9L Tumors Research Articles

In vivo enhancement of herpes simplex virus thymidine kinase/ganciclovir cancer gene therapy with polyamine biosynthesis inhibition

We have earlier demonstrated that inhibition of polyamine biosynthesis with difluoromethylornithine (DFMO) can be used to enhance the cytotoxicity of herpes simplex virus thymidine kinase/ganciclovir (HSV-TK/GCV) gene therapy in different tumor cell lines. Here, the utility of this treatment combination was tested in vivo in a nude mouse tumor model. First, the effect of DFMO was verified by treating mice bearing subcutaneous 9L rat glioma tumors with 2% DFMO in drinking water. The drug treatment induced almost complete suppression of ornithine decarboxylase activity, and as a result, a strong decrease in intratumoral putrescine and spermidine concentrations, which were normalized 4 days after drug removal. Consequently, the tumors displayed a significant reduction in the proliferation activity that was increased to 20% higher than the normal level at day 4 and returned to normal level 7 days after DFMO removal. Next, 9L tumors with 30% of TK-GFP fusion gene positive cells were induced and the animals were given DFMO and GCV in 2 treatment schemes, with the drug administration periods overlapping either 5 or 2 days. The analysis of tumor size at the end of the treatment revealed that DFMO can enhance HSV-TK/GCV cytotoxicity when the overlap between DFMO and GCV was 5 days, but the result was not significant. However, the 2-day overlap scheme yielded a significantly (p < 0.05, ANOVA) enhanced antitumor effect. In conclusion, the data here confirms that a novel combination of 2 clinically relevant treatment modalities, polyamine deprivation and HSV-TK/GCV suicide gene therapy, can be used synergistically in vivo.

9L Gliosarcoma Cell Proliferation and Tumor Growth in Rats Are Suppressed by N-Hydroxy-N′-(4-butyl-2-methylphenol) Formamidine (HET0016), a Selective Inhibitor of CYP4A

The present study examined the effects of N-hydroxy-N'-(4-butyl-2 methylphenyl) formamidine (HET0016), a selective inhibitor of the formation of 20-hydroxyeicosatrienoic acid (20-HETE) on the growth of 9L rat gliosarcoma cells in vitro and in vivo. After 48 h of incubation, HET0016 reduced the proliferation of 9L in vitro by 55%, and this was associated with a fall in p42/p44 mitogen-activated protein kinase and stress-activated protein kinase/c-Jun NH(2)-terminal kinase phosphorylation and increased apoptosis. HET0016 inhibited epidermal growth factor (EGF) and platelet-derived growth factor (PDGF)-induced proliferation and diminished phosphorylation of PDGF receptors. A stable 20-HETE analog increased 9L cell proliferation. In vivo, chronic administration of HET0016 (10 mg/kg/day i.p.) for 2 weeks reduced the volume of 9L tumors by 80%. This was accompanied by a 4-fold reduction in the mitotic index, a 3- to 4-fold increase in the apoptotic index, and a approximately 50% decrease in vascularization in the tumor. HET0016 treatment increased mean survival time of the animals from 17 to 22 days. Liquid chromatography/mass spectrometry experiments indicated that neither 9L cells grown in vitro nor 9L tumors removed produce 20-HETE when incubated with arachidonic acid. The normal surrounding brain tissue, however, avidly makes 20-HETE, and this activity is selectively inhibited by HET0016. These results suggest that HET0016 may be the prototype of a class of antigrowth compounds that may be efficacious for treating malignant brain tumors. In vivo, it may act in part by inhibiting the formation of 20-HETE by the surrounding tissue. However, the antiproliferative effects of HET0016 on 9L cells in vitro seem unrelated to its ability to inhibit the formation of 20-HETE.

Selective accumulation and strong photodynamic effects of a new photosensitizer, ATX-S10·Na (II), in experimental malignant glioma

We investigated the feasibility of a novel photosensitizer, ATX-S10.Na (II), in photodynamic therapy (PDT) for glioma. First, PDT was performed in various brain tumor cell lines in vitro. Cytotoxicity depended upon both drug concentration and laser energy and the 50% inhibitory concentration ranged from 3.5 to 20 microg/ml. Next, PDT was performed in the subcutaneous and intracranial 9L tumor models in Fischer rats using ATX-S10.Na (II) and light from a 670-nm diode laser delivered by intratumoral insertion of an optical fiber. The effect of PDT on brain tumors was evaluated using magnetic resonance imaging. Sequential changes of the ATX-S10.Na (II) concentrations were also measured quantitatively by fluorospectrometry up to 12 h after intravenous administration in rats with intracranial and subcutaneous tumors. The concentration of ATX-S10.Na (II) in the brain tumor reached a maximum at 2 h after administration and the tumor/normal brain concentration ratio was as high as 131 at 8 h. Intratumoral PDT for intracranial tumors irradiated at this timing showed an obvious anti-tumor effect without severe side effects. The present study demonstrated the highly selective accumulation of ATX-S10.Na (II) in tumor tissue and its potent photodynamic effect in an experimental malignant glioma model.

