Imaging Herpes Simplex Virus Type 1 Amplicon Vector–Mediated Gene Expression in Human Glioma Spheroids

Abstract

Vectors derived from herpes simplex virus type 1 (HSV-1) have great potential for transducing therapeutic genes into the central nervous system; however, inefficient distribution of vector particles in vivo may limit their therapeutic potential in patients with gliomas. This study was performed to investigate the extent of HSV-1 amplicon vector–mediated gene expression in a three-dimensional glioma model of multicellular spheroids by imaging highly infectious HSV-1 virions expressing green fluorescent protein (HSV-GFP). After infection or microscopy-guided vector injection of glioma spheroids at various spheroid sizes, injection pressures and injection times, the extent of HSV-1 vector–mediated gene expression was investigated via laser scanning microscopy. Infection of spheroids with HSV-GFP demonstrated a maximal depth of vector-mediated GFP expression at 70 to 80 μm. A > 80% transduction efficiency was reached only in small spheroids with a diameter of < 150 μm. Guided vector injection into the spheroids showed transduction efficiencies ranging between < 10 and > 90%. The results demonstrated that vector-mediated gene expression in glioma spheroids was strongly dependent on the mode of vector application—injection pressure and injection time being the most important parameters. The assessment of these vector application parameters in tissue models will contribute to the development of safe and efficient gene therapy protocols for clinical application.

IN VIEW OF the high incidence and poor prognosis of patients with malignant brain tumors treated by conventional therapies, research focusing on the elucidation of tumor biology and cell cycle regulation and on the establishment of clinically valuable gene therapies seems to be of the utmost importance.^1–3 The design of effective gene therapy strategies relies on concerted research to define the genetic and pathophysiologic alterations causing the disease, to understand the biologic characteristics of the target tissue, and to develop safe vectors and application systems to achieve efficient, targeted, and regulated alteration of specific therapeutic gene expression.^4,5 The suicide gene therapy based on retroviral vector–mediated expression of herpes simplex virus type 1 thymidine kinase (HSV-1-TK) and subsequent ganciclovir application was successful in various experimental brain tumor models.^6,7 Clinical studies revealed that this approach as an adjuvant to the surgical resection of recurrent high-grade brain tumors can be performed safely, although clinical responses were observed in only a few patients with small brain tumors.⁸ The lack of therapeutic efficiency of this replication-deficient retrovirus vector system in clinical settings may be due to vector-producer cell instability but even more to the inability to distribute vector-producer cells throughout the tumor, the low transduction efficiency of retrovirions, and the heterogeneity of tumor tissue.

To circumvent some of these problems and to serve improved distribution of vector particles within tumor tissue, replication-competent oncolytic viral vectors and certain vector infusion and application methods are being used as part of a multimodal gene therapy approach.^4,9–11 As the helper virus–free herpes simplex virus type 1 (HSV-1) amplicon is a safe and efficient vector system to transduce various central nervous system–derived cells, including human gliomas,^1,12 it holds great promise as part of a multimodal virus/gene therapy approach in the future. Therefore, this study aimed to further investigate the potential of how well HSV-1 vector particles can be distributed in a three-dimensional glioma tissue by using multicellular spheroids (MCSs).

MCSs are a cell culture model well suited to studying micrometastasis, avascular regions of large tumors, and alterations in metabolites and gene expression.^13,14 Moreover, the effects of radiation, chemotherapy, or immunotherapy can be efficiently tested in MCSs.^15–17 In comparison with single tumor cells grown in suspension or as monolayers, MCSs exhibit a higher complexity and, therefore, can be better compared to the situation in a three-dimensional tissue.¹⁵ In addition, the vitality patterns of MCSs can be studied using confocal laser scanning microscopy (CLSM) and fluorescent dyes.¹⁷

The aim of this study was to compare different methods of vector application and to investigate whether certain parameters of injection of HSV-1 vector particles into glioma MCSs would improve the extent of HSV-1 vector–mediated gene expression. The MCS model of human Gli36 glioma cells was used to serve as three-dimensional tumor tissue in culture and was studied using CLSM.

