Background
Glioblastoma multiforme (GBM) accounts for more than 70 % of all primary central nervous system neoplasms in adults [
1]. Despite advances in surgery, chemotherapy, and radiotherapy, the life expectancy of patients with GBM is still less than 1 year [
2]. The failure of current therapeutic approaches to treat GBM is attributed to the high proliferative and infiltrative nature of these neoplasms [
3]. Malignant cells are often seen surrounding the neurons and blood vessels and migrate through the white matter tracts to regions distant from the original tumor mass, thus the incidence for tumor recurrence is high. Herein, we explored the combination treatment of pre-established human glioma xenograft using multiple therapeutic genes whereby the gene expression is regulated by the status of cellular proliferation of the cancer cells.
We have previously constructed a Herpes Simplex Virus type 1 (HSV-1)-based amplicon vector in which the activation of the transgene expression is regulated by a G
0/G
1-specific transcriptional repressor protein termed cell cycle-dependent factor 1, CDF-1 [
4]. CDF-1 repressor protein binds to the CDE/CHR regulatory region located within the
cyclin A promoter. In quiescent cells, the transactivation of
cyclin A promoter could not take place due to the binding of the CDF-1 repressor protein onto the
cyclin A promoter. However, in actively proliferating cells, transcription of the luciferase reporter [
4,
5] or therapeutic gene [
6] is activated due to the absence of the CDF-1 repressor protein. As a proof-of-concept, we have chosen the
Fas ligand (
FasL) and the
Fas-associated death domain (
FADD/MORT1) as therapeutic genes because it is important that the effects derived from these genes should not mask the cell cycle-dependent function of the amplicon viral vectors.
Fas ligand/APO-1L (CD95L) is a ~ 40 kDa type II membrane protein belonging to the tumor necrosis factor (TNF) family. Full length FasL can be further processed to release a functional soluble 26 kDa molecule known as soluble FasL [
7]. Binding of FasL to its receptor Fas triggers the trimerization of the Fas receptors and initiates the recruitment of the cytoplasmic adaptor protein FADD through the interaction of the death domains [
8]. Recruited FADD then interacts with procaspase-8 via the death effector domain to form the death-inducing signaling complex (DISC). The close proximity of caspase-8 zymogens facilitates their autocatalytic cleavage, which subsequently trigger the downstream effector caspases resulting in apoptosis [
7,
9]. Both Fas and FasL expression are absent in normal astrocytes; however, the expression of Fas, but not FasL, in astrocytomas appear to correlate with neoplasm grade [
10‐
12]. Based on these findings, the Fas/FasL receptor system has been proposed as specific target for human brain tumor therapy. This contradicts another school of thought where the Fas/FasL receptor interaction grant the tumor cell an immune-privileged status, supported by studies demonstrating that the FasL expression in cancer cells deliver death signals to activated Fas-positive T lymphocytes [
13‐
15]. Aside from the possible role in immune surveillance, some of the glioma cells are resistant to Fas-induced apoptosis [
16,
17], possibly due to low levels of Fas expression [
16,
18], or absence of FADD [
19] or caspase 8 expression [
20]. Alternatively, epigenetic aberrations can select for glioma cells that possess several resistance mechanisms to conventional therapies [
21]. Interestingly, the overexpression of caspase 8 or FADD has been demonstrated to rescue the defect and rendered the cells sensitive to FasL-induced apoptosis [
19,
22]. Recently, inducible FADD was also shown to induce apoptosis in resistant glioma cells [
18].
