Effect of EBV LMP1 Targeted Dnazymes on Cell Proliferation And

Effect of EBV LMP1 targeted DNAzymes on cell proliferation and apoptosis Zhong-Xin Lu,1 Mao Ye,1 Guang-Rong Yan,1 Qun Li,1 Min Tang,1 Leo M Lee,2 Lun-Quan Sun,3 and Ya Cao1 1Cancer Research Institute, Xiangya School of Medicine, Central South University, Changsha, China; 2Laboratory of Molecular Technology SAIC-Frederick, National Cancer Institute, Frederick, Maryland, USA; and 3St Vincent’s Clinical School, Faculty of Medicine, The University of New South Wales, Sydney, Australia.

The latent membrane protein (LMP1) encoded by Epstein–Barr virus (EBV) has been suggested to be one of the major oncogenic factors in EBV-mediated carcinogenesis. RNA-cleaving DNA enzymes are catalytic nucleic acids that bind and cleave a target RNA in a highly sequence-specific manner. In this study, we explore the potential of using DNAzymes as a therapeutic approach to EBV- associated carcinomas by targeting the LMP1 gene. In all, 13 different phosphorothioate-modified ‘‘10–23’’ deoxyribozymes (DNAzymes) were designed and synthesized against the LMP1 mRNA and transfected into B95-8 cells, which constitutively express the LMP1. Fluorescence microscopy was used to examine the cellular uptake and distribution in B95-8 cells. As demonstrated in Western blots, three out of 13 deoxyribozymes significantly downregulated the expression of LMP1 in B95-8 cells. These DNAzymes were shown to markedly inhibit B95-8 cell growth compared with a disabled DNAzyme and untreated controls, as determined by an alamarBlue Assay. It was further demonstrated that these DNAzymes arrested the B95-8 cells in G0/G1 using flow cytometry. Interestingly, the active DNAzymes could also downregulate the expression of Bcl-2 gene in treated cells, suggesting a close association between the LMP1 and Bcl-2 genes and their involvement in apoptosis. This was further confirmed with the result that the DNAzymes could induce the release of cytochrome c from mitochondria, which is the hallmark of the apoptosis. The present results suggest that the LMP1 may present a potential target for DNAzymes towards the EBV-associated carcinoma through cell proliferation and apoptosis pathways. Cancer Gene Therapy (2005) 12, 647–654. doi:10.1038/sj.cgt.7700833 Published online 1 April 2005 Keywords: latent membrane protein; DNAzymes; EBV-associated carcinoma; targeted suppression