Fusion of the HSV-1 tegument protein vp22 to cytosine deaminase confers enhanced bystander effect and increased therapeutic benefit.

A major limitation in cancer gene therapy, specifically gene-dependent enzyme prodrug therapy (GDEPT), is inefficient gene delivery and expression. The suicide gene cytosine deaminase (CD) and its substrate, 5-fluorocytosine (5-FC), have been extensively explored due to the inherent 'bystander' effect achieved through diffusion of the toxic metabolite 5-fluorouracil (5-FU). In this study, we aimed to enhance this 'bystander' effect by fusing the Saccharomyces cerevisiae CD to the HSV-1 tegument protein vp22, a novel translocating protein. Two constructs were created: one with vp22 fused to CD (vp22CD) and a second wherein a truncated vp22, lacking the necessary residues for trafficking, fused to CD (delvp22CD). The generated 9L stable lines exhibited similar growth rates, enzyme expression, CD activity, and sensitivity to 5-FC and 5-FU. However, mixed population colony formation assays demonstrated greater bystander effect with the vp22CD fusion as compared to delvp22CD. This enhancement was maintained in vivo where 9L tumors expressing 20 or 50% vp22CD exhibited increased growth delay compared to the respective delvp22CD tumors. Moreover, adenoviral transduction of established wild-type 9L tumors showed increased growth delay with vp22CD (Ad-EF_vp22CD) as compared to equivalent CD (Ad-EF_CD) transduced tumors. Finally, confirming the increased efficacy, (19)F magnetic resonance spectroscopy (MRS) of vp22CD-expressing tumors demonstrated increased 5-FU levels as compared to tumors expressing the nontranslocating CD. These results together demonstrated that fusion of vp22 to CD resulted in CD translocation, which in turn amplified conversion of 5-FC to 5-FU in vivo and enhanced the therapeutic benefit of this GDEPT strategy.

Toxicity, biodistribution, and convection-enhanced delivery of the boronated porphyrin BOPP in the 9L intracerebral rat glioma model

To investigate the toxicity, biodistribution, and convection-enhanced delivery (CED) of a boronated porphyrin (BOPP) that was designed for boron neutron capture therapy and photodynamic therapy. For the toxicity study, Fischer 344 rats were injected with graded concentrations of BOPP (35-100 mg/kg) into the tail vein. For boron biodistribution studies, 9L tumor-bearing rats received BOPP either systematically (intravenously) or locally. All rats that received 70 mg/kg BOPP and 70% of rats that received < or = 60 mg/kg BOPP i.v. either had to be euthanized or died within 4 days of injection. In the biodistribution study, boron levels were relatively high in liver, kidney, spleen, and adrenal gland tissue, and moderate levels were found in all other organs. The maximum tumor boron concentration was 21.4 mug/g at 48 h after i.v. injection; this concentration of boron in brain tumors is at the low end of the range considered optimal for therapy. In addition, the tumor/blood ratio (approximately 1.2) was not optimal. When BOPP was delivered directly into intracerebral 9L tumors with CED, we obtained tumor boron concentrations much greater than those obtained by i.v. injection. Convection-enhanced delivery of 1.5 mg BOPP produced an average tumor boron level of 519 mug/g and a tumor/blood ratio of approximately 1850:1. Our study demonstrates that changing the method of BOPP delivery from i.v. to CED significantly enhances the boron concentration in tumors and produces very favorable tumor/brain and tumor/blood ratios.