Materials and Methods

Cell Culture

Human Gli36 glioma cells (provided by Dr. David Louis, Molecular Neurooncology Laboratory, Massachusetts General Hospital, Boston, MA) were grown as monolayers in Dulbecco's Modified Eagle medium (DMEM; Life Technologies, Karlsruhe, Germany) supplemented with 10% fetal bovine serum (FBS; Roche Diagnostics, Mannheim, Germany) and 100 U/mL penicillin and 100 μg/mL streptomycin (P/S; Life Technologies). African green monkey kidney 2–2 cells (kindly provided by Dr. Rozanne Sandri-Goldin, University of California Irvine) were grown in DMEM supplemented with 10% FBS, 1% P/S, and 500 μg/mL G418 (Life Technologies). Human DU-145 prostate cancer cells (provided by Dr. J. Carlsson, Uppsala, Sweden) were cultured in F10 medium (Life Technologies) supplemented with 10% FBS and 1% P/S. All cells were cultured at 37°C in a 5% CO₂/95% air atmosphere.

Spheroid Culture

MCSs were established from human Gli36 glioma and human DU-145 prostate cancer. Spheroids were grown from single cells. Cell monolayers were trypsinized with 0.2% trypsin and 0.05% ethylenediaminetetraacetic acid (INC Flow, Meckenheim, Germany) and seeded in 250 mL siliconated spinner flasks (Integra Bioscience, Fernwald, Germany) with 250 mL complete medium agitated at 40 rpm using a stirrer system (Integra Bioscience, Fernwald, Germany). Spheroids were cultured in DMEM supplemented with 10% FBS and 1% P/S, 2 mM glutamine (Life Technologies), and 100 μM β-mercaptoethanol (Sigma Aldrich, Taufkirchen, Germany) at 37°C in a 5% CO₂/95% air atmosphere. The medium was replaced daily.

HSV-1 Amplicon Plasmid Construction and Helper Virus–Free Packaging of HSV-1 Amplicons

The HSV-1 amplicon plasmid pHSV-GN (obtained from X.O. Breakefield, Neurogenetics Unit, Massachusetts General Hospital¹⁸) carries the green fluorescence protein gene (gfp) under transcriptional control of a cytomegalovirus immediate early gene promoter (CMV¹²). After packaging, the gfp-transducing HSV-1 amplicon vector particles are depicted as HSV-GFP. Helper virus–free stocks of HSV-GFP amplicons were generated as described previously.^19,20 In brief, 10 cm plates were seeded with 3.6 × 10⁶ vero 2–2 cells. Sixteen hours later, the cells were transfected with 6 μg of Pac1 digested cosmids and 1.8 μg of the respective HSV-1 amplicon plasmid using lipofectamine (Life Technologies) for 5.5 hours. Thereafter, cells were washed (serum-free medium; three times) and incubated in 10 mL packaging medium (DMEM, 6% FBS, 25 mM HEPES [pH 7.3]) at 37°C for 24 hours. Thereafter, the medium was replaced by fresh packaging medium and cells were incubated for 24 hours at 34°C and then for 24 hours at 37°C. After harvesting (72–80 hours after transfection), vector stocks were purified and concentrated. The virus pellet was resuspended in phosphate-buffered saline (PBS), and virus stocks were stored at −80°C for titration and final use. Purified vector stocks were titered (transducing units per milliliter [TU/mL]) on 2–2 and Gli36 cells by infecting confluent monolayers in 24-well plates (Falcon, Becton Dickinson, Heidelberg, Germany) and counting GFP-positive cells 24 hours later. Vector titers of 1 to 4 × 10⁸ TU/mL were used for further experiments.

GFP Fluorescence and Quantification

For determination of green fluorescence, human Gli36 cells were plated in 24-well plates (2 × 10⁵ cells/well) to form confluent monolayers and were infected with HSV-GFP at a multiplicity of infection of 1 the next day. GFP-expressing cells were examined by use of a standard fluorescence microscope (Zeiss Axiovert 135, Carl Zeiss, Heidenheim, Germany) 24 hours after infection.²⁰

Characterization of Viable Target Tissue within Glioma Spheroids

To identify viable glioma tissue and to distinguish this from necrosis, staining with Lucifer yellow and Ki-67 was performed.²¹

Lucifer Yellow

The highly charged polar tracer Lucifer yellow (Sigma, Deishofen, Germany) was used to stain lethal cells within necrotic areas.²¹ Incubation of MCSs with Lucifer yellow (20 μM) was performed for 40 minutes at room temperature. The spheroids were washed in PBS three times, placed in DMEM, and examined by CLSM (excitation by 488 nm line of an argon-ion laser, LP 515 nm emission filter set).

