A Novel Micro-Linear Vector for In Vitro and In Vivo Gene Delivery and Its Application for EBV Positive Tumors

Background The gene delivery vector for DNA-based therapy should ensure its transfection efficiency and safety for clinical application. The Micro-Linear vector (MiLV) was developed to improve the limitations of traditional vectors such as viral vectors and plasmids. Methods The MiLV which contained only the gene expression cassette was amplified by polymerase chain reaction (PCR). Its cytotoxicity, transfection efficiency in vitro and in vivo, duration of expression, pro-inflammatory responses and potential application for Epstein-Barr virus (EBV) positive tumors were evaluated. Results Transfection efficiency for exogenous genes transferred by MiLV was at least comparable with or even greater than their corresponding plasmids in eukaryotic cell lines. MiLV elevated the expression and prolonged the duration of genes in vitro and in vivo when compared with that of the plasmid. The in vivo pro-inflammatory response of MiLV group was lower than that of the plasmid group. The MEKK1 gene transferred by MiLV significantly elevated the sensitivity of B95-8 cells and transplanted tumor to the treatment of Ganciclovir (GCV) and sodium butyrate (NaB). Conclusions The present study provides a safer, more efficient and stable MiLV gene delivery vector than plasmid. These advantages encourage further development and the preferential use of this novel vector type for clinical gene therapy studies.


Introduction
An effective vector for gene delivery should afford satisfactory transfection efficiency while assuring safety clinical application [1]. Traditionally, plasmid and viral-based vectors are two commonly used vectors. However, limitations such as immunogenicity and cytotoxicity reduce the clinical viability of viral vectors [2]. While plasmid-based gene transfection is considered to be less toxic, the relatively small transfection efficiency and short duration of transgene expression of plasmids have limited the feasibility of this method in clinical applications [3]. Furthermore, the numerous CpG sequences contained in the plasmid backbone can cause immunotoxic effects, including the elimination of transfected cells by the host immune responses [4]. The immune responses caused by unmethylated CpG dinucleotide motifs can further decrease the efficiency of gene transfection [5]. Another significant disadvantage for propagation of plasmids in bacteria is that it contains bacterial remnants such as lipopolysaccharides (LPS) or endotoxins, which can cause adverse clinical effects [6]. Therefore, these traditional vectors should be improved before clinical translation.
Recently, several novel approaches have been used to improve the traditional vectors applied in gene therapy. For example, previous studies have revealed that minicircle DNA was less immunogenic, had greater in vivo diffusivity and more stability than conventional plasmids [7][8][9]. These characteristics are attributed to the smaller size of the molecules and little contamination with DNA sequences that originated in the bacteria. The minimalistic immunologically defined gene expression (MIDGE) vectors created by Witting and his colleagues [10] are linear molecules containing only a promoter, a target gene and an RNA stabilizing sequence, flanked by two short hairpin oligonucleotide sequences. Each MIDGE, particularly when it was conjugated with nuclear localization signal (NLS) peptides, has greater transfection efficiency as compared with its corresponding plasmid, both in vitro and in vivo [11][12][13][14]. However, production of MIDGEs are costly and time consuming, particularly for the conjugation of NLS peptides. PCR-amplified DNA fragments, used as a model for double-stranded synthetic genes in gene therapy, have been proven to be efficient for both in vitro and in vivo gene delivery [15,16]. However, transfection efficiency of PCR-amplified DNA fragments is lower than that of plasmid, especially when the DNA fragment is delivered as cationic complexes [16]. This is likely due to the instability and poor rate of transcription of the DNA fragment when incorporated into cells. Therefore, all of these recently developed novel vectors require further modifications before they would be feasible for clinical application.
Here, we report the development of a novel linear DNA delivery vector. Briefly, the gene expression cassette was ligated with hairpin oligodeoxynucleotides (ODNs), amplified by PCR by use of a ligation mixture as the template, and purified by use of PCR cleanup kits. We have named the process the Micro-Linear Vector (MiLV). The capability and pro-inflammatory responses of the MiLV to deliver genes were investigated both in vitro and in vivo. As a proof of concept, the MiLV was evaluated as a vector in gene therapy for Epstein-Barr virus (EBV) positive cancer cells.