Intratumoral injection of BCNU in ethanol (DTI-015) results in enhanced delivery to tumor – a pharmacokinetic study

Solvent facilitated perfusion (SFP) has been proposed as a technique to increase the delivery of chemotherapeutic agents to tumors. SFP entails direct injection of the agent into the tumor in a water-miscible organic solvent, and because the solvent moves easily through both aqueous solutions and cellular membranes it drives the penetration of the solubilized anticancer agent throughout the tumor. To test this hypothesis, we compared the pharmacokinetics (PK) of 14C-labeled 1,3-bis-chlorethyl-1-nitrosourea (BCNU) in intra-cerebral 9L rat gliomas after intravenous (IV) infusion in 90% saline--10% ethanol or direct intratumoral (IT) injection of 14C-BCNU in 100% ethanol (DTI-015). Treatment with DTI-015 yielded a peak radioactive count (Cmax) for the 14C label that was 100-1000 fold higher in the tumor than in all other tissues in addition to a concentration in the tumor that was 100-fold higher than that achieved following IV infusion of 14C-BCNU. Pathologic and auto-radiographic analysis of tissue sections following IT injection of 14C-BCNU in ethanol into either tumor or normal rat brain revealed both an enhanced local volume of distribution and an increased concentration of BCNU delivered to tumor compared to non-tumor bearing brain. To investigate the mechanism behind the SFP of BCNU to the tumor both dynamic contrast and perfusion MRI were performed on 9L tumors before and after treatment and demonstrated a decrease in tumor perfusion following IT injection of DTI-015. Finally, initial PK of patient blood samples following administration of DTI-015 into relapsed high-grade glioma indicated a 20-fold decrease in systemic exposure to BCNU compared to IV infusion of BCNU providing further evidence for the enhanced therapeutic ratio observed for DTI-015.

Gliosarcoma Cell Death After Radiosurgery in a Rat Model

Therapeutic radiation and subsequent detection of tumor cell death has been performed mainly in vitro systems, making it difficult to accurately characterize the mechanisms of tumor cell death after radiosurgery. To better characterize what occurs to glioma cells after radiation therapy, we developed a rat model using the 9L gliosarcoma cell line implanted reproducibly to the caudate nucleus in rats. After 1 Gy radiation, 9L tumors in vivo induced mainly necrosis (determined by trypan blue exclusion) of 10 - 74 % at 6 - 72 hours post-radiation. This is in contrast to a previous in vitro study which demonstrated that 18 Gy of radiation induces considerably less cell death as determined by trypan blue exclusion (approximately 20 - 25 % at 6 - 72 hours post-radiation). However, significant amounts of apoptosis were detected as early as 6 hours after radiation. Apoptosis determination was by annexin V (marker of early apoptosis) and propidium iodide (marker of membrane stability) staining followed by flow cytometry detection. When caspase 3 and caspase 8 enzymatic activities (mediators of apoptosis) were measured from freshly explanted tumor cells, peak activity was found 6 hours after 1 Gy radiation (p < 0.01). Taken together, these data indicate the presence of apoptosis early after radiation therapy (1 Gy) which progressed to necrosis in a unique in vivo model of gliosarcoma that may prove useful in determining new therapeutic approaches to radiation therapy and tumor cell biology.

A novel brainstem tumor model: guide screw technology with functional, radiological, and histopathological characterization

Survival rates for high-grade brainstem tumors are approximately 10% and optimal therapy has yet to be determined. Development of a satisfactory brainstem tumor model is necessary for testing new therapeutic paradigms that may prolong survival. The authors report the technique, functional progression, radiological appearance, and histopathological features of a novel brainstem tumor model in rats. Thirty female Fischer 344 rats were randomized (10 animals/group) to receive an injection of either 3 ml of 9L gliosarcoma cells (100,000 cells), 3 ml of F98 glioma cells (100,000 cells), or 3 ml of medium (Dulbecco modified Eagle medium) into the pontine tegmentum of the brainstem. Using a cannulated guide screw system implanted in the skull of the animal, rats in each group were injected at coordinates 1.4 mm to the right of the sagittal and 1 mm anterior to the lambdoid sutures, at a depth of 7 mm from the dura mater. The angle of the syringe during injection was anteflexed 5 degrees from the vertical. Postoperatively, the rats were evaluated for neurological deficits by using an automated rotarod test. High-resolution [18F]fluorodeoxyglucose-positron emission tomography (FDG-PET) fused with computerized tomography (CT) scans were acquired pre- and postoperatively through the onset of hemiparesis and correlated accordingly. Kaplan-Meier curves were generated for survival and disease progression, and brains were processed postmortem for histopathological investigation. The 9L and F98 tumor cells grew in 95% of the animals in which they were injected and resulted in a statistically significant mean onset of hemiparesis of 16.5 6 0.56 days (p = 0.001, log-rank test), compared with animals in the control group, which had no neurological deficits by Day 45. The FDG-PET studies coregistered with CT scans demonstrated space-occupying brainstem lesions, and this finding was confirmed by histological studies. Animals in the control group showed no functional, radiological, or pathological signs of tumor. Progression to hemiparesis was consistent in all tumor-injected animals, with predictable onset of symptoms occurring approximately 17 days postsurgery. The histopathological and radiological characteristics of the 9L and F98 brainstem tumors were comparable to those of aggressive primary human brainstem tumors. Establishment of this animal tumor model will facilitate the testing of new therapeutic paradigms for the treatment of these lesions.