pstein–Barr virus (EBV) is a gamma herpesvirus role in a varietyof human epithelial cancers and possibly Einfecting more than 90% of adults worldwide and is also in hyperplasias and certain dysplasias preceding the first human virus identified to be associated with carcinomas. human cancers.1 EBV has been implicated in the Latent membrane protein 1 (LMP1) is considered as pathogenesis of several human malignancies including a major oncogenic protein encoded byEBV because lymphoproliferative malignancies such as Burkitt’s and it could transform rodent fibroblasts and was tumorigenic Hodgkin’s lymphomas as well as epithelial tumors, in the nude mice.4 LMP1 is an integral membrane protein including nasopharyngeal carcinoma (NPC) and gastric composed of a short cytoplasmic N-terminus of 24 amino carcinoma.2 More recently, there have been scattered acids, six transmembrane domains of 162 amino acids reports linking EBV with typical epithelial cancers of and a cytoplasmic C-terminus of 200 amino acids. The other primarysites including breast, lung, colon and six hydrophobic transmembrane domains facilitate prostate carcinomas.3 These studies suggested that EBV, spontaneous self-oligomerization and create a complex rather than being associated with restricted groups of that constitutivelyactivates cell-signaling pathways uncommon malignancies, might in fact playan oncogenic via a cytoplasmic C-terminus that mimics tumor necrosis factor (TNF) receptor superfamilymembers. While the C-terminal activating region 1 (CTAR1) of LMP1 is Received September 23, 2004. located proximal to the membrane and binds to the Address correspondence and reprint requests to: Dr Lun-Quan Sun, Faculty of Medicine, University of New South Wales, Sydney, TNF receptor-associated factors (TRAFs), the CTAR2 Australia. E-mail: [email protected] or Professor Ya Cao, is located toward the far C-terminus of LMP1 and Cancer Research Institute, Xiangya School of Medicine, Central binds the TNF receptor-associated death domain protein South University, Changsha, Hunan, 410078, PR China. (TRADD) and receptor-interacting protein (RIP). E-mail: [email protected] Consequently, LMP1 can trigger several signaling DNAzyme-mediated suppression of LMP1 expression Z-X Lu et al 648 pathways that lead to the activation of several transcrip- Design and construction of DNAzymes tion factors, including NF-kB, STATs and AP-1.5–7 EBV has been shown to possess some mutations in LMP1 Activation of NF-kB or AP-1 byLMP1 has been linked and these sequence changes maybe associated with the to the upregulation of some cellular proteins and viral oncogenic potential. In order to identifysome inhibition of apoptosis. Depending on the cell types, conserved regions in the LMP1 gene for DNAzyme expression of LMP1 has been shown to playdifferent targeting, we aligned the LMP1 sequences from B95-8 roles in response to biological and physiological stimulus. cells with a panel of eight EBV isolates, including For example, the LMP1 expression was essential in the representatives of four defined groups of European immortalization of the EBV-infected B cells, and in LMP1 variants (groups A–D) and Chinese NPC-derived human epithelial cells LMP1 was suggested to be involved LMP1.18 All DNAzymes contained the 10–23 catalytic in tumor progression and invasions. More significantly, motif (ggctagctacaacga) and 9 bp hybridizing arms LMP1 is expressed in the majorityof EBV-associated specific for RU cleavage sites (R ¼ G or A) in the human malignancies, which supports the notion that the conserved regions of the LMP1 transcript. DNAzymes LMP1 maybe one of the keyfactors contributing to 8 were selected in the studyif their hybridizationstability EBV-mediated tumorigenesis. Therefore, genetic manip- 19 with their targets (D G1)is À25 kcal/mol. Based on ulation of the LMP1 expression mayprovide a novel o the analyses of sequences, thermodynamics and site strategyfor the treatment of EBV-associated human distribution within the LMP1 gene, 13 DNAzymes were cancers. synthesized with two phosphorothioate linkages on both Currently, there are several commonly used technologies arms (Table 1). The control oligonucleotide of DZ10 was for gene suppression, which include antisense oligonucleo- designed byinverting catalyticcore sequence. Where tides, ribozymes, DNAzymes and RNA interference indicated, DZ10 DNAzyme was labeled with a fluorescein (RNAi). The relative merits of antisense, RNAi, ribozymes isothiocyanate (FITC) at the 50-end for cellular uptake and DNAzymes for sequence-specific knockdown of target study. RNA in different situations were recentlycompared by Scherer and Rossi.9 DNAzymes are synthetic, single- stranded DNA catalysts that can be engineered to bind DNAzyme transfection of B95-8 cells to their complementarysequence in a target messenger Cells were split the daybefore transfection. On the dayof RNA (mRNA) through Watson–Crick base pairing.10 A transfection, cells were harvested and washed once with general model for the DNAzyme has been proposed, and is cold PBS. 2 Â 106 cells per well were seeded in a six-well known as the ‘‘10–23’’ model. DNAzymes following the plate in 1 ml of fresh RPMI 1640 medium containing 10% ‘‘10–23’’ model, also referred to simplyas ‘‘10–23 FBS. The DNAzymes/tetra(4-methylpyridyl) porphyrine DNAzymes’’, have a catalytic domain of 15 deoxyribonu- (TMP) mixtures were made at a charge ratio of 1 with cleotides, flanked bytwo substrate-recognition domains of 2 mM DNAzyme oligonucleotides as described.20 The seven to nine deoxyribonucleotides each. In vitro analyses mixture was then added to the cells and incubated at 371C show that this type of DNAzyme can effectively cleave its for 4 hours, followed bythe addition of complete medium substrate RNA at purine:pyrimidine junctions under to the wells and further incubation for 20 hours. physiological conditions.11 These agents have been used in a number of in vitro and in vivo applications to inhibit Fluorescence microscopic analysis the expression of their target gene12 and their dependent At 20 hours after transfection, cells were rinsed twice with genes.13 Their capacityto block development of a diverse PBS, once with 2 ml of .2 M glycine (pH 2.8) buffer, and range of pathologies in animal models suggests a potential then once with PBS. Cell nuclei were stained with 5 ng/ml use of DNAzymes as therapeutic tools.12–17 Hoescht 33258 or 50 ng/ml PI in PBS for 5 minutes prior In this report, we designed and constructed the specific to fixing with 4% paraformaldehyde in PBS for 30 DNAzymes that were shown to be effective in suppressing minutes at ambient temperature.21 Cells were rinsed with expression of the target protein LMP1 and affecting some PBS and then mounted in 75% glycerol. Microphoto- LMP1-associated biological pathways. These results graphs were acquired under UV (for Hoescht staining) demonstrated the feasibilityof targeting the LMP1 and B2 (for FITC-labeled DNAzymes) filters using a mRNA as a potential therapeutic strategyfor EBV- Nikon Optiphot-2 microscope (Tokyo, Japan) equipped associated carcinomas. with a Nikon Microflex UFX-DX photomicrographic attachment.