Cell Culture
Human embryo kidney cell line 293 (HEK 293), mouse embryonic fibroblast cell line NIH 3T3, human nasopharyngeal carcinoma line CNE2 and EBV-positive monkey (tamarin) lymphocyte cell line B95-8 were purchased from the American Type Culture Collection (ATCC) (Rockville, MD) and maintained in our laboratory. The HEK 293, NIH 3T3 and CNE2 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) and B95-8 cell line was maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) and antibiotics at 37uC in a 5% CO 2 incubator. The medium was replaced until cells became 80% confluent and then passaged using 0.25% trypsin/EDTA.

Construction of eGFP-MiLV
Procedures for constructing the eGFP-MiLV are illustrated in Figure 1. Briefly, the eGFP expression cassette, including CMV promoter (pCMV), eGFP gene and RNA-stabilizing sequences (polyadenylic acid, SV40), was bi-digested by AseI and Afl II from pEGFP-N3 plasmid (BD Biosciences Clontech, NJ) at 37uC for 4 h. Two ODN caps containing AseI and Afl II restriction enzyme site respectively were designed as structural analogues of tRNA's Dloop in eukaryotic cells and synthesized by the Shanghai Sangon Company (Shanghai, China). The sequences of ODNs were: Cap 1:59-TAG CGC TCA GTT GGG AGA GCG CTA AT -39; Cap 2:59-TTA AGG CGC TCA GTT GGG AGA GCG CC-39. The eGFP expression cassette, Cap1, and Cap2 were mixed (molar ratio, 1:10:10) and ligated overnight with T4 DNA ligase at 16uC. Then, 1 ml ligation mixture was used as the template to amplify eGFP-MiLV by PCR by use of a single primer (59 ACA AGT TCA GCG TGT CCG 39, annealing to 751 to 768 of the eGFP expression cassette) in a 50 ml reaction system. Amplification of eGFP-MiLV was performed by PCR under the following conditions: initial 94uC denaturation (2 min), followed by 35 cycles of three PCR steps (each cycle: 30 s at 94uC, 30 s at 54uC and 80 s at 72uC), and terminated with an extension prolongation for 5 min at 72uC. High success-rate DNA polymerase (Toyobo Co., Ltd., Osaka, Japan) was used to obtain sufficient DNA. After PCR, the products were purified by use of the E.Z.N.A.H Cycle-Pure Kit (Omega, USA). After the purified eGFP-MiLV was checked by DNA sequencing, it was used for cell transfection and intramuscular injection. To examine the in vitro resistance of MiLV to exonuclease, the eGFP-MiLV and eGFP fragment were incubated with Exonuclease III (Takara, Japan) at 37uC for 2 to 24 h and detected by 1% agarose gel electrophoresis.

Construction of pLMP1-MEKK1-MiLV
The EBV genome was extracted from B95-8 cells using phenol/ chloroform and purified by ethanol precipitation. The promoter of latent membrane protein 1 (pLMP1) was amplified from the EBV genome by PCR with the following primers: forward: 59 GAC ATT AAT CTC AGG GCA GTG TGT CA G 39; reverse: 59 CCG CTC GAG TTG TGC AGA TTA CAC TGC 39. The restriction enzyme sites of AseI and XhoI were contained at the 59 and 39 ends of pLMP1, respectively. The mitogen-activated protein kinase kinase (MEK) kinase 1 (MEKK1) gene with XhoI and Afl II restriction enzyme sites was amplified from the active form of pCMV/MEKK1D plasmid (Palo Alto, CA) with the following primers: forward: 59 CCG CTC GAG CCA CCA TGG CGA TGT CAG CGT CTC 39; reverse: 59 CAG CTT AAG TTT ATT TGT GAA ATT TGT GAT GC 39. Then, pLMP1 and MEKK1 were digested by Xho I at 37uC for 4 h and ligated by T4 DNA ligase at 16uC for 12 h. The pLMP1-MEKK1 gene was amplified by PCR and ligated with two caps as mentioned above. The MEKK1-MiLV was amplified with the following primer: 59 GAG TAA ATA CGG AGC TTT CAA GGA G 39. The MEKK1-MiLV is approximately 1.5 kb, including the promoter (pLMP1) and MEKK1 gene.

