regulate expression in an epigenetic manner. It has been reported that the expression of the SSX 1, 2, 4, and 5 genes is induced by the inhibition of DNA methylation or histone acetylation in colon cancer cell lines [27]. The SSX4 promoter is affected by epigenetic regulation, as are other CTA genes. However, the epigenetic influence on the activity of the SSX4 promoter was very limited in the present study because the promoter assays used transiently transfected cells. One explanation for the tumor-specific expression of the SSX4 gene is that trans-acting element(s) recruit at the SSX4 promoter in tumor cells. Determination of the SSX4 promoter region using unidirectional deleted mutants showed a gradual decrease of activity in tumor cells as the promoter fragments were shortened (Fig. 3). We cannot rule out the possibility that suppressive cis-elements located nearby are required for promoter activity, although deleted fragments of the SSX4 promoter never showed activation in NHA or BJ-5ta cells. In addition, CTA gene expression is not induced by treatment with the methylation inhibitor 5-aza-20 -deoxycytidine in various normal cells [28]. This supports the idea that a tumor- or testis-specific factor(s) is required for the expression of CTA genes including SSX4.

505

Transcription factors involved in tumor- or testis-specific expression of CTA genes are largely unknown. Transcription factors that specifically recruit CTA genes promoters are related to regulator-imprinted sites (BORIS), a cancer-testis gene that regulates the expression of the CTA gene NY-ESO-1 [29]. However, the expression of BORIS was not detected in our brain tumor cell lines (Fig. 1); hence, the effects appear to be limited by tissue or cell type. Sp1 cooperates with the hTERT or urokinasetype plasminogen activator receptor (u-PAR) gene for tumor-specific expression [30, 31], and u-PAR promoter regions also contain Sp1- and AP-1-binding sites involved in tumor-specific gene expression. On the basis of computational analysis of the SSX4 promoter using the Transcription Element Search System (TESS; http://www.cbil. upenn.edu/cgi-bin/tess/tess), we predict that the minimal SSX4 promoter contains several Sp1- and c-Jun/AP-1binding sites (data not shown). Frequent expression of AP1 in high-grade astrocytoma has been reported, and the constitutively active form of c-Jun NH2-terminal kinase is expressed in glial tumors [32, 33]. AP-1 may be responsible for SSX4 promoter activity. Further analysis is required to identify the transcription factor(s) inducing the testis- and tumor-specific expression of SSX4. We have also reported that glioma stem cells preferentially express the SSX4 gene in some glioma cell lines [26]. Cancer-testis genes are expressed in immature cells and are hallmarks of tumor malignancy. The promoter activity of the SSX4 gene in tumor stem cells remains to be elucidated; the promoter may be active in such undifferentiated cells involved in tumor malignancy. Hence, the SSX4 promoter may be useful in controlling gene expression in immature tumor cells and maintain potential therapeutic benefit. In the past, a clinical trial of HSVtk/GCV for malignant glioma was held at the National Institutes of Health in the USA [34]. The protocol consists of transplantation of retrovirus-producing cells harboring HSVtk and GCV administration. The outcome was limited to partial tumor regression in these patients, and prognosis did not improve substantially. The reasons for this lack of success are problems associated with the instability of cells transplanted into the patient’s brain and the low transduction rate with a low viral titer. To solve these problems, we have developed a retrovirus packaging cell line that yields hightiter retrovirus and showed complete remission using suicide gene therapy in a mouse glioma model [9]. Using this system, we are establishing a retrovirus-producing cell line harboring HSVtk under the SSX4 promoter. In another approach, lentivirus vector is a very attractive candidate for improving the transduction rate, because this vector can infect both dividing and quiescent cells. The transduction efficacy of the glioma tissue was greatly improved by

123

506

lentivirus vector [35]. However, the safety of gene therapy with lentivirus vector should be investigated, because normal cells including neurons are also transduced at high level. To preserve normal tissues in the gene therapy with lentivirus vector, SSX4 promoter may be suitable for restriction of the suicide gene expression. For clinical application of the suicide gene therapy using SSX4 promoter, further investigation of intracranial transplantation of glioma cells in mouse models and analysis of the frequency of SSX4 expression in brain tumor tissues are still needed.