Materials and methods Western blot analysis Cell culture B95-8 cells grown in six-well plates were transfected as The B95-8 cell line (ATCC CRL-1612) was grown in described above and incubated for 20 hours at 371Cin RPMI1640 medium supplemented with 10% fetal bovine medium supplemented with 10% FBS. Cells were then serum (FBS, GIBCO). All media were supplemented with washed with ice-cold PBS and lysed in the lysis buffer 10 mg/ml streptomycin and 10 IU/ml penicillin and all (10 mM Tris-HCl, pH 8.0, 1 mM EDTA, 2% SDS, 5 mM cultures were incubated in a humidified atmosphere of DTT, 10 mM PMSF, Proteinase inhibitors mix) for 30 5% CO2 at 371C. minutes on ice. The lysate was centrifuged for 5 minutes

Cancer Gene Therapy DNAzyme-mediated suppression of LMP1 expression Z-X Lu et al 649 Table 1 Location of DNAzyme target sites in the LMP1 mRNA and DNAzyme design

Name Cleavage sitea DNAzyme sequences Modificationb

Dz10a a (1) u 50cgtgttccaGGCTAGCTACAACGAggtcagggt30 50FITC/2+2PS Dz10 a (1) u 50cgtgttccaGGCTAGCTACAACGAggtcagggt30 2+2 PS Dz1 a (167) u 50gcaaaggaaGGCTAGCTACAACGAagaggacaa30 2+2 PS Dz2 a (211) u 50ttctgaagaGGCTAGCTACAACGAaaagatgat30 2+2 PS Dz3 g (233) u 50ccaagtggaGGCTAGCTACAACGAagagaaggt30 2+2 PS Dz4 a (476) u 50tgttgtagaGGCTAGCTACAACGAagagagcaa30 2+2 PS Dz5 g (492) u 50tagagtccaGGCTAGCTACAACGAcagttttgt30 2+2 PS Dz6 g (505) u 50ggagatcaaGGCTAGCTACAACGAcaatagagt30 2+2 PS Dz7 a (593) u 50agggagtcaGGCTAGCTACAACGAcgtggtggt30 2+2 PS Dz8 a (785) u 50tgtgggccaGGCTAGCTACAACGAtgtcatcag30 2+2 PS Dz9 a (893) u 50ccattgtcaGGCTAGCTACAACGAcagtgttgt30 2+2 PS Dz11 a (950) u 50gggcctccaGGCTAGCTACAACGAcatttccag30 2+2 PS Dz12 g (1109) u 50tcatctccaGGCTAGCTACAACGAcggaaccag30 2+2 PS Dz13 g (1138) u 50ttagctgaaGGCTAGCTACAACGAtgggccgtg30 2+2 PS Control ODN 50cgtgttccaGGTATGTACAACGAggtcagggt30 2+2 PS aCleavage sites are defined as AU or GU along the LMP1 mRNA. Numbers in brackets are the nucleotide positions of A or G. b2+2 PS, 2 phosphorothioate linkages at each end of DNAzyme oligonucleotides.