Cell Transfection
For transfection, cells were seeded into six-well plates at a density of 1610 4 cells/cm 2 , cultured for 24 h, and transfected with MiLV or plasmid DNA (0.1 mM DNA/cm 2 per plasmid) with Lipofectamine reagent (Gibco BRL, Gaithersburg, MD) according to the manufacturer's instructions. Briefly, cells were washed with phosphate-buffer saline (PBS) and before transfection cells resuspended in 800 ml culture medium without FBS or antibiotics. DNA and transfection reagent were diluted with 100 ml culture medium and incubated for 15 min at room temperature. Lipofectamine solution was added to the DNA solution and mixed, and then the mixture was incubated for a further 30 min (room temperature). The DNA/reagent mixture was added dropwise into the cell culture supernatant. Medium was replaced by 2 ml fresh medium supplemented with 10% FBS 2 h later. The transfection efficiency was defined as described previously [10] and evaluated 48 h after the transfection.

Flow Cytometry
The green fluorescence in transfected cells was quantified by flow cytometry by use of an Epics XL (Coulter Immunotech, Hamburg, Germany). Cells were washed twice, resuspended with ice-cold PBS, and fixed with 70% ethanol. The fluorescence was measured by use of a 530-nm/30-nm band pass filter after illumination with an argon ion laser tuned at 488 nm. Cells transfected with transfection reagent served as the control. Magnitude of expression of GFP was reported as the percentage of GFP+ cells [10].

Cell Proliferation and Cytotoxicity Assay
For cell proliferation assays, HEK 293 cells were inoculated into 6-well plates and incubated for 24 h before exposure to 1 mM plasmid or MiLV. Cells were harvested every 24 h, and then cell density was calculated by use of a hemacytometer. Cytotoxicity was determined by use of the MTT (3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide) (Sigma, St. Louis, MO) assay. Briefly, after being transfected by the MEKK1-MiLV or pCMV/ MEKK1 plasmid for 24 h, B95-8 cells were treated with 1 mM NaB for 18 h followed by treatment with 100 mg/ml ganciclovir (GCV) for 3 or 6 days. MTT was then added to each well to make a final concentration of 0.5 g/L, and then cells were incubated for a further 4 h. Supernatant solutions were then aspirated, and the cells solubilized in 200 mL dimethyl sulfoxide (DMSO). Optical density was measured at 570 nm.

In vivo Gene Transfer
BALB/c male mice (6-8 weeks, Experiment Animal Center of Sun Yat-sen University, Guangzhou, China) were reared and maintained under conventional breeding conditions with food and water ad libitum, on a 12:12 h light: dark cycle. The experimental protocol was approved by the Ethics Committee for Animal Research at Sun Yat-sen University. Twenty micrograms eGFP-MiLV and 60 mg pEGFP-N3 plasmid (equal molar) were packaged with Lipofectamine TM 2000 Reagent (1:1.5, g/ml) in a total volume of 100 ml. The mixtures were injected intramuscularly within 5 s. The control group was injected with PBS. To investigate expression of GFP, in each group, at least one mouse was euthanized each week for a total of 8 weeks. Mice were euthanized by use of standard surgical procedures. Muscle was sectioned transversely (5 mm) with a Leica CM 1850 cryostat (Leica, Nussloch, Germany) maintained at 220uC. Sections were examined for expression of GFP by use of a laser scanning confocal microscopy. The calculation of fluorescence intensity were processed with previously published method [17] with slight modification. Briefly, both the normalized photon counts and the area of eGFP signals were quantified. Then we subtracted the photon counts/second/mm 2 of region of interest (ROI) by the photon counts/second/mm 2 of the eGFP2 area and calculated the total photon counts of generated by eGFP+ cells by timing the normalized intensity with the area of eGFP+ region.