Dexamethasone Normalizes Brain Tumor Hemodynamics as Indicated by Dynamic Susceptibility Contrast MRI Perfusion Parameters

The purpose of this study is to demonstrate the utility of dynamic susceptibility contrast (DSC) MRI-derived perfusion parameters to characterize the hemodynamic effects of dexamethasone in a 9L gliosarcoma tumor model. Twenty-four rats underwent intracerebral inoculation with 9L tumor cells. Fifteen were treated with a total of 3mg/kg of dexamethasone on days 10-14 post-inoculation, while the remaining 9 rats served as controls. Fourteen days post-inoculation, MRI images, sensitive to total and micro-vascular cerebral blood flow (CBF), mean transit time (MTT), and intravoxel transit time distributions (TTD)s were obtained using a simultaneous gradient-echo(GE)/spin-echo(SE) DSC-MRI method. Dexamethasone-treated animals had a microvascular (SE) tumor CBF that was 45.9% higher (p = 0.0008) and a MTT that was 47.8% lower (p = 0.0005) than untreated animals. With treatment, there was a non-significant 91.3% increase in total (GE) vascular CBF (p = 0.35), and a significant decrease in MTT (49.1%, p = 0.02). The total vascular and microvascular TTDs from the treated tumors were similar to normal brain, unlike the TTDs in the untreated tumors. These findings demonstrate that DSC-MRI perfusion methods can be used to non-invasively detect the morphological and functional changes in tumor vasculature that occur in response to dexamethasone treatment.

77. Efficient Transduction and Therapy of Malignant Glioma by Lentiviral Vectors Pseudotyped with LCMV Glycoproteins

Malignant gliomas, the most frequent primary brain tumors, still have a dismal prognosis despite advances in neurosurgery, radiation, and chemotherapy. Gene therapy using viral vectors represents an attractive alternative to conventional cancer therapies. In particular, retroviral vectors have been regarded as potent candidates as they transduce only dividing cells, thereby achieving specificity for tumor cells. However, clinical trials have failed, because in a given treatment window, only the minority of tumor cells is dividing and therefore susceptible to retroviral infection. To overcome this problem we used lentiviral vectors, which can transduce dividing as well as quiescent cells, for gene therapy of malignant glioma. In a previous study we established lentiviral vectors pseudotyped with lymphocytic choriomeningitis (LCMV) glycoproteins (GP) and demonstrated efficient transduction of human malignant glioma cells in culture. In the current investigation, we compared the transduction efficacy of LCMV GP- and vesicular stomatitis virus glycoprotein (VSV G)-pseudotyped lentiviral vectors for malignant glioma cells and normal brain cells in vitro and in vivo (Miletic et al., Human Gene Therapy 2004). LCMV pseudotypes transduced predominantly astrocytes, whereas VSV-G pseudotypes infected neurons as well as astrocytes in vitro and in the hippocampus and striatum of Fischer rats. LCMV pseudotypes showed an efficient transduction of solid glioma parts of 9L tumors (69.3 |[plusmn]| 10.6%) and a very specific transduction of infiltrating tumor cells (71.5 |[plusmn]| 10.6%). In contrast, VSV-G pseudotyped lentiviral vectors transduced only a few tumor cells in solid (6.8 |[plusmn]| 1.9%) and infiltrating (2.2 |[plusmn]| 1.9%) tumor parts and infected mostly normal brain cells in infiltrating tumor areas (42.9 |[plusmn]| 11.4%). To investigate the therapeutic impact of LCMV pseudotypes, we used the suicide gene HSV-tk fused to eGFP as therapeutic construct. We could demonstrate a high transduction efficiency of 9L tumors (Fig. 1A; Methionin-PET) using [18F]FHBG-PET to detect HSV-tk expression (Fig. 1B) and by fluorescence microscopy of histological brain sections (Fig. 1C). The Kaplan-Meier survival plot (Fig. 1D) revealed over 90% of long time survivors (100 days) in the therapeutic group (n=13). In conclusion, lentiviral vectors pseudotyped with LCMV glycoproteins represent an attractive option for gene therapy of malignant glioma.