Since the Fas/FasL receptor pathway converges at FADD, we hypothesized that the overexpression of FADD could sensitize glioma cells to FasL-induced apoptosis. In view of the multifaceted roles of FasL and FADD in keeping the homeostasis of immune cells, these genes were inserted into a previously generated cell-cycle regulatable HSV-1 amplicon vector under a glial cell-specific GFAP promoter. We demonstrated that the newly generated therapeutic vectors are capable of inducing cell death in proliferating primary human glioma cells derived from patients, suggesting that these vectors are functional in a clinical scenario. Furthermore, these vectors are stable, elicit minimal immune response, and are not significantly hampered by chemotherapy or irradiation in vivo. More importantly, we showed that the co-expression of FasL and FADD could elicit potent anti-tumor effect, which was enhanced in the presence of temozolomide, resulting in prolonged survival of mice bearing orthotopic gliomas. Taken together, our results demonstrated that the glial-specific, cell cycle-regulatable HSV-1 amplicon viral vectors may prove useful in enhancing the efficacy of glioma treatment.
Discussion
GBM have retained their dismal prognosis despite advances in neurosurgical techniques, radiation and drug therapies. Some of the difficulties encountered include inaccessibility to resective surgery because of anatomical location and tumor recurrences. Based on a model that predict the number of tumor cells distributed around the primary tumor bed, the percentage of tumor cells found at a distance more than 2 cm from the tumor edge is at ~2 % prior to surgery and increased to ~ 23 % post-surgical resection [
28]. Thus, a strategy that can effectively target the highly proliferating tumor cells is urgently in need. We have previously generated a HSV-1 amplicon viral vectors whereby the transgene expression is regulated by cell proliferation [
4,
6]. The present study aimed to explore the clinical feasibility of this vector in the treatment of human brain tumors by placing the
FasL or
FADD genes under the regulation of a glial cell-specific promoter. We demonstrated the therapeutic efficacy of these vectors in primary cultures of human brain tumors, and showed their ability to mediate cell-type specific transgene expression
in vivo. More importantly, the efficacy of these therapeutic viruses was greatly enhanced by TMZ, resulting in prolonged survival of glioma-bearing mice.
The ligands of the TNF family (e.g, FasL, TRAIL and TNF-α) and the members of the corresponding TNF receptor superfamily are known to exhibit pleiotrophic activities in mammalian cells. They can induce cellular proliferation, differentiation or cell death depending on the responding cell type and the microenvironment [
7], for e.g., Fas/FasL interaction has been shown to be involved in neurogenesis [
29]; the Fas/FasL system was also demonstrated to confer immune privileged status to tumor cells due to the expression of FasL on the tumor cells and the tumor endothelium [
30,
31], which induces cell death in the Fas-expressing T cells. However, how this process is regulated is still unknown. Despite these issues, several groups have generated recombinant viral vectors that deliver the
FasL gene to eradicate glioma cells and have demonstrated prolongation of survival [
32,
33]. However, high level of FasL expression has been shown to induce liver failure [
34]. Thus, restricting the FasL expression to tumor cells is essential if FasL is to be employed for cancer therapy. We have chosen
FasL as our therapeutic gene because in our cell cycle-dependent transgene activation system, the therapeutic effect cannot be overwhelming as this could potentially mask the cell cycle-regulatory property of the vectors. To circumvent the possible complications of the immune system, we have chosen the immunodeficient nude mice as our mouse model; hence, the
FasL-induced apoptosis could still serve as a good tool to assess the therapeutic efficacy of our dual-function viral vectors.