Immunization of BALB/c Mice and Cell-mediated Immune Response
Forty micrograms eGFP-MiLV and pEGFP-N3 plasmid were precipitated on 20 ml of 1 mm gold beads, respectively, according to the instructions of manufacturer (BioRad, USA). A suspension of gold beads carrying the DNA was made in 99.5% ethanol and the suspension was used to coat a 50 cm of tefzel tubing. The tubing was cut into 12.7 mm pieces and stored at 220uC before being used for gene gun immunizations. Then, 10-week-old female BALB/c mice were immunized with eGFP-MiLV and pEGFP-N3 plasmid using a gene gun (HeliosTM, BioRad, USA). Three groups (Blank, MiLV and plasmid), each containing eight rodents, were vaccinated 4 times with 2 week intervals (0, 2, 4 and 6 weeks). The primary immunization (day 0) was performed with four cartridges, and the following immunizations were each performed using two cartridges. Blood samples were collected by orbital puncture of two individuals from each group every second day during the first 14 days, and at later time-points from all mice at days 28, 42 and 56, before doing the gene gun immunization.

Measurement of Pro-inflammatory Cytokines in the Blood
At 2 h after injection of 40 mg eGFP-MiLV and pEGFP-N3 in lipoplex form (1:1.5, g/ml, in a total volume of 50 ml) into the tail vein, the blood was collected by saphenous venepuncture. Blood samples were allowed to coagulate at 4uC for 4 h and then centrifuged at 40006g for 10 min. Serum was collected, diluted with PBS and kept at 280uC until analysis. The concentrations of tumor necrosis factor (TNF)-a, interleukin (IL)-6 and IL-12 were determined using enzyme-linked immunosorbent assay (ELISA) kits.

Immunoblot Analysis
The immunoblot assays were performed as described previously [18]. In brief, cells were lysed in buffer (1% Nonidet P-40, 20 mM Tris-HCl (pH 7.6), 0.15 M NaCl, 3 mM EDTA, 3 mM ethylene glycol tetraacetic acid (EGTA), 1 mM phenylmethylsulfonyl fluoride, 2 mM sodium vanadate, 20 mg/ml aprotinin, and 5 mg/ml leupeptin). The lysates were purified initially by centrifugation and denatured by boiling in Laemmli buffer, separated on 8% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and electrophoretically transferred to a nitrocellulose membrane. Following blocking with 5% non-fat milk at room temperature for 2 h, the membrane was incubated with the patient's EBV-TK serum at 1:1000 dilution overnight at 4uC. Membranes were then incubated with a 1:5000 dilution of horseradish peroxidase conjugated secondary antibodies for 1 h at room temperature, and detected with the Western Lightning Chemiluminescent detection reagent (Perkin-Elmer Life Sciences, Wellesley, MA).

In vivo Treatment Efficacy of MiLV
To evaluate the in vivo treatment efficacy of MiLV, 1610 7 B95-8 cells transfected with equal mole of MEKK1-MiLV or pCMV/ MEKK1 plasmid were inoculated subcutaneously into both flanks of 10-week-old male BALB/c nude mice (8 mice for each treatment group). When tumors had become palpable (7-10 days later), they were treated with a single intraperitoneal injection of NaB (500 ml of 50 mM sodium butyrate in PBS) and intraperitoneal injection of GCV (100 mg/kg twice a day for 5 days). Tumor size was monitored by measuring the length and width with calipers, and volumes were calculated with the formula: (L6W 2 ) 60.5, where L is length and W is width of each tumor. When tumors became extremely large (greater than 1 cm 3 ) or the mice appeared ill, mice were sacrificed by cervical dislocation, and the tumors were excised and weighed.

Statistical Analysis
Statistical comparisons of differences between treatments were made by use of the paired t test. A p-value of ,0.05 was considered to be statistically significant. The statistical analyses were performed using SPSS 17.0 for Windows.