at 14,000 rpm in a microfuge. After protein quantification ice-cold PBS and suspended in 75% ethanol at À201C bythe BCA AssayReagent (Pierce Chemical, Inc.), 50 mg overnight. Fixed cells were centrifuged and washed with of the total proteins from the DNAzyme-, control ODN- PBS twice. For cell cycle analysis, cells were stained with or mock-transfected cells were mixed with sample buffer 50 mg/ml of PI and .1% of RNase A in 400 ml PBS in a and boiled for 5 minutes. The samples were then resolved light-proof tube at 251C for 30 minutes. Stained cells were on 10% polyacrylamide SDS gel and transferred onto a assayed on FACSort (Becton Dickinson) and the cell nitrocellulose membrane byelectroblotting. The mem- cycle parameters were determined using the CellQuest brane was incubated in blocking buffer (TBS containing software program (Becton Dickinson). 5% skim milk and .1% Tween 20) for 2 hours, followed byincubation with primaryantibodydiluted in the same buffer (1:1000). The membrane was washed in TBS Cytochrome c release assay containing .1% Tween 20 and further incubated with a The cells were transfected as described above. The secondaryantibodythat could be detected using a proteins from cytosolic fraction and mitochondrial peroxidase-conjugated anti-IgG at 1:10,000. Protein fraction were extracted using cytochrome c apoptosis expressions were determined using a supersignal chemi- assaykit (BioVision), according to the manufacturer’s luminescence system (ECL, Pierce), followed by exposure instruction. Western blots were performed using a CytoC- to autoradiographic film. The studyemployedantibodies specific antibody(K257-100, BioVision). against LMP1 (cs1-4, DAKO), a-tubulin (sc-5286, Santa Cruz BiotechnologyInc.), Bcl-2 (C-2, sc-7382, Santa Cruz BiotechnologyInc.) and CytoC(K257-100, BioVision), Ik-Ba (sc-203, Santa Cruz BiotechnologyInc.) and NF- Results and discussion kB p65 (sc-109, Santa Cruz BiotechnologyInc.). Cellular uptake and distribution of LMP1 DNAzymes Cell proliferation assays A cell line B95-8 was used to examine DNAzyme efficacy in downregulation of the LMP1 gene expression and B95-8 cells grown in 96-well micro-titer plates were impact on cellular functions. To facilitate deliveryof transfected at .5, 1 and 2 mM of DNAzymes or control DNAzyme oligonucleotides into cells, a cationic porphy- ODN as described above. Cells were then incubated at rin, TMP, was used as a transfection reagent for 371C and 25 ml of alamarBlue Assayreagent (Serotec Ltd) intracellular delivery.20 In order to increase DNAzyme was added to each well at different time points. The plates stabilityin cells, two phosphorothioate linkages were were allowed for a further incubation of 3 hours and incorporated into each of the arms in DNAzymes, which analyzed spectrophotometrically (absorbance at 570 and has been shown to be an efficient wayto protect 600 nm, Bio-Tek Inc.). DNAzyme oligonucleotides from enzymatic degradation.22 When the transfected cells were subjected to fluorescent Cell cycle analysis microscopic analysis, efficient delivery was observed using The transfected B95-8 cells were incubated for different the TMP-complexed phosphorothioate-Dz. Importantly, a times at 371C in medium supplemented with 10% FBS. significant nuclear deliverywas seen at 24-hour time point Untreated and treated cells were then washed with (Fig 1).

Cancer Gene Therapy DNAzyme-mediated suppression of LMP1 expression Z-X Lu et al 650

Figure 1 Cellular distribution of fluorescent labeled DNAzyme. B95-8 cells were incubated with 2 mM 33-mer FITC-labeled phosphorothioate- modified LQ10 DNAzyme complexed with TMP at C/R of 1. Cells were transfected for 24 hours, treated with glycine (pH 2.8) buffer, stained with 5 ng/ml Hoescht 33258 or 50 mg/ml PI, and fixed with 4% paraformaldehyde. Nuclei stained with Hoescht dye display a bright blue color, and the PI dye exhibits a red hue. (a) Nuclei stained with Hoescht dye. (b, d) FITC-labeled DNAzyme. (c) Nuclei stained with PI dye. Â 150.