References 1. DeAngelis LM, Gutin PH, Leibel SA, Posner JB (2002) Intracranial tumors: diagnosis and treatment. Martin Dunitz, London, pp 367–394 2. Mahaley MS Jr, Mettlin C, Natarajan N, Laws ER Jr, Peace BB (1989) National survey of patterns of care for brain-tumor patients. J Neurosurg 71:826–836 3. Deen DF, Chiarodo A, Grimm EA et al (1993) Brain tumor working group report on the 9th international conference on brain tumor research and therapy. Organ system program, National Cancer Institute. J Neurooncol 16:243–272 4. Culver KW, Ram Z, Walbridge S, Ishii H, Oldfield EH, Blaese RM (1992) In vivo gene transfer with retroviral vector producer cells for treatment of experimental brain tumors. Science 256: 1550–1552 5. Vincent E, Vogels R, Van Someren G et al (1996) Herpes simplex virus thymidine kinase gene therapy for rat malignant brain tumors. Human Gene Ther 7:197–205 6. Okada H, Miyamura K, Itoh T et al (1996) Gene therapy against an experimental glioma using adeno-associated virus vectors. Gene Ther 3:957–964 7. Kramm CM, Chase M, Herrlinger U et al (1997) Therapeutic efficiency and safety of a second-generation replication-conditional HSV1 vector for brain tumor gene therapy. Hum Gene Ther 8:2057–2068 8. Nettelbeck DM, Je´roˆme V, Mu¨ller R (2000) Gene therapy: designer promoters for tumour targeting. Trends Genet 16:174–181 9. Tamura K, Tamura M, Ikenaka K et al (2001) Eradication of murine brain tumors by direct inoculation of concentrated high titer-recombinant retrovirus harboring the herpes simplex virus thymidine kinase gene. Gene Ther 8:215–222 10. Majumdar AS, Hughes DE, Lichtsteiner SP, Wang Z, Lebkowski JS, Vasserot AP (2001) The telomerase reverse transcriptase promoter drives efficacious tumor suicide gene therapy while preventing hepatotoxicity encountered with constitutive promoters. Gene Ther 8:568–578 11. Song JS, Kim HP, Yoon WS et al (2003) Adenovirus-mediated suicide gene therapy using the human telomerase catalytic subunit (hTERT) gene promoter induced apoptosis of ovarian cancer cell line. Biosci Biotechnol Biochem 67:2344–2350 12. Hakin-Smith V, Jellinek DA, Levy D et al (2003) Alternative lengthening of telomeres and survival in patients with glioblastoma multiforme. Lancet 361:836–838 13. Yun HJ, Cho YH, Moon Y et al (2008) Transcriptional targeting of gene expression in breast cancer by the promoters of protein regulator of cytokinesis 1 and ribonuclease reductase 2. Exp Mol Med 40:345–353