Increased oxygenation of intracranial tumors by efaproxyn (efaproxiral), an allosteric hemoglobin modifier: In vivo EPR oximetry study

To determine quantitatively the changes in oxygenation of intracranial tumors induced by efaproxiral, an allosteric hemoglobin modifier. Efaproxiral reduces hemoglobin-oxygen binding affinity, which facilitates oxygen release from hemoglobin into surrounding tissues and potentially increases the pO(2) of the tumors. The study was performed on 10 male Fisher 344 rats with 9L intracranial tumors. Electron paramagnetic resonance (EPR) oximetry was used to measure quantitatively the changes in the pO(2) in the tumors. Lithium phthalocyanine (LiPc) crystals were implanted in the tumors and in the normal brain tissue in the opposite hemispheres. We monitored the cerebral pO(2) starting 7 to 10 days after the tumor cells were implanted. NMR imaging determined the position and size of tumor in the brain. After an initial baseline EPR measurement, efaproxiral (150 mg/kg) was injected intravenously over 15 minutes, and measurements of tumor and normal brain oxygen tension were made alternately at 10-minute intervals for the next 60 minutes; the procedure was repeated for 6 consecutive days. Efaproxiral significantly increased the pO(2) of both the intracranial tumors and the normal brain tissue on all days. The maximum increase was reached at 52.9 to 59.7 minutes and 54.1 to 63.2 minutes after injection, respectively. The pO(2) returned to baseline values at 106 to 126.5 minutes after treatment. The maximum tumor and normal tissue pO(2) values achieved after efaproxiral treatment from Day 1 through Day 6 ranged from 139.7 to 197.7 mm Hg and 103.0 to 135.9 mm Hg, respectively. The maximum increase in tumor tissue pO(2) values from Day 2 to Day 5 was greater than the maximum increase in normal tissue pO(2). We obtained quantitative data on the timing and extent of efaproxiral-induced changes in the pO(2) of intracerebral 9L tumors. These results illustrate a unique and useful capability of in vivo EPR oximetry to obtain repeated noninvasive measurements of tumor oxygenation over a number of days. The information on the dynamics of tumor pO(2) after efaproxiral administration illustrates the ability of efaproxiral to increase intracranial tumor oxygenation.

Therapeutic Efficacy of DTI-015 using Diffusion Magnetic Resonance Imaging as an Early Surrogate Marker

To investigate diffusion weighted magnetic resonance imaging as a quantitative surrogate marker for evaluating the therapy-induced cellular changes in an orthotopic experimental glioma model, tumors were treated with direct intratumoral administration of DTI-015, a solution of 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) in 100% EtOH. Intracerebral 9L tumors were induced in Fischer 344 rats, and three treatment groups were established: DTI-015, EtOH, and sham. Two groups of rats received intratumoral injection of either 67 mg/mL BCNU in EtOH or EtOH alone at 50% of the tumor volume up to a maximum of 30 mul under stereotactic guidance. Diffusion magnetic resonance images were acquired before treatment and after treatment at 1, 24, 48, and 72 hours and then 3 times per week thereafter. Tumor cell viability was examined using multislice diffusion weighted magnetic resonance imaging with diffusion weighted transverse magnetic resonance images and histogram plots of each tumor quantified over time. Control animals (EtOH- or sham-treated animals) showed mean apparent diffusion coefficients (ADCs) that remained essentially unchanged over the experimental time course. In contrast, rats treated with DTI-015 showed a significant increase in ADC relative to the pretreatment within 24 hours, which further increased over time, followed by a significant therapeutic response as evidenced by subsequent tumor volume shrinkage, development of a cystic region, and enhanced animal survival. Finally, not only were ADC measurements predictive of differences between treatment groups, but they also yielded spatial and temporal data regarding the efficacy of treatment within individual treated animals that could be used to guide subsequent therapy.

Highly Successful Therapeutic Vaccinations Combining Dendritic Cells and Tumor Cells Secreting Granulocyte Macrophage Colony-stimulating Factor