Ki-67

The mouse monoclonal Ki-67 antibody (Boehringer, Mannheim, Germany) recognizes a cell cycle–associated protein of 345 kDa identical to the Ki-67 antigen. The immunoreactivity of Ki-67 is confined to the nuclei of proliferating cells. The Ki-67 antigen is preferentially expressed during late G₁, S, G₂, and M phases of the cell cycle, whereas resting, noncycling cells (G₀ phase) lack Ki-67 expression. For Ki-67 staining, MCSs were fixed in ice-cold methanol/acetone (30 minutes at −20°C), washed (PBS with 0.1% Triton X10), embedded in Tissue Tek (Reichert-Jung, Heidelberg, Germany), and frozen in liquid nitrogen. Cryosections of 12 μm were performed on a Reichert-Jung 2800 Frigocut. Sections were mounted, blocked against unspecific binding of MCSs (1% BSA 60 minutes), incubated with 7.5 μg/mL Ki-67 antibody (60 minutes), washed three times, and incubated with 4.6 μg/mL Cy5-conjugated F(ab')₂ fragment goat antimouse IgG (Boehringer) (60 minutes). Ki-67-positive nuclei were visualized by CLSM using fluorescence excitation at 633 nm (helium-neon-laser) and a longpass 655 nm filter set.

Infection versus Injection of Spheroids

Spheroids were either infected with 5 × 10⁶ TU HSV-GFP amplicon vector or directly injected by use of a microinjector (Eppendorf, Hamburg, Germany) at various spheroid sizes (diameter 100–800 μm), injection pressures (50, 100, 300, 1,000 hPa), and injection times (5, 15, 45 seconds). The extent of HSV-1 amplicon vector–mediated gene expression was investigated by use of CLSM (CLSM 410 microscope, Zeiss, Jena, Germany) one day after vector application. Ten-micrometer slices were acquired starting at the surface of the spheroid for a maximal depth of 160 mm. After infection, the maximum depth of GFP expression was registered using the CLSM imaging analysis software. After injection, the slice nearest the equatorial plane was used to evaluate the extent of vector-mediated gene expression by means of a regions of interest (ROI) analysis using the CLSM imaging analysis software. The volume of transduced tumor tissue was calculated by 4/3^*π ^*r³. For each infection experiment, 15 to 20 spheroids were investigated. For each injection experiment, 3 to 6 spheroids were investigated for each parameter. Experiments were repeated three times. Targeted spheroid injections were performed by holding the spheroid by a sterile vacuum tip (diameter 100 μm) and by injecting HSV-GFP amplicon vector via the transfer tip spike (diameter 20 μm) into the center of the spheroid.

CLSM and Optical Probe Technique

MCSs were investigated under the optical control of an inverted CLSM (LSM 410, Zeiss) using a 25× objective as described previously.^17,22 The CLSM allows detection of fluorescence signals up to a depth of 180 mm in living tissue. To investigate the extent of vector-mediated expression, the optical probe technique was used. The principle of the optical probe technique is that fluorescence intensity is recorded within a selected ROI (600 μm², 40 × 40 pixels) along the radial axis starting in the center of the spheroid and moving to the periphery. The motor commands of the stepper motor of the confocal setup were set to steps of 10 μm in z-direction between two adjacent ROI. The pinhole settings of the confocal setup were chosen to give a full-width half-maximum of 8 mm. Each ROI was scanned once for 0.064 seconds. A whole z-series of 16 ROI was achieved in approximately 7 seconds, the convolution time of the stepper motor thereby being the rate-limiting step. The mean field intensity of the fluorescence signal was determined in each ROI, and the fluorescence intensity data were plotted as a function of the penetration depth of the laser beam in the tissue.