As for all the oligonucleotide-based agents such as for their effect on the LMP1 expression at the protein antisense, ribozyme or siRNA, the DNAzyme delivery level using Western blotting. As shown in Figure 2, has been the major hurdle in assayof its efficacy.Many DNAzymes DZ1, DZ7 and DZ10 (2 mM) strongly types of delivery agents have been widely used for this inhibited LMP1 protein expression in B95-8 cells, purpose, including liposomes, nanoparticles or electro- compared with a control ODN, TMP transfection reagent poration.23 In this study, TMP was shown to be a simple and untreated controls. Based on the initial screening and efficient transfection reagent due to its high complex- results, one of the efficient DNAzymes, DZ10, was used ing efficiency, low cellular toxicity and preferential for further studyand confirmed that DZ10 could inhibit nuclear uptake. LMP1 expression in B95-8 cells in a dose-dependent manner (Fig 3). Inhibition of LMP1 protein expression in B95-8 cells using DNAzymes Impact of LMP1 DNAzymes on B95-8 cell growth DNAzymes are enzymatic molecules composed entirely There have been manyreports showing that the EBV of DNA that can be made to bind and cleave any LMP1 is one of several latentlyexpressed genes essential RNA molecule both in vitro and in vivo. However, like for the transformation of B lymphocytes and affects other agents that function byhybridizationwith single- cellular growth and differentiation of epithelial cells in stranded RNA, the DNAzyme must compete with the vitro.25,26 The broad expression of LMP1 in undiffer- target’s own stable intramolecular base pairing, which entiated NPC suggests this viral oncoprotein as a forms its characteristic secondarystructures. Fortunately, keyeffector molecule in NPC pathogenesis. 8 We pre- the 10–23 DNAzyme cleavage sites are plentiful in viouslyshowed that the EBV LMP1 might modulate most biological substrates and thus provide a host EGFR promoter activityin a NF- kB-dependent manner of opportunities to achieve maximum cleavage efficiency. in NPC.27 Thus, downregulation of the LMP1 expression In addition to target site accessibility, the thermody- could result in a growth inhibition in LMP1-expressing namics of enzyme–substrate interactions also have cells. a strong influence on the efficiencyof DNAzyme- To examine the effect of downregulation of LMP1 on mediated hydrolysis of a long RNA.24 Taking into cell growth, active DNAzymes DZ1, DZ7 and DZ10 were consideration these factors, we designed 13 different transfected into B95-8 cells and the proliferation rate was 10–23 DNAzymes targeted against AU or GU sites assessed byalamarBlue assay.As shown in Table 2, the along the full-length LMP1 mRNA (Table 1). These proliferation rate was inhibited in all the Dz-treated cells DNAzymes were transfected into B95-8 cells and assayed at 24-hour time point, followed bya recoveryat 48- and

Cancer Gene Therapy DNAzyme-mediated suppression of LMP1 expression Z-X Lu et al 651

Figure 3 Dose–response analysis of anti-LMP1 Dz10. B95-8 cells grown in six-well plates were transfected with different concentra- tions of DZ10 (.5, 1, 2), or mock-transfected and incubated in medium containing 10% FBS at 371C for 24 hours. Total cell lysates were analyzed by Western blots. Expression level for each protein was estimated by densitometry and presented as a ratio to the loading control a-tubulin (lower panel).

phase and a marginal reduction of cells in S phase (Table 3). This result correlated well to the data from the cell proliferation assayat 24 hours time point (Table 2), further suggesting an involvement of the LMP1 gene in cell proliferation in NPC development.

Inhibition of Bcl-2 protein expression and induced apoptosis in B95-8 cells with LMP1 DNAzymes There have been reports that the tumorigenicityof the EBV-positive cells was associated with an increase Figure 2 Inhibition of LMP1 protein expression in B95-8 cells by in the antiapoptotic protein Bcl-2 and resistance DNAzymes. B95-8 cells grown in six-well plates were transfected to apoptosis.28 Abdulkarim et al.29 reported that antiviral with DNAzymes (DZ1-13 or Control ODN, 2 mM), or mock-trans- agent Cidofovir could downregulate the EBV oncoprotein fected and incubated in medium containing 10% FBS at 371C for 24 LMP1 expression, which was associated with a decrease hours. Total cell lysates were analyzed by Western blots using in antiapoptotic Bcl-2 protein and an increase of the mouse monoclonal antibodies targeting EBV LMP1 (a), and the proapoptotic Bax protein in Raji (BL) and C15 (NPC) LMP1 expression level was quantitated by densitometry (b). cells. However, the mechanism of Codofovir’s effect on the LMP1 and bcl-2 protein expression was not defined. Here, we used the LMP1-specific DNAzymes to further investigate the association of the viral 72-hour points, which maysuggest a gradual degradation LMP1 and cellular Bcl-2 protein in B95-8 cells. Western of the transfected oligonucleotides within the transfected blotting results indicated that the active LMP1 cells. When three DNAzymes were combined and DNAzymes could concomitantly downregulate the ex- transfected into the cells, an increase in the inhibitory pression of Bcl-2 gene in the treated cells (Fig 4a and b). effect was observed, compared with anyof individual Analyses of the proteins of cytosolic fraction and DNAzyme treatment. This combined effect may be due to mitochondrial fraction, respectively, using cytochrome c the multiple targeting of the accessible site within the apoptosis assaykit (BioVision) confirmed that the active LMP1 mRNA structures. DNAzymes could also induce the release of cytochrome c To further investigate the cause of the reduced cell from mitochondria, which is the hallmark of the proliferation rate in the DZ-treated B95-8 cells, flow apoptosis (Fig 5a and b). cytometry was employed to study the effect of the Taken together, these results, for the first time, DNAzymes on cell cycle. The results showed that DZ-7- demonstrated a direct relationship between the LMP1 treated cells were arrested in G0/G1 at 24 hours post and Bcl-2 at a gene-specific level and their involvement in transfection, with a significant reduction of cells in G2/M apoptosis.