In an attempt to induce potent immune antitumor activities, we investigated, within the rat 9L gliosarcoma model, distal therapeutic vaccinations associating three therapies: dendritic cell vaccination, intratumoral granulocyte macrophage colony-stimulating factor (GM-CSF) gene transfer, and tumor apoptosis induction. Vaccines of dendritic cells coinjected with processed GM-CSF secreting 9L cells induced systemic responses, resulting in the complete regression of distant preimplanted 9L tumor masses in, with the best strategy, 94% of male rats. All of the cured rats developed a long-term resistance to a rechallenge with parental cells. The curative responses were correlated with the detection of elevated specific cytotoxic activities and a CD4+, CD8+ T cell-, and natural killer (NK) cell-mediated IFN-gamma production. The survival rate of the rat seemed more directly linked to the amount of GM-CSF secreted by the transduced tumor cells, which in turn depended on the toxicity of the apoptosis-inducing treatment, than to the level of apoptosis induced. Unexpectedly, alive GM-CSF secreting 9L cells became apoptotic when injected in vivo. Thus we documented the positive role of apoptosis in the induction of therapeutic antitumor responses by comparing, at equal GM-CSF exogenous supply, the effects of dendritic cells coinjected with apoptotic or necrotic 9L cells. The data showed the superior therapeutic efficiency of combined vaccines containing apoptotic tumor cells. In conclusion, vaccinations with dendritic cells associated with apoptotic tumor cells secreting GM-CSF show a very high therapeutic potency that should show promise for the treatment of human cancer.

Invasive phenotype observed in 1,3-bis(2-chloroethyl)-1-nitrosourea—resistant sublines of 9L rat glioma cells: a tumor model mimicking a recurrent malignant glioma

Chemotherapy is suspected of having an effect on the generation of phenotypical heterogeneity and the development of drug resistance in tumors. Recurrent gliomas feature drug resistance as well as greater invasive growth than original tumors. The authors investigated phenotypical changes in invasion observed in 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU)-resistant sublines of the 9L rat gliosarcoma. Two established BCNU-resistant sublines, derived from 9L gliosarcoma cells by treating these cells with BCNU in vivo or in vitro, were used in the study. An in vitro examination confirmed the resistance of the cells to BCNU treatment. The cells were implanted into the striatum of Fisher 344 rats, and histological examinations were performed to compare the growth patterns of the resultant tumors. A new brain tumor model was established by implanting 9L-2 cells in Fisher 344 rats. The 9L-2 and BTRC-19 cells displayed a distinct increase in BCNU resistance compared with the 9L cells. Both BCNU-resistant sublines developed a tumor mass with invasive margins, which is not the case with 9L tumor models. The newly developed 9L-2 tumor model demonstrated 100% tumor uptake with consistent growth patterns. Cells that acquire drug resistance also demonstrated invasive growth. Because the 9L-2 and BTRC-19 cells were derived from 9L cells that had been treated with BCNU in vivo and in vitro, this change in phenotype was likely caused by the drug treatment, which may have implications for chemotherapy of gliomas. The tumor model that developed from the 9L-2 cells can be used as a model of a recurrent glioma, which features drug resistance and invasive growth.

Tumor-dependent kinetics of partial pressure of oxygen fluctuations during air and oxygen breathing.

The primary purpose of this study was to examine the kinetics of partial pressure of oxygen (pO2) fluctuations in fibrosarcoma (FSA) and 9L tumors under air and O2 breathing conditions. The overall hypothesis was that key factors relating to oxygen tension fluctuations would vary between the two tumor types and as a function of the oxygen content of the breathing gas. To assist in the interpretation of the temporal data, spatial pO2 distributions were measured in 10 FSA and 8 9L tumors transplanted into the subcutis of the hind leg of Nembutal-anesthetized (50 mg/kg) Fischer 344 rats. Recessed-tip oxygen microelectrodes were inserted into the tumor, and linear pO2 measurements were recorded in 50-microm steps along a 3-mm path, and blood pressure was simultaneously measured via femoral arterial access. Additionally, pO2 was measured at a single location for 90 to 120 minutes in FSA (n=11) or 9L tumors (n=12). Rats were switched from air to 100% O2 breathing after 45 minutes. Temporal pO2 records were evaluated for their potential radiobiological significance by assessing the number of times they crossed a 10-mm-Hg threshold. In addition, the data were subjected to Fourier analysis for air and O2 breathing. FSA and 9L tumors had spatial median pO2 measurements of 4 and 1 mm Hg, respectively. 9L had more low pO2 measurements < or =2.5 mm Hg than did FSA, whereas between 2.5 and 10 mm Hg this pattern was reversed. Pimonidazole staining patterns in FSA and 9L tumors supported these results. Temporal pO2 instability was observed in all experiments during air and O2 breathing. Threshold analyses indicated that the 10 mm Hg threshold was crossed 2 to 5 times per hour, independent of tumor type. However, the magnitude of 9L pO2 fluctuations was approximately eight times greater than FSA fluctuations, as assessed with Fourier transform analysis (Wilcoxon, P < 0.005). O2 breathing significantly increased median pO2 in FSA from 3 to 8 mm Hg (P < 0.005) and caused a significant increase in frequency and magnitude of pO2 fluctuations. One hundred percent O2 breathing had no effect on 9L tumor pO2, and it decreased the magnitude of pO2 fluctuations with borderline significance. These results show that these two tumors differ significantly with respect to spatial and temporal oxygenation conditions under air and O2 breathing. Fluctuations of pO2 of the type reported herein are predicted to significantly affect radiotherapy response and could be a source for genetic instability, increased angiogenesis, and metastases.