Statistics

Analysis of variance (ANOVA) was performed by Superior Performance Software, SPSS, version 10.0 for Windows (SPSS Inc., UK Ltd, Surrey, England), and for regression analysis, KaleidaGraph 3.08 (Synergy Software, Reading, PA) was used.

Results

A new three-dimensional tumor spheroid model from human Gli36 glioma cells was established to determine injection parameters for HSV-1 vector particles that lead to an improved distribution of vector-mediated gene expression. Spheroids were grown in spinner flasks for up to 22 days, and spheroid sizes were measured every day. The growth curve (Figure 1A) of human Gli36 glioma spheroids demonstrates that an exponential growth is kept for at least 18 days and that the size of spheroids derived from human Gli36 glioma cells may reach a diameter of up to 1,000 μm. In comparison, control spheroids grown from human DU-145 prostate cancer cells reach a size of up to 400 μm in diameter.

Figure 1.

Characterization of three-dimensional spheroids. A, Growth curve of human Gli36 glioma spheroids. Human Gli36 glioma spheroids were grown for 22 days in a spinner flask, and their size was recorded over time. An exponential growth was observed for at least 18 days. B, Lucifer yellow staining. Lucifer yellow was used to stain lethal cells within necrotic areas. Different age and size stages of human Gli36 glioma spheroids are shown in comparison with DU-145 prostate cancer spheroids incubated with Lucifer yellow. The critical diameter for the first appearance of necrotic areas proved to be greater than 900 μm for human Gli36 glioma spheroids and greater than 350 μm for control DU-145 spheroids. C, Ki-67 immunostaining. Ki-67 expression indicates that multicellular spheroids of various sizes consist mostly of proliferating cells up to day 4 and a rim of proliferating tumor cells at larger spheroid sizes. In comparison, Ki-67 staining is shown in a 7-day-old DU-145 prostate spheroid.

To characterize the physiologic status and viability of human Gli36 glioma spheroids, Lucifer yellow and Ki-67 staining was performed (Figure 1, B and C). The critical diameter for the first appearance of necrotic areas in Gli36 spheroids was greater than 600 μm. In comparison, signs of necrosis were found in control DU-145 prostate cancer spheroids at a diameter of > 300 μm. Proliferating tumor cells were found throughout the spheroid at a diameter of < 300 μm or at the surface for those spheroids with diameters > 300 μm.

To determine the extent of vector-mediated GFP expression after incubation of spheroids with vector particles (infection), multiple spheroids were investigated by CLSM 24 hours after infection of spheroids with 4 × 10⁶ TU HSV-GFP amplicons (24-well plate, total infection volume 200 μL). The maximal depth of vector-mediated GFP-expression was 70 to 80 μm, as measured from the surface, with maximum levels of GFP expression at 40 to 50 μm (Figure 2, A and B). These results were independent of the size of the spheroid, resulting in relatively lower transduction efficiency with increasing spheroid size (Figure 2C). A > 60% transduction efficiency was reached only in small spheroids with a radius of less than 150 μm. Owing to the lower transduction efficiency of HSV-1 vector particles into human DU-145 prostate cancer cells, only single DU-145 cells at the surface of DU-145 spheroids were GFP positive.

Figure 2.

Transduction efficiency of HSV-GFP in human Gli36 glioma spheroids. After infection of spheroids with HSV-GFP amplicons, the maximal depth of vector-mediated GFP expression is limited to superficial cell layers (70–80 μm), as measured from the surface, with maximum levels at 40 to 50 μm. A, Representative figures were obtained by CLSM. B, Quantitative data were obtained by the optical probe technique with fluorescence distribution within the spheroid corresponding to the representative figures in A. C, The spheroid size–dependent transduction efficiency is depicted with percent transduced spheroid volume plotted over the radius of spheroids. A > 60% transduction efficiency could be reached only in small spheroids with a radius of less than 150 μm.