Characteristics of MiLV
Compared with its progenitor pEGFP-N3 plasmid (4.7 kb) which contains antibiotic makers and other bacterial originated genes, the eGFP-MiLV (1.7 kb) is only about one third the size (Figure 1). MiLV and PCR fragment were incubated with Exonuclease III to investigate the resistance of MiLV to in vitro degradation. The results indicated that the half-life of eGFP-MiLV was 10 to 15 folds greater than the DNA fragment alone (data not shown). More than 85% of MiLV were resistant against exonuclease digestion after 2 h. MTT assay showed that the viability of HEK 293 cells transfected with eGFP-MiLV (95.860.70%) were significantly (p,0.05) higher than that transfected with the pEGFP-N3 plasmid (92.860.67%). There was no significant (p.0.05) difference between cytostatic effects of MiLV or plasmid (data not shown).

In vivo GFP Expression of MiLV and Plasmid
To assess in vivo expression of the GFP gene, 20 mg eGFP-MiLV or 60 mg pEGFP-N3 plasmid (equal molar) was injected into a hind leg of mice. After durations of 1, 2, 4 or 8 weeks after intramuscular injection, at least one mouse of each group was euthanized. Green fluorescence was detected in muscle of mice more than two months after injection of eGFP-MiLV. Maximum fluorescence was observed 2-4 weeks after injection ( Figure 4). However, there was just limited fluorescence observed 4 weeks after injection with the pEGFP-N3 plasmid. The results suggested that MiLV is a more stable vector than plasmid for in vivo gene transfection. Transfection efficiency was quantified by measuring the fluorescence in muscle 2 weeks after injection. Fluorescence intensity in leg muscle of mice injected with eGFP-MiLV was significantly greater (3.2 fold, p,0.05, n = 3 for each group) than in the muscle of mice injected with the pEGFP-N3 plasmid.
To compare the immune responses to encoded antigen of plasmid and MiLV, the eGFP-MiLV and pEGFP-N3 plasmid were precipitated on gold particles and coated onto tefzel-tubings before gene gun immunization. Antibodies directly against the encoded antigen (GFP) were examined by ELISA in mouse sera taken at the indicated time-points after the primary immunization. Both the MiLV and plasmid group showed a positive antibody response towards the GFP antigen in serum samples taken 2 weeks after the first immunization. Mice immunized with MiLV showed a significantly (p,0.05) higher antibody response after the second immunization compared with the plasmid group ( Figure 5). After four times immunization, the immune response of plasmid group is only about 38% of the MiLV group. It suggested that the gene delivery and expression efficiency of equal weight MiLV is significantly greater than that of plasmid.

In vivo Inflammatory Response
The immunostimulatory activities of the eGFP-MiLV and pEGFP-N3 formulations were tested by determining serum TNFa, IL-6 and IL-12 levels after injection (40 mg DNA) for 2 h. Figure 6 shows the levels of TNF-a, IL-6 and IL-12 levels in blood at 2 h after injection of eGFP-MiLV and pEGFP-N3. The levels of TNF-a, IL-6 and IL-12 in pEGFP-N3 group were 1.5-, 1.4-, and 1.2-fold higher than that in eGFP-MiLV group, respectively. The results revealed that MiLV, which contains much less CpG motifs than plasmid, reduces the in vivo inflammatory responses to gene delivery vector.