Microglia phagocytose alloreactive CTL-damaged 9L gliosarcoma cells

Intracranial adoptive transfers of alloreactive cytotoxic T lymphocytes (aCTL) for brain tumor treatment were safe and showed promise in preclinical and early clinical trials. To better understand the endogenous immune responses that may ensue following cellular therapy with aCTL, we examined the ability of microglia to phagocytose aCTL-damaged and undamaged rat 9L gliosarcoma cells in vitro and in vivo. In vitro, 5.5±0.9% of microglial cells isolated from adult tumor-bearing rat brains phagocytosed aCTL-damaged 9L cells, whereas microglia did not bind to or ingest undamaged 9L cells. Addition of supernates from either 9L cell cultures or from aCTL+9L co-incubate cell cultures to microglia did not significantly alter their ability to bind to or phagocytose damaged glioma cells even though the latter contained T helper 1 and 2 cytokines. At 3 days following intracranial 9L cell infusion, 17.5±0.1% of the microglia phagocytosed CFSE-labeled aCTL-damaged 9L tumor cells within the adult rat brain, confirming the in vitro data. The results suggest that microglia within the tumor microenvironment of the adult rat glioma model selectively remove damaged, but not undamaged, glioma cells.

Biopolymer-mediated suramin chemotherapy in the treatment of experimental brain tumours

Suramin inhibits tumour growth and neoangiogenesis by blocking several growth factor receptors. In this study the toxicity and efficacy of intralesional delivery of suramin incorporated in a controlled-release polymer were assessed in a rat 9L tumour model. Initially, the toxicity of the compound was evaluated in adult Fisher 344 rats. The animals were intracerebrally implanted with an ethylene vinyl acetate copolymer. These experiments showed early toxicity in the rats implanted with a 50% load-polymer and 100% mortality within 48 h, whereas in rats implanted with a 33% load-polymer only transient behavioural changes were observed. In a second experiment the rats were stereotactically implanted with 9L cells in the frontal region. Two days after inoculation of cells, the animals were divided into two groups: one group received a 33% suramin load-biopolymer at the tumour implantation site, while the control group received polymer implants only. The interstitial release of suramin in the brain did not produce any improvement in survival of 9L tumour-bearing rats, with a mean survival of 14.2±1 days for the suramin-treated group versus 13.8±2 for the control group (p=0.82). We conclude that intralesional polymer-mediated chemotherapy with suramin does not prolong survival in rats with intracerebral 9L tumours.

In situ generation of pseudotyped retroviral progeny by adenovirus-mediated transduction of tumor cells enhances the killing effect of HSV-tk suicide gene therapy in vitro and in vivo.

Hybrid adeno-retroviral vector systems utilize the high efficiency of adenovirus transduction to direct the in situ production of retroviral progeny. In this study, we show that a single-step transduction of glioma cells with trans-complementing hybrid adeno-retroviral vectors effectively turns these cells into retrovirus vector-producing cells, which in turn facilitates the transduction of adjacent cells. We have adapted the adeno-retroviral hybrid viral vector system to enhance the ganciclovir (GCV) killing of glioma cells following transfer of the herpes simplex virus thymidine kinase (HSV-tk) gene. To assess the effect of the in situ production of retroviral vectors on the transduction efficiency of glioma cells, 9L cells were transduced with adeno-retroviral hybrid vectors that separately express a retroviral genome (AVC2.GCEGFP or AVC2.GCTK) and retroviral packaging proteins (AxTetGP and AxTetVSVG). The generation of an integrated HSV-tk provirus by trans-complementation of the adeno-retroviral vectors was verified by analysis of the flanking retroviral LTR sequences. Tumors established on nu/nu mice were injected with the viruses followed by intraperitoneal injections of either PBS or GCV. We also estimated the copy numbers of the HSV-tk transgene present in the tumors of the treated mice. To determine the expression pattern of the HSV-tk transcripts within a tumor, in situ hybridization analysis was performed using an RNA probe specific for HSV-tk. The co-transduction of rat 9L glioma cells with AVC2.GCEGFP together with vectors expressing packaging proteins of retroviruses increased the transduction efficiency. Transduction with AVC2.GCTK together with packaging vectors increased the in vitro sensitivity of cells to the pro-drug GCV by one log compared with control cells that were incapable of generating retrovirus. In vivo, the injection of established subcutaneous 9L tumors on athymic mice with a combination of AVC2.GCTK and packaging vectors followed by GCV treatment resulted in complete tumor regression in 50% of tumors at day 22, while no tumor regression was observed in control animals. Retroviral sequences diagnostic of 3' LTR reduplication in vivo were detected in genomic DNA extracted from the transduced tumors, indicating pro-viral integration of the retroviral genome derived from the adeno-retroviral hybrid vector. Furthermore, the relative copy number of the HSV-tk gene in tumors treated with the adeno-retroviral vectors was up to approximately 250-fold higher than in control tumors. In situ hybridization suggested dispersion of the HSV-tk product across a wider area of the tumor than in control tumors, which indicates the spread of the in situ generated retroviruses. Although the efficacy of this system has to be evaluated in orthotopic models, our observations suggest that this hybrid adeno-retroviral vector system could improve the suicide gene therapy of tumors.