Cancer Gene Therapy DNAzyme-mediated suppression of LMP1 expression Z-X Lu et al 652 Table 2 Inhibition of B95-8 cell proliferation by active LMP1 DNAzymes

Proliferation rate (%)a

DNAzymes Concentration (mM) 12 hours 24 hours 36 hours 48 hours 72 hours

DZ1 .5 99.0 63.9 87.6 89.8 87.7 1 95.2 55.4 87.1 87.3 91.3 2 89.4 46.3 82.4 75.5 89.5

DZ7 .5 96.1 68.5 86.4 90.6 93.3 1 88.2 56.8 85.4 87.5 88.2 2 85.5 42.4 88.0 85.6 87.3

DZ10 .5 87.6 66.7 82.1 84.1 85.7 1 81.0 51.3 78.7 77.0 81.9 2 72.3 39.6 69.8 79.9 81.1

DZ1+7+10 .5 95.9 59.6 73.8 85.3 88.5 1 92.4 51.0 73.0 79.6 77.3 2 85.9 28.1 61.2 67.9 81.5

Control ODN .5 98.8 95.6 87.9 99.4 89.4 1 93.0 98.7 86.9 94.6 86.7 2 96.8 98.6 89.6 86.8 85.5 aProliferation rate is calculated as a relative value of OD readings in each well to untreated cells.

Table 3 Effect of LMP1 DNAzyme on cell cycle in B95-8 cellsa Phases Cell control TMP control DZ-7 treatment

G0/G1 41.82 46.46 54.23 S 44.10 44.01 40.26 G2/M 14.08 9.53 5.52 aThe numbers represent the percentage of cells in different phases of cell cycle.

Modulation of LMP1-induced cellular NF-kB signal pathway in CNE1-LMP1 cells by DNAzymes LMP1-mediated activation of NF-kB is one of the major signal transduction pathways involved in cell growth and apoptosis in the EBV-associated NPC.30 In normal cells, NF-kB is sequestered byinhibitorymolecules, IkBs. When LMP1 activates the NF-kB pathway, one of the inhibitorymolecules, I kBa, is degraded by uniquitination, which leads to release and nuclear transportation of NF-kB.31 To elucidate the mechanism Figure 4 Inhibition of Bcl-2 protein expression in B95-8 cells by DNAzymes. B95-8 cells grown in six-well plates were transfected wherebythe LMP1-specific DNAzymespromote with DNAzymes (DZ1, DZ7 or DZ10 at 2 mM), or mock-transfected apoptotosis in LMP1-expressing cells, we examined if and incubated in medium containing 10% FBS at 371C for 24 hours. downregulation of LMP1 byDNAzymescould affect the Total cell lysates were analyzed by Western blots (a). Expression NF-kB pathway. As shown in Figure 6, the NF-kB level level for each protein was estimated by densitometry and presented in DNAzyme-treated cells remained unchanged while the as a ratio to the loading control a-tubulin (b). IkBa level was significantlyincreased compared with controls. This result suggests that suppression of LMP1 expression leads to accumulation of the IkBs in cells, In conclusion, we have developed the active DNA- which could have a global effect on cell functions zymes that could suppress the EBV LMP1 expression, regulated via the NF-kB pathway, including cell prolif- inhibit the cell proliferation in the LMP-expressing eration and apoptosis. B95-8 cells and increase apoptosis bydownregulation

Cancer Gene Therapy DNAzyme-mediated suppression of LMP1 expression Z-X Lu et al 653

Figure 5 Induction the release of cytochrome c from mitochondria in B95-8 cells. B95-8 cells grown in six-well plates were transfected with DNAzymes (DZ1, DZ7 or DZ10 at 2 mM), or mock-transfected and incubated in medium containing 10% FBS at 371C for 24 hours. Cell lysates were analyzed by Western blot using mouse monoclonal antibodies targeting human Cyt c (a). Expression level for each protein was estimated by densitometry and presented as a ratio to the loading control a-tubulin (b).