To improve transduction efficiency in larger spheroids, guided vector injections were performed by use of a micromanipulator (Figure 3A). A spheroid was fixed by use of a vacuum tip, and the transfer tip was placed into the center of the spheroid using an automatic guidance system. Figure 3A shows examples of the guided injections of vector and control dye (methylene blue). After guided vector injection into the spheroids, injection time– and injection pressure–dependent transduction efficiencies ranged between < 10% and > 90%. The maximal transduction efficiencies were reached at injection pressures of 100 hPa for smaller spheroids (3 days) to 300 hPa for larger spheroids (7–10 days) (ANOVA; p < .05; Figure 3B). Moreover, injection time–dependent increases in vector distribution were observed at 50 hPa at all spheroid ages (3, 7, and 10 days) and at 100 hPa and 300 hPa only at larger spheroids (10 and 7 days, respectively) (Figure 3C). Using a higher pressure (1,000 hPa) failed to further increase the transduction rate, most likely owing to leakage from the injection site. A further increase in the time period for injection (over 45 seconds) in the same location of the transfer tip did not have a positive effect on the transduction rate, which may be due to saturation or obstruction of the interstitial space by vector particles. In comparison with infection, especially in larger spheroids, higher transduction efficiency could be reached only by targeted microinjection (Figure 4).

Figure 3.

Targeted microinjection of human Gli36 glioma spheroids. A, Transmission figures of an 8-day-old Gli36 spheroid. The spheroid is fixed by means of a vacuum tip (left) with a diameter of 100 μm. The transfer tip (right) with a diameter of 20 μm is directly placed into the center of the spheroid. The upper picture shows the transfer tip entering the spheroid. The middle picture was being performed during HSV-GFP vector injection with parts of the solution leaking out of the injection site. The lower picture demonstrates injection of the control dye methylene blue into a spheroid. B and C, Pressure- (B) and time- (C) dependent transduction efficiency of targeted HSV-GFP injection. Data summarize the dependency of transduction efficiency of 3-, 7-, and 10-day-old spheroids depending on application pressure (B) at various times used for injection (5, 15, and 45 seconds) with maximal transduction efficiencies at injection pressures of 100 and 300 hPa (ANOVA; p < .05) and depending on the application time (C) at various pressures used for injection (50, 100, 300, 1,000 hPa), with a linear increase in transduction volume observed at injection pressures of 50 to 300 hPa.

Figure 4.

Difference in transduction efficiency obtained by targeted application of HSV-1 amplicon vectors. Direct comparison of transduction efficiency achieved by infection (left) versus targeted injection (right) of HSV-GFP amplicon vectors in human Gli36 glioma spheroids.

Discussion

The purpose of this study was to establish a three-dimensional human glioma model in culture using the human glioma Gli36 cell line to investigate vector distribution and transduction efficiency by use of HSV-1 amplicon vectors employing different vector application methods. Injection time– and pressure–dependent increase in transduction efficiency could be achieved compared to infection. This type of study might be important for planning vector applications in animal models and in the clinical application to serve for optimal vector distribution in vivo.

With regard to spheroid characterization, the viability and proliferative activity of human Gli36 glioma spheroids could be demonstrated even in larger spheroids by Ki-67 staining.^21,23 The Gli36 spheroids provide a stable model to study targeted three-dimensional vector application. As had been shown in a previous study, Gli36 glioma spheroids reach a considerably larger size than DU145 prostate cancer spheroids.²⁴ The distribution of vector-mediated gene expression was investigated by use of the optical probe technique. This technique permits the quantitative noninvasive analysis of the distribution of HSV-1 viral particle in MCSs by means of CLSM.²³ The glioma spheroids in culture can be compared to gliomas grown in vivo with respect to their three-dimensional tumor structures, which exhibit a higher complexity than those of single tumor cells grown in suspension or monolayers. Therefore, new methods with regard to targeted application of vector particles can be tested in these three-dimensional models before taking the step into animal models or human application in vivo. It should be pointed out that not every cell line is able to form spheroids and that spheroids in culture may contain glioma cells, astrocytes, neurons, and blood vessels.^13,25

Several gene therapy studies indicate that gene transfer with regard to distribution of vector particles within solid tumors is one of the main limiting factors.^26–28 For safe and efficient gene therapy in clinical application, technology is required that allows the noninvasive monitoring of the level and distribution of vector-mediated gene expression in vivo. With the help of imaging, one could follow critical steps of the gene therapy protocol, such as vector distribution volume²⁹ and the tissue dose of vector-mediated gene expression in vivo.^20,28,29 Furthermore, the delivery and function of the therapeutic gene HSV-1-TK using HSV amplicon vectors have been demonstrated in subcutaneous gliomas in nude mice¹⁹; however, vector distribution within the tumor still needs to be improved. In recent years, the group of Bankiewicz, as discussed in reference 30, investigated and improved the use of convection-enhanced delivery (CED) as a method to deliver therapeutics (in particular liposomes) to the central nervous system. They further developed a magnetic resonance imaging–based imaging approach whereby gadolinium-loaded liposomes were coinfused with therapeutic agents³⁰ to visualize their delivery.