Effects of combined granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-2, and interleukin-12 based immunotherapy against intracranial glioma in the rat.

Cytokines play a major role in the regulation of the immune system. Granulocyte-macrophage colony-stimulating factor (GM-CSF) has been shown to be useful for immunotherapy against glioma because it can stimulate dendritic cells to present tumor antigen. Interleukin-2 (IL-2) is involved in T-cell expansion, and interleukin-12 (IL-12) drives the T-helper cell type I response. Previous studies have shown that each of these cytokines alone can induce the regression of tumor cells. In the present study we postulated that peripheral infusion of GM-CSF along with either IL-2 or IL-12 and irradiated tumor cells can lead to increased survival from 9L brain tumors. 9L gliosarcoma cells (10(6)) were implanted in the brains of syngeneic Fischer 344 rats. Osmotic minipumps were utilized for subcutaneous, continuous delivery of GM-CSF, either alone or with IL-2 or IL-12. Irradiated 9L cells were injected subcutaneously at various time points during treatment. Delayed-type hypersensitivity (DTH) and immunohistological analysis were used to further characterize the anti-tumor response. Treatment with GM-CSF and irradiated tumor cells led to an increase in survival rate in rats with intracranial 9L tumors when compared to untreated animals. The addition of IL-2 or IL-12 to the GM-CSF/tumor cell therapy further increased the survival rate up to 90%. The anti-tumor response was associated with vigorous DTH against 9L cells and increased infiltration of CD4+ and CD8+ lymphocytes into the tumor. These results suggest that the combined infusion of GM-CSF and other cytokines may be effective adjuvants in treating brain tumors.

TK-GFP fusion gene virus vectors as tools for studying the features of HSV-TK/ganciclovir cancer gene therapy in vivo

The fusion gene of herpes simplex virus thymidine kinase and green fluorescent protein (TK-GFP) was shown to be a versatile tool for examining the features of thymidine kinase/ganciclovir gene therapy in vitro. In this study, we used viral vectors carrying the fusion gene to characterize the aspects of this gene therapy form in rodent tumor models. Growth of subcutaneous 9L rat tumors transduced ex vivo with TK-GFP gene was prevented when ganciclovir (GCV) treatment was initiated immediately after tumor inoculation. Established tumors (>100 mm(3)), however, were untreatable despite the initial 55% proportion of TK-GFP positive cells. This was due to a rapid clearance of TK-GFP positive cells, but not GFP positive cells. Propidium iodide staining revealed that TK-GFP lentivirus vector was able to induce apoptosis/necrosis in 9L cells, as opposed to the respective GFP vector. Furthermore, when a subcutaneous nude mouse tumor model was used, the percentage of TK-GFP positive cells in vivo was maintained similarly as in cultured cells, suggesting contribution of a fully functional immune response to the disappearance of fusion gene positive cells. In vivo gene transfer studies: adenovirus TK-GFP vector injections resulted in about 25% gene transfer efficiency to 9L tumors and showed that their growth could be significantly reduced even when the tumor volumes were already >120 mm(3). Part of the effect was shown to be due to cytotoxicity of the vector. In summary, our results demonstrate the utility of TK-GFP fusion gene-carrying viral vectors in animal studies and show that readily detectable therapeutic genes can help us to understand the complicated nature of in vivo cancer gene therapy experiments.