LMP1 could be a potential molecular target for nucleic acid-based therapeutics, such as DNAzymes, towards the EBV-associated carcinomas through cell proliferation and apoptosis pathways.

Acknowledgments

This research was supported bythe National Natural Science Foundation of China for Distinguished Young Scholars (No. 39525022), State KeyResearch Program, Fundamental Investigation on Human Carcinogenesis of China (No. G1998051201), and the National Natural Science Foundation of China (No. 30200009).

References

1. Epstein MA, Acjpmg BG, Barr YM, Zajac B, Henle G, Henle W. Morphological and virological investigation in cultured Burkitt tumor lymphoblasts (strain Raji). J Natl Cancer Inst. 1966;333:693–698. Figure 6 Effect of DNAzymes on NF-kB signal pathway. CNE1- 2. Rickinson AB. Epstein–Barr virus in action in vivo. N Engl J LMP1 cells grown in six-well plates were transfected with DNAzymes Med. 1998;338:1461–1463. (DZ1, DZ7 or DZ10 at 2 mM), or mock-transfected and incubated in 3. Grinstein S, Preciado MV, Gattuso P, et al. Demonstration 1 medium containing 10% FBS at 37 C for 24 hours. Total cell lysates of Epstein–Barr virus in carcinomas of various sites. Cancer were analyzed by Western blots (a). Expression level for each Res. 2002;62:4876–4878. protein was estimated by densitometry and presented as a ratio to 4. Eliopoulos AG, Young LS. Activation of the c-Jun N- the loading control a-tubulin (b). Antibodies: IkB-a (C-15): sc-203, terminal kinase (JNK) pathwaybythe Epstein–Barr virus- NF-kB p65 (a): sc-109, from Santa Cruz. encoded latent membrane protein 1 (LMP1). Oncogene. 1998;16:1731–1742. 5. Eliopoulos AG, Young LS. LMP1 structure and signal transduction. Semin Cancer Biol. 2001;11:435–444. of the bcl-2 expression. The biological activityof the 6. Hatzivassiliou E, Mosialos G. Cellular signaling pathways DNAzymes was shown to be mediated via the NF-kB engaged bythe Epstein–Barr virus transforming protein signaling pathway. The present results suggest that the lmp1. Front Biosci. 2002;7:D319–D329.