For the spheroid model, we could demonstrate that the efficiency of HSV-1 amplicon vector–mediated gene expression is dependent on application parameters such as time, pressure, and volume, which seem to be critical to reach efficient gene expression. Another way to improve the distribution of vector-mediated gene expression is to use either replication-conditional HSV-1 vectors^1,31 or herpes simplex virus/Epstein-Barr virus hybrid amplicon vectors containing retrovirus vector components to achieve amplification of stable transgene expression in tumors by conversion of tumor cells into retroviral producer cells.³²

In our infection studies using HSV-1 amplicons, we demonstrated that the maximal depth of vector-mediated GFP expression is 70 to 80 μm from the surface of the spheroid. The reason for the limited transduction efficiency of Gli36 spheroids is most likely due to the limited penetration of HSV-1 vector particles within the three-dimensional tumor tissue. Similar results were observed by using adenovirus vectors.²⁷ Obviously, the particle size, constitution of the extracellular space, and expression of cell surface receptors of the vector particle seem to play critical roles. Adeno-associated virus (AAV)-based vectors are very small particles (20 nm) compared to AdV (80 nm) and HSV-1 (250 nm). Hadaczek and colleagues showed that CED of AAV-based vectors resulted in a significantly improved distribution of AAV vector–mediated gene expression when compared to simple injection techniques in nonhuman primates.³³ Most importantly, blocking AAV receptors at the neuronal membrane by heparin extended vector distribution and resulted in a significantly larger area of transduction of a rat brain. Enger and colleagues used AAV- and AdV-based vectors and demonstrated the dependency of vector-mediated gene expression on the mode of vector application using CLSM.²⁷ We also demonstrated that transduction efficiency could be significantly improved by application of HSV-1 amplicon vectors using a micromanipulator. By varying the time and injection pressure for application of vector particles into the spheroids, the transduced spheroid volume could be significantly improved, especially in older or larger spheroids. However, 100% transduction efficiency could not be reached. In contrast, Betz and colleagues investigated the effects of infusion parameters, infectious titers, and virus concentration on transduction volume by gene transfer to rodent brain using recombinant adenoviral vectors and came to the conclusion that changes of infusion parameters have only limited effects on transduction efficiency.³⁴ However, other findings report the effect of variables affecting, for example, AAV CED, such as rate of infusion; optimal cannula size, type, or placement; infusion volume; and viral serotype.^35–38

The discrepant findings between different studies indicate that the lack of widespread vector distribution throughout the tumor tissue is still one of the current limitations for successful clinical application of gene therapy vectors. Therefore, various modes of vector application should be tested in three-dimensional tumor models for each specific vector, which will be applied in clinical application to serve for the determination of application parameters that lead to optimal vector distribution and gene transduction.

With regard to clinical applicability and “potential use,” we would like to point out that normal intracranial pressure is 0 to 10 mm Hg. The injection pressures used in our work (50, 100, 300, 1,000 hPa) are equivalent to 37.5, 75, 250, and 750 mm Hg, respectively. Up to now, only CED has been successfully used by our group to deliver vector particles into patients with gliomas.²⁹ Unlike local injection techniques, CED uses a pressure gradient established at the tip of an infusion catheter that literally pushes the infusate through the interstitial space.³⁹ With 50 to 100 hPa (37.5–75 mm Hg), we seem to be at the same infusion line pressures described by Bobo and colleagues in their initial studies of CED, but with higher pressures, we may be at the limit of clinical tolerance.⁴⁰

Conclusion

Our results demonstrate that the extent of HSV-1 amplicon vector–mediated gene expression in three-dimensional human Gli36 glioma spheroids is greatly dependent on the mode of vector application (infection versus injection)—injection pressure and injection time being the most important parameters. The assessment of vector application parameters in tissue models will contribute to the development of safe and efficient gene therapy and vector application protocols for clinical application. A combination of novel modes of vector delivery and improved vector systems shall increase the level of clinical relevance.