Case Study: MEKK1-MiLV to EBV Positive Cells and Tumor
Because the results of the in vitro and in vivo proof of concept studies suggested that MiLV had good potential application prospects at gene therapy, the MEKK1 gene, which can elevate the sensitivity of B95-8 cells to NaB/GCV and the expression of thymidine kinase (TK) gene [18], was chosen to construct MEKK1-MiLV (Figure 7 A). The pLMP1 promoter was included to ensure that MEKK1 would only be expressed in cells that are EBV positive. The immunoblot results revealed that MEKK1 delivered by MiLV or plasmid could significantly enhance the TK expression in B95-8 cells (Figure 7 B), which was consistent with our previous study [18]. The results indicated that MEKK1 transfection efficiency of MEKK1-MiLV was at least comparable with pCMV/MEKK1 plasmid.
After being transfected with either MEKK1-MiLV or MEKK1 plasmid for 24 h, B95-8 cells were treated with 1 mM NaB for 18 h followed by exposure to 100 mg/ml GCV for 3 days. Nuclei of cells transfected with the MEKK1 plasmid or MEKK1-MiLV became smaller, elongated and fusiform (Figure 8 A). The relative viability of MEKK1-MiLV transfected cells (Figure 8 A, MiLV group ) was less than MEKK1 plasmid transfected cells (Figure 8 A, Plasmid group). This result was confirmed by the results of the MTT assay. B95-8 cells transfected with MEKK1-MiLV were significantly (p,0.05, paired t test) more sensitive than cells transfected with the plasmid to co-exposure to GCV/NaB for 3 or 6 days (Figure 8  B). Thus, it was concluded that transfection with MEKK1 could enhance the sensitivity of EBV positive cells to GCV/NaB, particularly when MEKK1 was transferred by MiLV.
We tested the effects of MEKK1 gene delivered by MiLV or plasmid on established nasopharyngeal carcinoma growth in mice. As shown in Figure 9, tumor cells transfected with MEKK1-MiLV or pCMV/MEKK1 plasmid both significantly suppressed the growth of B95-8 subcutaneous tumors when compared with that of control (not transfected) (p,0.01), suggesting that MEKK1 gene could enhance the in vivo sensitivity of EBV positive tumor cells to GCV/NaB. Furthermore, the tumor volumes were significantly (p,0.05) reduced in the group treated with MEKK1-MiLV compared with the group treated with pCMV/MEKK1 plasmid. These results suggested that the MEKK1-MiLV has a more favorable antitumor effect than plasmid in vivo.