Cancer Gene Therapy DNAzyme-mediated suppression of LMP1 expression Z-X Lu et al 654 7. Gires O, Kohlhuber F, Kilger E, et al. Latent membrane 20. Benimetskaya L, Takle GB, Vilenchik M, Lebedeva I, Miller protein 1 of Epstein–Barr virus interacts with JAK3 and P, Stein CA. Cationic porphyrins: novel delivery vehicles activates STAT proteins. EMBO J. 1999;18:3064–3073. for antisense oligodeoxynucleotides. Nucleic Acids Res. 8. Tsao SW, Tramoutanis G, Dawson CW, Lo AK, Huang 1998;26:5310–5317. DP. The significance of LMP1 expression in nasopharyngeal 21. Pichon C, MonsignyM, Roche AC. Intracellular localiza- carcinoma. Semin Cancer Biol. 2002;12:473–487. tion of oligonucleotides: influence of fixative protocols. 9. Scherer LJ, Rossi JJ. Approaches for the sequence-specific Antisense Nucleic Acid Drug Dev. 1999;9:89–93. knockdown of mRNA. Nat Biotechnol. 2003;21:1457–1465. 22. Nunamaker EA, Zhang HY, Shirasawa Y, Benoit JN, Dean 10. Sun LQ, Cairns MJ, Saravolac EG, Baker A, Gerlach WL. DA. Electroporation-mediated deliveryof catalyticoligo- Catalytic nucleic acids: from lab to applications. Pharmacol deoxynucleotides for manipulation of vascular gene expres- Rev. 2000;52:325–347. sion. Am J Physiol Heart Circ Physiol. 2003;285:H2240– 11. Santoro SW, Joyce GF. A general purpose RNA-cleaving H2247. DNA enzyme. Proc Natl Acad Sci USA. 1997;94:4262–4266. 23. Dass CR, Saravolac EG, Li Y, Sun LQ. Cellular uptake, 12. Santiago FS, Lowe HC, Kavurma MM, et al. New DNA distribution, and stability of 10–23 deoxyribozymes. Anti- enzyme targeting Egr-1 mRNA inhibits vascular smooth sense Nucleic Acid Drug Dev. 2002;12:289–299. muscle proliferation and regrowth factor injury. Nat Med. 24. Cairns MJ, Hopkins TM, Witherington C, Wang L, Sun 1999;5:1264–1269. LQ. Target site selection for an RNA-cleaving catalytic 13. FahmyRG, Dass CR, Chesterman CN. Transcription DNA. Nat Biotechnol. 1999;17:480–486. factor Egr-1 supports FGFdependent angiogenesis during 25. Kaye KM, Izumi KM, Li H, et al. An Epstein–Barr virus neovascularization and tumor growth. Nat Med. that expresses onlythe first 231 LMP1 amino acids 2003;9:1026–1032. efficiently initiates primary B lymphocyte growth transfor- 14. Lowe HC, Chesterman CN, Khachigian LM. Catalytic mation. J Virol. 1999;73:10525–10530. antisense DNA molecules targeting Egr-1 inhibit neointima 26. Dawson CW, Eliopoulos AG, Blake SM, Barker R, Young formation following permanent ligation of rat common LS. Identification of functional differences between proto- cartoid arteries. Thromb Haemost. 2002;87:134–140. type Epstein–Barr virus-encoded LMP1 and a nasophar- 15. Zhang L, Gasper WJ, Stass SA, Ioffe OB, Davis MA, yngeal carcinoma derived LMP1 in human epithelial cells. Mixson AJ. Angiogenic inhibition mediated bya DNAzyme Virology. 2000;272:204–217. that targets vascular endothelial growth factor receptor 2. 27. Tao YG, Tan YN, Liu YP, et al. Nuclear factor kappa B Cancer Res. 2002;62:5463–5469. (NF-kB) dependent modulation of Epstein–Barr virus latent 16. Mitchell A, Dass CR, Sun LQ, Khachigian LM. Inhibition membrane protein 1 (LMP1) in epidermal growth factor of human breast carcinoma proliferation, migration, che- receptor (EGFR) promoter activity. Virus Res. 2004;104:61– moinvasion and solid tumour growth byDNAzymes 70. targeting the zinc finger transcription factor EGR-1. Nucleic 28. Komano J, Sugiura M, Takada K. Epstein–Barr virus Acids Res. 2004;32:3065–3069. contributes to the malignant phenotype and to apoptosis 17. Zhang G, Dass CR, Sumithran E, Girolamo ND, Sun LQ, resistance in Burkitt’s lymphoma cell line Akata. J Virol. Khachigian LM. Effect of deoxyribozymes targeting c-Jun 1998;72:9150–9156. on solid tumor growth and angiogenesis in rodents. J Natl 29. Abdulkarim B, Sabri S, Zelenika D, et al. Antiviral agent Cancer Inst. 2004;96:683–696. cidofovir decreases Epstein–Barr virus (EBV) oncoproteins 18. Fielding CA, Sandvej K, Mehl A, Brennan P, Jones M, and enhances the radiosensitivityin EBV-related malignan- Rowe M. Epstein–Barr virus LMP-1 natural sequence cies. Oncogene. 2003;17:2260–2271. variants differ in their potential to activate cellular signaling 30. Yoshizaki T, Sato H, Furukuwa M, Pagano JS. The pathways. J Virol. 2001;75:9129–9141. expression of matrix metalloproteinase 9 is enhanced by 19. Cairns MJ, King A, Sun LQ. Optimisation of the 10–23 Epstein–Barr virus latent membrane protein 1. Proc Natl DNAzyme–substrate pairing interactions enhanced RNA Acad Sci USA. 1998;95:3621–3626. cleavage activityat purine-cytosinetarget sites. Nucleic 31. Baldwin AS. The NF-kB and IkB proteins: new discoveries Acids Res. 2003;31:2883–2889. and insights. Annu Rev Immunol. 1996;14:649–681.

Cancer Gene Therapy