Footnotes

Acknowledgment

Financial disclosure of authors: This work was supported in part by the Center of Molecular Medicine Cologne (ZMMK***TV46), the Ministerium für Schule, Wissenschaft und Forschung NRW (MSWF516-40000299), and the 6th FW EU Grants DiMI (LSHB-CT-2005-512146) and CliniGene NoE (LSHB-CT-2006-018933).

Financial disclosure of reviewers: None reported.

References

Jacobs

Breakefield

Fraefel

. HSV-1-based vectors for gene therapy of neurological diseases and brain tumors: part II. Vector systems and applications. Neoplasia 1999;1:402–16.

Kramm

Chase

Herrlinger

. Therapeutic efficiency and safety of a second-generation replication-conditional HSV1 vector for brain tumor gene therapy. Hum Gene Ther 1997;8:2057–68.

Martuza

Malick

Markert

. Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science 1991;252:854–6.

Selznick

Shamji

Fecci

. Molecular strategies for the treatment of malignant glioma—genes, viruses, and vaccines. Neurosurg Rev 2008;31:141–55; discussion 155.

Young

. Neurological disorders: an overview. In: Chiocca

Breakefield

, editors. Gene therapy for neurological disorders and brain tumors. Totowa, NJ: Humana Press; 1998. p. 341–3.

Culver

Ram

Wallbridge

. In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumors. Science 1992;256:1550–2.

Ezzeddine

Martuza

Platika

. Selective killing of glioma cells in culture and in vivo by retrovirus transfer of the herpes simplex virus thymidine kinase gene. New Biol 1991;3:608–14.

Barzon

Zanusso

Colombo

. Clinical trials of gene therapy, virotherapy, and immunotherapy for malignant gliomas. Cancer Gene Ther 2006;13:539–54.

Hadjipanayis

Fellows-Mayle

Deluca

. Therapeutic efficacy of a herpes simplex virus with radiation or temozolomide for intracranial glioblastoma after convection-enhanced delivery. Mol Ther 2008;16:1783–8.

10.

Meijer

Maguire

LeRoy

. Controlling brain tumor growth by intraventricular administration of an AAV vector encoding IFN-beta. Cancer Gene Ther 2009;16:664–71.

11.

Tyminski

Leroy

Terada

. Brain tumor oncolysis with replication-conditional herpes simplex virus type 1 expressing the prodrug-activating genes, CYP2B1 and secreted human intestinal carboxylesterase, in combination with cyclophosphamide and irinotecan. Cancer Res 2005;65:6850–7.

12.

Rueger

Winkeler

Miletic

. Variability in infectivity of primary cell cultures of human brain tumors with HSV-1 amplicon vectors. Gene Ther 2005;12:588–96.

13.

Sutherland

. Cell and environment interactions in tumor microregions: the multicell spheroid model. Science 1988;240:177–84.

14.

Sutherland

Eddy

Bareham

. Resistance to Adriamycin in multicellular spheroids. Int J Radiat Oncol Biol Phys 1979;5:1225–30.

15.

Carlsson

Nilsson

Westermark

. Formation and growth of multicellular spheroids of human origin. Int J Cancer 1983;31:523–33.

16.

Soranzo

Ingrosso

. A comparative study of the effects of anthracycline derivatives on a human adenocarcinoma cell line (LoVo) grown as a monolayer and as spheroids. Anticancer Res 1988;8:369–73.

17.

Wartenberg

Acker

. Induction of cell death by doxorubicin in multicellular spheroids as studied by confocal laser scanning microscopy. Anticancer Res 1996;16:573–9.

18.

Fraefel

Jacoby

Lage

. Gene transfer into hepatocytes mediated by helper virus-free HSV/AAV hybrid vectors. Mol Med 1997;3:813–25.

19.

Jacobs

Rueger

Winkeler

. Imaging-guided gene therapy of experimental gliomas. Cancer Res 2007;67:1706–15.

20.