Many clinical trials in human brain tumors are conducted by injecting recombinant viral vectors into the tumor cavity margins following surgical resection [
35]. It is therefore important that (i) the tumor cells are susceptible to viral infection; (ii) the viruses are stable without causing adverse cytotoxic effects; and (iii) the transgene expression is restricted to only tumor cells. We have demonstrated that transduction efficiency of pG8-18 amplicon viruses was relatively high in proliferating GFAP-positive primary human glioma cells. However, the transduction efficiency of HSV-1 amplicon viral vectors has been reported to vary in different primary glioma cell cultures, possibly due to the heterogeneity of the glioma cells and the variation in the cell surface receptors required for viral entry [
36]. Thus, it may be necessary to pre-examine the efficiency of infection on a patient's tumor sample. We have also demonstrated that the amplicon viruses are relatively stable (Figure
3D) although the transgene expression mediated by these vectors maybe transient due to the increasing tumor cells to vector ratio. The
GFAP enhancer sequence has been shown to confer glial-cell specificity to T98 [
37], ΔGli36, U251 and SF767 [
4]. Because the
GFAP enhancer elements drive FasL and FADD expression specifically in glial cells, the packaging efficiency of the virions was unaffected (data not shown). Furthermore, FasL expression mediated by pG8-
FasL vectors was higher in proliferating versus growth arrested ΔGli36 human glioma cells (Figure
2B), which correlated with the enhanced apoptosis observed in the proliferating GFAP-positive ΔGli36 cells (Figure
2A). By contrast, FasL expression did not differ in the proliferating HeLa cells versus the G
1-arrested cells (Figure
1D), indicating that the transgene expression mediated by pG8-
FasL is regulated by type of cells under proliferating conditions. This is further supported by similar finding
in vivo (Figure
1E).
One of the major obstacles encountered in targeting death receptors in tumor cells is that the cells are usually resistant to apoptosis induced by death receptor ligands [
17,
38]. A recent report has shown that CD133-positive cells isolated from human glioma cells are also resistant to
Fas-induced apoptosis [
39]. Likewise, cells derived from human patients have been shown to be resistant to etoposide, paclitaxel, TMZ and carboplatin [
40]. These findings suggest that the immature stem cells in glioma could be an important factor of resistance to
Fas signaling pathway. Thus, enhanced therapeutic efficacy is much desired. The co-expression of FADD and caspase-8 are reported to be required for the synergistic cytotoxicity induced by combined IR/TRAIL treatment [
41]. As such, we explored whether the therapeutic efficacy of pG8-
FasL could be improved in the presence of
FADD. Our results showed that the co-expression of FasL and FADD in primary glioma cells enhanced apoptosis by 20%
in vitro (Figure
2D) and prolonged the survival of intracranial glioma bearing mice (Figure
3A and
3B). However, the therapeutic efficacy varies depending on whether the viruses were used to infect human glioma cells prior to tumor cell implantation (Paradigm 1) or after the establishment of the tumor mass (Paradigm 2). In both scenarios, significantly prolonged survival was observed in mice treated with amplicon viruses albeit paradigm2 was substantially less effective. The latter was attributed to the limited vector spread and mode of vector delivery but not due to the instability of the amplicon virions (Figure
3A, B and
3D) or possible immunocytotoxicity elicited by the vectors (Figure
3E). In fact, Suzuki et al. has reported the persistent transgene expression conferred by HSV-1 amplicon vectors in the brains of immunocompetent C57BL/6 mice (up to 385 days post viral injection) through the incorporation of the Epstein Barr Virus (EBV) episomal elements [
42].
TMZ in combination with IR are currently the first-line treatment for recurrent GBM and when used concurrently, have been shown to improve the median survival time of glioma patients for up to 5 years of follow-up [
43]. With that in mind, we investigated whether TMZ and IR can improve the overall cell death induced by pG8-
FasL and pG8-
FADD amplicon viruses. Indeed, combination treatment of ΔGli36 cells with pG8-
FasL/FADD amplicon viral vectors, TMZ and IR markedly enhanced the percentage of cell death by ~ 40 % (Figure
4C). We further challenged the effectiveness of TMZ and IR
in vivo (Additional file
3). This time, the suppressive effect of the therapeutic amplicon viruses was not as remarkable as shown previously (Figure
1E), possibly due to the lower dose of viruses used, a different derivative of glioma cells used and a different strain of mice (SCID mice versus nude mice). Despite these variable parameters, pG8-
FasL/FADD amplicon viruses can still mediate a suppressive effect on the tumor growth. Irradiation, however, did not significantly enhance the overall therapeutic efficacy mediated by the pG8-
FasL/FADD viruses (Additional file
3). This is similar to a report by Yamini et al in that IR alone with adenovirus-delivered tumor necrosis factor (TNF) did not improve the survival of glioma-bearing mice [
44]. Since the concomitant and adjuvant dosage of TMZ and IR with pG8-
FasL/
FADD is difficult to manipulate
in vivo, we decide to focus our study on the effect of TMZ and pG8-
FasL/
FADD in mice bearing intracranial gliomas, which are of more clinical relevance. Our results showed that adjuvant TMZ boosted the therapeutic efficacy of pG8-
FasL/
FADD; the survival time was markedly prolonged in comparison to mice receiving either TMZ or pG8-
FasL/
FADD viruses alone (Figure
5).