123

J Neurooncol (2011) 104:497–507 14. Caballero OL, Chen YT (2009) Cancer/testis (CT) antigens: potential targets for immunotherapy. Cancer Sci 100:2014–2021 15. Kumanishi T, Ikuta F, Yamamoto T (1973) Brain tumors induced by Rous sarcoma virus, Schmidt–Ruppin stain III. Morphology of brain tumors induced in adult mice. J Natl Cancer Inst 51:95–109 16. Yoshimatsu T, Tamura M, Kuriyama S, Ikenaka K (1998) Improvement of retroviral packaging cell lines by introducing the polyomavirus early region. Hum Gene Ther 9:161–172 17. Scanlan MJ, Gure AO, Jungbluth AA et al (2002) Cancer/testis antigens: an expanding family of targets for cancer immunotherapy. Immunol Rev 188:22–32 18. Volloch V, Schweitzer B, Rits S (1994) Ligation-mediated amplification of RNA from murine erythroid cells reveals a novel class of beta globin mRNA with an extended 50 -untranslated region. Nucl Acids Res 22:2507–2511 19. Suzuki Y, Yoshitomo-Nakagawa K, Maruyama K, Suyama A, Sugano S (1997) Construction and characterization of a full length-enriched and a 50 -end-enriched cDNA library. Gene 200: 149–156 20. Emerman M, Temin HM (1984) Genes with promoters in retrovirus vectors can be independently suppressed by an epigenetic mechanism. Cell 39:449–467 21. Panagopoulos I, Mertens F, Isaksson M et al (2001) Clinical impact of molecular and cytogenetic findings in synovial sarcoma. Genes Chromosomes Cancer 31:362–372 22. Brett D, Whitehouse S, Antonson P, Shipley J, Cooper C, Goodwon G (2007) The SYT protein involved in the t(X;18) synovial sarcoma translocation is a transcriptional activator localised in nuclear bodies. Hum Mol Genet 6:1559–1564 23. Lim FI, Soulez M, Koczan D, Thiesen HJ, Knight JC (1998) A KRAB-related domain and a novel transcription repression domain in proteins encoded by SSX genes that are disrupted in human sarcomas. Oncogene 17:2013–2018 24. Cronwright G, Le Blanc K, Go¨therstro¨m C, Darcy P, Ehnman M, Brodin B (2005) Cancer/testis antigen expression in human mesenchymal stem cells: down-regulation of SSX impairs cell migration and matrix metalloproteinase 2 expression. Cancer Res 65:2207–2215 25. Maio M, Coral S, Fratta E, Altomonte M, Sigalotti L (2003) Epigenetic targets for immune intervention in human malignancies. Oncogene 22:6484–6488 26. Yawata T, Nakai E, Park KC et al (2010) Enhanced expression of cancer testis antigen genes in glioma stem cells. Mol Carcinog 49:532–544 27. Gu¨re AO, Wei IJ, Old LJ, Chen YT (2002) The SSX gene family: characterization of 9 complete genes. Int J Cancer 101:448–453 28. Natsume A, Wakabayashi T, Tsujimura K et al (2008) The DNA demethylating agent 5-aza-20 -deoxycytidine activates NY-ESO-1 antigenicity in orthotopic human glioma. Int J Cancer 122: 2542–2553 29. Kang Y, Hong JA, Chen GA, Nguyen DM, Schrump DS (2007) Dynamic transcriptional regulatory complexes including BORIS, CTCF and Sp1 modulate NY-ESO-1 expression in lung cancer cells. Oncogene 26:4394–4403 30. Liu L, Ishihara K, Ichimura T et al (2009) MCAF1/AM is involved in Sp1-mediated maintenance of cancer-associated telomerase activity. J Biol Chem 284:5165–5174 31. Leupold JH, Asangani I, Maurer GD, Lengyel E, Post S, Allgayer H (2007) Src induces urokinase receptor gene expression and invasion/intravasation via activator protein-1/p-c-Jun in colorectal cancer. Mol Cancer Res 5:485–496 32. Assimakopoulou M, Varakis J (2001) AP-1 and heat shock protein 27 expression in human astrocytoma. J Cancer Res Clin Oncol 127:727–732

J Neurooncol (2011) 104:497–507 33. Tsuiki H, Tnani M, Okamoto I et al (2003) Constitutively active forms of c-Jun NH2-terminal kinase are expressed in primary glial tumors. Cancer Res 63:250–255 34. Ram Z, Culver KW, Oshiro EM et al (1997) Therapy of malignant brain tumors by intratumoral implantation of retroviral vector-producing cells. Nat Med 12:1354–1361

507 35. Beyer WR, Westphal M, Ostertag W, von Laer D (2002) Oncoretrovirus and lentivirus vectors pseudotyped with lymphocytic choriomeningitis virus glycoprotein: generation, concentration, and broad host range. J Virol 76:1488–1495

123