Jacobs

Winkeler

Hartung

. Improved herpes simplex virus type 1 amplicon vectors for proportional coexpression of positron emission tomography marker and therapeutic genes. Hum Gene Ther 2003;14:277–97.

21.

Wartenberg

Acker

. Quantitative recording of vitality patterns in living multicellular spheroids by confocal microscopy. Micron 1995;26:395–404.

22.

Meschini

Molinari

Calcabrini

. Intracellular localization of the antitumour drug Adriamycin in living cultured cells: a confocal microscopy study. J Microsc 1994;176:204–10.

23.

Wartenberg

Hescheler

Acker

. Doxorubicin distribution in multicellular prostate cancer spheroids evaluated by confocal laser scanning microscopy and the “optical probe technique.” Cytometry 1998;31:137–45.

24.

Wartenberg

Ling

Schallenberg

. Down-regulation of intrinsic P-glycoprotein expression in multicellular prostate tumor spheroids by reactive oxygen species. J Biol Chem 2001;276:17420–8.

25.

Bjerkvig

Tonnesen

Laerum

. Multicellular tumor spheroids from human gliomas maintained in organ culture. J Neurosurg 1990;72:463–75.

26.

Benedetti

Dimeco

Pollo

. Limited efficacy of the HSV-TK/GCV system for gene therapy of malignant gliomas and perspectives for the combined transduction of the interleukin-4 gene. Hum Gene Ther 1997;8:1345–53.

27.

Enger

Thorsen

Lonning

. Adeno-associated viral vectors penetrate human solid tumor tissue in vivo more effectively than adenoviral vectors. Hum Gene Ther 2002;13:1115–25.

28.

Jacobs

Voges

Reszka

. Positron-emission tomography of vector-mediated gene expression in gene therapy for gliomas. Lancet 2001;358:727–9.

29.

Voges

Reszka

Gossmann

. Imaging-guided convection-enhanced delivery and gene therapy of glioblastoma. Ann Neurol 2003;54:479–87.

30.

Krauze

Forsayeth

Yin

. Convection-enhanced delivery of liposomes to primate brain. Methods Enzymol 2009; 465:349–62.

31.

Jacobs

Tjuvajev

Dubrovin

. Positron emission tomography-based imaging of transgene expression mediated by replication-conditional, oncolytic herpes simplex virus type 1 mutant vectors in vivo. Cancer Res 2001;61:2983–95.

32.

Hampl

Camp

Mydlarz

. Potentiated gene delivery to tumors using herpes simplex virus/Epstein-Barr virus/RV tribrid amplicon vectors. Hum Gene Ther 2003;14:611–26.

33.

Hadaczek

Mirek

Berger

. Limited efficacy of gene transfer in herpes simplex virus-thymidine kinase/ganciclovir gene therapy for brain tumors. J Neurosurg 2005;102:328–35.

34.

Betz

Shakui

Davidson

. Gene transfer to rodent brain with recombinant adenoviral vectors: effects of infusion parameters, infectious titer, and virus concentration on transduction volume. Exp Neurol 1998;150:136–42.

35.

Chen

Lonser

Morrison

. Variables affecting convection-enhanced delivery to the striatum: a systematic examination of rate of infusion, cannula size, infusate concentration, and tissue-cannula sealing time. J Neurosurg 1999;90:315–20.

36.

Fiandaca

Varenika

Eberling

. Real-time MR imaging of adeno-associated viral vector delivery to the primate brain. Neuroimage 2009;47 Suppl 2:T27–35.

37.

Krauze

Saito

Noble

. Reflux-free cannula for convection-enhanced high-speed delivery of therapeutic agents. J Neurosurg 2005;103:923–9.

38.

Szerlip

Walbridge

Yang

. Real-time imaging of convection-enhanced delivery of viruses and virus-sized particles. J Neurosurg 2007;107:560–7.

39.

Dickinson

LeCouteur

Higgins

. Canine model of convection-enhanced delivery of liposomes containing CPT-11 monitored with real-time magnetic resonance imaging: laboratory investigation. J Neurosurg 2008;108:989–98.

40.

Bobo

Laske

Akbasak

. Convection-enhanced delivery of macromolecules in the brain. Proc Natl Acad Sci U S A 1994;91:2076–80.