The effectiveness of TMZ is largely determined by the status and expression level of the O
6-methylguanine-DNA methyltransferase (MGMT) [
45]. Silencing of the
MGMT promoter has been shown to confer therapeutic benefits by inhibiting DNA repair upon DNA damage induced by TMZ [
46]. Moreover, Hegi et al showed that the patients with methylated
MGMT promoter has better survival than those without after TMZ and IR treatment [
47]. Therefore, we speculated that an even greater therapeutic efficacy of pG8-
FasL/FADD and TMZ could be achieved in human glioma cells with generally low MGMT activity. Alternatively, a greater effect may be seen by increasing the viral dosages, or using a more potent therapeutic gene such as the
caspase-8 or bacterial exotoxin. Caspase-8 is frequently lost or silenced in human gliomas [
20]. Inducible caspase-8 has been shown to be effective in prostate cancer gene therapy [
48] and malignant brain tumors [
49]. A fusion protein consisting of interleukin 13 (IL-13) and a mutated form of
Pseudomonas exotoxin (IL-13-PE) has also been shown to induce potent and specific cytotoxicity in glioma cells that overexpresses the receptor for IL-13, IL13 receptor-α2 (IL13-Rα2) [
50]. Since our cell cycle-regulatable HSV-1 amplicon viral vectors have been shown to confer relatively tight regulation of gene expression, it will be interesting to study the potential efficacy in these settings.
From a clinical application point of view, there are two ways one could use these vectors as gene delivery vehicles. They can either be used directly to infect cells surrounding the margins of tumor resection followed by adjuvant/concurrent treatment with TMZ and/or IR, or to infect
ex-vivo cultured adult human mesenchymal stem cells (MSC), which has been shown to be resistant to chemotherapy drugs such as cisplatin, vincristine, and etoposide [
51] and IR [
52]. Although the latter strategy needs to be stringently evaluated, the inherent tumor tracking properties of MSC is extremely attractive especially since the incidence of metastatic brain tumors with high proliferative potential is predicted to increase [
53]. We have performed independent studies to show that HSV-1 amplicon viral vectors can infect MSC efficiently without affecting the cellular proliferation, tumor homing and multilineage differentiation potential of MSC [
54]. Thus, further studies are required to couple the homing potential of MSC with the cell cycle-regulatable HSV-1 amplicon vectors.
In summary, we have demonstrated the therapeutic efficacies of pG8-FasL/FADD amplicon viruses in human glioma cells derived from established cell lines and patients biopsy samples. The vectors are relatively stable with minimal cytotoxicity and remained functional in the presence of chemotherapy and ionizing radiation treatment. More importantly, combined treatments of these therapeutic viruses with TMZ prolonged the survival of intracranial glioma-bearing mice. Given that gliomas are heterogeneous in nature, the combination of TMZ and our cell cycle-regulated FasL and FADD vector should confer added survival benefits.