Discussion
The design and optimization of the expression system is a major part of the development of successful gene therapy [19]. One reasonable approach to enhance the efficiency of transfection is to remove the non therapeutic genes from the plasmid, especially immunostimulatory CpG motifs that originate in bacteria [20]. In the present study, the ligation product was used directly as the template for PCR amplification. The vector was duplicated at the end of each PCR cycle. After the vector was purified using a PCR cleanup kit and ensured by DNA sequencing, it could be used for in vitro and in vivo experiments. A possible concern about using   PCR amplification for MiLV is that errors occur during amplification. Two or more polymerases (such as Taq and Pyrobest polymerases) with different fidelities can reduce these errors. Furthermore, DNA sequencing after PCR amplification was used to ensure that there was no mutation for DNA-MiLV transfection. The PCR-amplified MiLV has several significant advantages over plasmid DNA and other similar gene delivery vectors.
The MiLV is safer than plasmids and other vectors. The MiLV reduces numbers of inflammatory unmethylated CpG motifs which are contained in the skeleton of plasmid. Unmethylated CpG dinucleotides, or CpG motifs, which are uncommon in mammalian DNA, stimulate immune cells through Toll-like receptor 9 (TLR9). This recognition results in the production of pro-inflammatory cytokines, especially when DNA is administered as a DNA/cationic liposome complex [21,22]. Our results revealed that in vivo inflammatory response levels (levels of TNF-a, IL-6 and IL-12) of pEGFP-N3 group were significantly higher than in the eGFP-MiLV group, which elevates the in vivo gene delivery safety of MiLV. In addition, the PCR-amplified MiLV can avoid contaminations of bacterial origin during plasmid extraction [15]. Minicircle DNA has been approved to be an efficient DNA vector which is a double-stranded circular DNA with reduced size [7,23]. The in vitro and in vivo efficiency of gene transfer for MiLV and minicircle DNA were not compared in the present study. However, the MiLV could avoid contaminations of the bacterial originated endotoxin such as LPS which might be involved during the production processes of minicircle DNA [5,24]. Furthermore, MiLV might reduce the chances of chromosomal integration into mammalian genomes than that of plasmid DNA, which may cause toxic adverse effects [25]. The results of previous studies have shown that transcriptionally active linear DNA fragments were not well integrated into the genome and remained predominately extrachromosomal in mammalian organs [26,27]. It is reasonable to assume that genes delivered by MiLV are less likely to be integrated into the genome because the expression cassette flanked by two caps is linear fragment. Therefore, adverse effects of MiLV being integrated into the genome might be less than those caused by use of the plasmid.
MiLV improves the gene delivery efficiency. The results of in vitro experiments revealed that efficiencies of transfection of the eGFP gene delivered to eukaryotic cell lines such as HEK 293, NIH 3T3 and CNE2 by MiLV were at least comparable or greater than the plasmid. In order to efficiently transfect cells, gene delivery vectors have to pass through nuclepores; which favor small and actively transported molecules [6,28]. Previous studies indicated that DNA fragments greater than 1 kb remain in the cytoplasm rather than entering the nucleus [29]. Therefore, larger DNA molecules have less opportunity to enter the nucleus. The improvement of gene transfection efficiency by MiLV might be due to the smaller size. Furthermore, pro-inflammatory cytokines, which are stimulated by unmethylated CpG motifs as mentioned above, have been reported to reduce the transgene expression in later periods of gene transfer [30,31]. MiLV, which reduces the in vivo inflammatory response as compared to plasmid, would elevate the gene transfection efficiency. In addition, the CpG motifs can be the target for methylation by DNA methyltransferases. About 70% of CpG sites in the CMV enhancer were methylated at days 7 after intramuscular injection of adenoviral vectors [32]. Methylation is a major mechanism responsible for the reduced gene expression in eukaryotic cells [33]. One recent study revealed that the deletion of CpG motifs in plasmid improved the duration of in vivo transgene expression when administered as a DNA/ polymer complex [34]. Collectively, the greater in vitro and in vivo gene delivery efficiency of the MiLV than plasmid in the present study might be due to the reduction of bacterial CpG motifs as well as its small size.
Stability is another important factor affecting expression of transgenes. In the present study, new ODN caps were designed according to the D loop of tRNA in eukaryotic cells. The results of in vitro digestion experiments revealed that the cap could protect the vector from exonuclease effectively (85% of MiLV were resistant against exonuclease digestion for 2 h). Furthermore, the duration of GFP expression in vitro or in vivo that had been transfected by use of MiLV was significantly greater than that transfected by use of the plasmid. Fluorescence was quenched after 4 weeks in mice transfected by using the plasmid; while in mice transfected with MiLV, fluorescence lasted for more than 2 months. In addition to small molecules enhancing the transfection efficiency of MiLV, another reason for lower stability of plasmid is that the presence of CpG motifs, which triggers the induction of inflammatory cytokines upon administration to animals [35]. This feature is a drawback for the sustained expression of transgenes incorporated using plasmid. In the case of the MiLV, the continuous expression of the foreign genes might be due to the fact that almost all nontherapeutic sequences have been removed. Therefore, mice immunized with MiLV showed a significantly (p,0.05) lower pro-inflammatory response compared with the plasmid group. Further studies will be needed to demonstrate more detailed reasons for the prolonged transgene expression of MiLV.
EBV is a ubiquitous human herpes virus that is associated with variety of human malignancies, including nasopharyngeal carcinoma (NPC), Burkitt's lymphoma (BLs), T cell lymphoma and gastric carcinoma [36]. Nearly 100% of NPCs and 90% of BLs contain EBV episomes [37]. Our previous study indicated that the constitutive activation of MEKK1 can increase the sensitivity of EBV positive cells to GCV/NaB via a TK-dependent mechanism [18]. In the present study, the cells transfected with MEKK1-MiLV were more sensitive to GCV/NaB than that transfected with plasmid. Furthermore, our results showed that tumor cells transfected with MEKK1-MiLV or pCMV/MEKK1 plasmid had significantly smaller B95-8 subcutaneous tumors when compared with that of the control. The in vivo antitumor efficacy of MEKK1-MiLV was significantly greater than its corresponding plasmid. These results also suggested that interfering with the MEKK1 signaling pathway may be a useful therapeutic strategy to enhancing the sensitivity of EBV-positive tumor cells to GCV/ NaB.
In summary, this study provides proof of the efficacy of a safer gene delivery vector with satisfactory transfection efficiency both in vitro and in vivo. This PCR generated vector does not require bacteria for production. Therefore, it removes the possibility of LPS contamination during plasmid preparation. These advantages