Materials and methods
Isolation of primary human glioma cells
This study has been approved by the SingHealth Centralized Institutional Review Board, Singapore. Primary human glioma cells were isolated, after informed consent, from the brain tumor tissues of patients undergoing brain tumor surgery at the National Neuroscience Institute, Singapore. The harvested tissue was separated into small pieces in the presence of complete medium (Astrocyte Basal Medium (ABM) supplemented with 10% FBS, Penicillin/Streptomycin, normocin and L-Glucose; Cambrex Bio Science Walkersville, Inc., Walkersville, MD). The tissue suspensions were first passed through a 5 ml serological pipette, followed by a 1 ml pipette and finally a flame-polished pasteur pipette until no clumps were visible. Following trypsin digestion, the homogenate was filtered through a 70-μm cell strainer (BD Biosciences, San Jose CA), and then subjected to centrifugation. The collected cells were cultured in complete ABM. All cells were maintained at 37°C in a humidified incubator with 5% CO
2. The culture of ΔGli36 and HeLa cells was performed as described previously [
6].
Plasmid constructions
The construction of pG8-18, pIH8Gal
Luc, pC8-36, pC8-
FasL, and pIH8Gal
FasL plasmid were described previously [
4,
6]. To generate pG8-
FasL, the entire DNA fragment encoding the
Gal4/NF-YA fusion protein and the 8Gal
FasL region from pC8-
FasL vector [
6] was excised using
Pme I and inserted into the same restriction enzyme site on pG8-18. A similar subcloning strategy was used for the construction of pG8-
FADD from pC8-
FADD[
6]. All plasmids were amplified in
E. coli STBL-2 (Invitrogen, Grand Island, NY) and the DNA was extracted using a QIAprep Spin Miniprep kit (Qiagen GmbH, Hilden, Germany) and verified by DNA sequencing (Applied Biosystem Inc., USA).
Synchronization of cells for cell cycle analysis
Synchronization of cells in the G
1 phase of the cell cycle was performed by treating the cells with 40-60 μM of lovastatin (Merck, Singapore) in the presence of 0.1% FBS for 48 h. Cell cycle analysis was performed as described previously [
4].
Packaging of helper virus free HSV-1 amplicon viral vectors
Packaging of the HSV-1 amplicon vector was performed as described previously using the helper virus-free packaging method [
55]. The titer obtained for the resulting packaged amplicon viral vectors ranged from 1×10
7 to 1×10
8 TU/ml after concentration through a sucrose gradient. Infection of viral vectors on ΔGli36 and HeLa cells were performed at a multiplicity of infection (MOI) of 1.0 and the transduction efficiency was determined by flow cytometry for the presence of eGFP+ cells.
Immunohistochemistry, immunofluorescence and TUNEL staining
Immunohistochemistry and immunofluorescence staining were performed as previously described [
4]. Antibodies (GFAP, Ki67, CD4, CD8, and CD11b) were purchased from BD Biosciences and used at 2 μg/ml concentration. TUNEL staining was performed using the In situ cell death detection kit (Roche) according to manufacturer's instruction. Briefly, fixed cells were permeabilized with 0.1% triton in 0.1% sodium citrate solution prior to incubating in solution containing the terminal deoxynucleotidyl transferase enzyme and nucleotide mixture. Staining of cells was carried out at 37°C for 1 h. After which, non-specific staining was removed by rinsing the cells in PBS twice. Samples were then visualized under fluorescence microscope. All images were either acquired on the CCD digital camera (Olympus DP11, Olympus, Japan) mounted on the upright microscope (Olympus BX41) or the Nikon TE300 Eclipse fluorescence microscope (Nikon, Japan).
Treatment with TMZ and γ-irradiation
Temozolomide (Temodal; Schering Plough, Belgium) was dissolved in DMSO (Sigma Aldrich) to produce a 100 mM stock solution for
in vitro experiments. For
in vitro experiments, TMZ was diluted with PBS to obtaine 75 μM solutions. For
in vivo experiments, stock solution was diluted in PBS to a final concentration of 5 mg/ml. A dose of 5 mg/kg body weight was used, which is equivalent to half of the recommended dosage of 25 mg/kg/m
2 in adult humans [
56].
To assess the effect of TMZ treatment on luciferase gene expression, cells were first infected with the respective amplicon viral vectors for 6 h. The transduced cells were divided into two groups, one portion of the cells were cultured in complete medium, while the other portion was treated with 75 μM of TMZ at 37°C. After 1 h of treatment, the cells were rinsed twice with PBS and replenished with complete medium containing 10% serum. To assess the effect of IR, similar procedure was performed as treatment with TMZ, except that the cells were transfected with pG8-18 or pIH8GalLuc plasmid. Transfected cells were exposed to 3 Gy of γ-irradiation, followed by incubating the cells in either fresh complete medium or medium containing lovastatin. Luciferase activities were measured after 48 h of transfection.
To assess the effect of TMZ and IR on pG8-FasL/FADD-mediated cell death, ΔGli36 cells was infected with MOI of 1.0 of pG8-FasL and pG8-FADD or pG8-18 amplicon viruses. The viral supernatant was removed after 6 h and cells were cultured in medium containing 75 μM of TMZ. The cells were subsequently subjected to 3Gy of IR treatment. TMZ-containing medium was removed after 1 h of incubation. The percentage of cell death was assessed after 72 h by trypan blue exclusion assay.
Animal Experiments
All animal experiments were performed according to the guidelines and protocols approved by the SingHealth Institutional Animal Care and Use Committee, Singapore.
To determine the efficacy of FasL
in vivo, 6-8 weeks old CB-17 SCID mice (Animal Resource Centre, Australia), inoculated with either HeLa or ΔGli36-SCID8 cells (2 × 10
6) at their right flank, were divided into 4 groups. One day following tumor inoculation, 2 × 10
6 TU of viral vector was administered intratumorally. Injections of viral vectors were repeated every 10 days until tumor necrosis was observed in the non-treated groups. The tumor volume was measured and calculated according to the formula
volume = 0.52 × length × width
2
. At the end of the experimental period, all animals were sacrificed and tumor nodules were harvested. Analysis of FasL expression was also performed as described previously [
6].
To determine the synergistic effect of FasL and FADD in intracranial tumor-bearing mice, two experimental paradigms were designed. In Paradigm 1, ΔGli36 human glioma cells (1×106) were pre-infected with equal ratios (5×105 TU each) of pG8-FasL and pG8-FADD amplicon viral vectors followed by implantation into the right hemisphere (bregma (0,0) lateral 2.0 mm and depth 2.5 mm) of immunodeficient nude mice on the next day. For Paradigm 2, mice were inoculated with ΔGli36 intracranially followed by inoculation of amplicon viral vectors (MOI of 1.0) intratumorally 7 days later. Mice were monitored weekly for changes in body weight.
For investigating the combined therapeutic effect of TMZ with pG8-FasL and pG8-FADD in vivo, immuno-incompetent nude mice were first inoculated with 2×105 ΔGli36 cells intracranially. Amplicon viruses, either pG8-18 (2×105 TU) or pG8-FasL and pG8-FADD (MOI of 1.0, 1×105 TU each), were injected intratumorally (i.t.) into the same co-ordinates after 1 week. TMZ was administered after 18 h and on a daily basis for 5 days (5 mg/kg which total to 25 mg/kg) via an intraperitoneal (i.p.) route.
To assess the effect of TMZ and/or IR on FasL and FADD-mediated tumor regression in a subcutaneous glioma model, mice bearing ΔGli36 human glioma xenograft (5×10
5) at their hind limbs were randomized into groups indicated in Additional File
3. One week post tumor cell implantation, pG8-
FasL and pG8-
FADD amplicon viruses (combined MOI = 1) were injected intratumorally (i.t). Treatment with TMZ, IR or TMZ and IR was initiated 18 h post-virus inoculation. TMZ was delivered i.p. at a dose of 10 mg/kg/day for 5 doses, and IR (2Gy/day) was given to the mice, which amount to a total of 10 Gy. Tumor volume was measured every 3-4 days as detailed above.
To determine the stability of the pG8-18 amplicon viral vectors
in vivo, 1×10
6 TU of pG8-18 vector was injected into the left (control) and right hemisphere of the same immunodeficient nude mice. The right hemisphere consisted of tumor mass (1 × 10
6 ΔGli36), which was implanted a week before viral injection. The brains were then harvested on days 4, 10, and 28 post-injection and prepared for extraction of viral DNA. Viral DNA was recovered from brain tissues using Hirt's method [
57] with slight modifications. Briefly, the tissues were first frozen with liquid nitrogen and ground into powder using a mortar and pestle. After that, the tissues were incubated in 500 μL of lysis buffer (0.6 % SDS, 10 mM EDTA, pH 8.0, 10 mM Tris-HCl, pH 7.5) for 20 min at room temperature followed by addition of 125 μl of 5 M NaCl and incubated at 4°C overnight. The next day, the extract was subjected to centrifugation at 13 000 × g at 4°C for 30 min. The supernatant was recovered and extracted with phenol, phenol-chloroform and chloroform. The DNA was precipitated using isopropanol, rinsed with 70% ethanol, and dissolved in 35 μl of TE (10 mM Tris-Cl pH8.0, 1 mM EDTA) buffer.
To determine the transduction efficiency of pG8-18 amplicon viral vectors, immunodeficient nude mice were separated into 2 groups and inoculated with 10 μl of pG8-18 (1 × 106 TU) viral vector. The mouse brains were harvested 1 day post-injection. On the day of harvesting, the mice were first perfused through the heart with PBS, and the brains were harvested and processed to single cell suspensions. The brains were homogenized in a 50 ml falcon tube with 12 ml of HBSS (Invitrogen) using a 5 ml serological pipette, followed by a 1 ml serological pipette and a flamed-polished Pasteur pipette, until no clumps were visible. Cells were trypsinized and incubated for 15 min at 37°C, with mixing every 5 min. The homogenates were then filtered through a 70-μm pore size nylon cell strainer (BD Biosciences), and the filtrates were subjected to centrifugation at 500 rpm for 15 min at 4°C (without brake) (Beckman Coulter). The supernatant was removed and the cell pellet resuspended in DMEM containing 10% serum. The percentage of eGFP+ cells was analyzed using FACS. For the other group of mice, the brains were fixed in 4 % PFA solution overnight, followed by 30 % sucrose for 48 h, and then sliced into 10-micron sections. The eGFP+ cells were visualized using a LSM 510 Meta confocal microscope (Carl Zeiss Microscopy, Göttingen, Germany) with the appropriate filters.
To determine the immunogenicity of the HSV-1 amplicon viruses, immunocompetent Balb/c mice (6 weeks old) was inoculated with either saline or pG8-18 viruses (1×104 TU) in the right hemisphere of the mouse brain. Brains were harvested on day 1 and 4 post-injections. On the day of harvesting, mice were perfused through the heart with PBS followed by 4% paraformaldehyde. Brains were processed and cryosectioned at 10-micron thickness. Immunohistochemical staining was performed on consecutive sections.
Statistical analysis
The data are presented throughout this study as means + standard error of the mean. The statistical significance was evaluated by an unpaired t-test, and p < 0.05 was considered significant. Kaplan-Meier survival analysis was used to calculate the percentage of survival as a function of time, and the survival curves were compared using the log-rank test.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
IH designed and executed the experiments and took part in writing the manuscript; WHN provided the primary human glioma samples together with relevant clinical information and took part in proofreading the manuscript; PL was involved in the overall design of the experiments, established collaboration, and wrote the manuscript. All authors have read and approved of the final manuscript