Skip to main content
NCBI home page
Molecular Therapy logoLink to Molecular Therapy
. 2012 Nov 13;21(2):300–308. doi: 10.1038/mt.2012.229

Systemic TNFα Gene Therapy Synergizes With Liposomal Doxorubicine in the Treatment of Metastatic Cancer

Baowei Su 1, Arzu Cengizeroglu 1, Katarina Farkasova 1, Joana R Viola 1, Martina Anton 2, Joachim W Ellwart 3, Rudolf Haase 1, Ernst Wagner 1,4, Manfred Ogris 1,4,*
PMCID: PMC3594014  PMID: 23299796

Abstract

Tumor necrosis factor alpha (TNFα) is a potent antitumoral cytokine, either killing tumor cells directly or affecting the tumor vasculature leading to enhanced accumulation of macromolecular drugs. Due to dose limiting side effects systemic administration of TNFα protein at therapeutically active doses is precluded. With gene vectors, tumor restricted TNFα expression can be achieved and in principle synergize with chemotherapy. Synthetic gene carriers based on polyamines were intravenously injected, which either passively accumulate within the tumor or specifically target the epidermal growth factor receptor. A single intravenous injection of TNFα gene vector promoted accumulation of liposomal doxorubicine (Doxil) in murine neuroblastoma and human hepatoma by enhancing tumor endothelium permeability. The expression of transgenic TNFα was restricted to tumor tissue. Three treatment cycles with TNFα gene vectors and Doxil significantly delayed tumor growth in subcutaneous murine Neuro2A neuroblastoma. Also tumors re-growing after initial treatment were successfully treated in a fourth cycle pointing at the absence of resistance mechanisms. Systemic Neuro2A metastases or human LS174T colon carcinoma metastases in liver were also successfully treated with this combined approach. In conclusion, this schedule opens the possibility for the efficient treatment of tumors metastases otherwise not accessible for macromolecular drug carriers.

Introduction

Above a minimal size tumor lesions rely on a functional blood supply for access to oxygen and nutrients. Albeit being well vascularized, solid tumors often exhibit a disorganized vessel structure and increased interstitial pressure, which precludes efficient systemic treatment with chemotherapeutics. To achieve sufficient drug concentrations in tumor tissue, elevated doses have to be applied leading to unwanted side effects in nontarget organs. Improving the accessibility of tumors for chemotherapeutic drugs is hence a major aim in tumor therapy. One attempt is to attack the tumor vasculature: e.g., combining the anti-VEGF antibody bevacizumab with chemotherapeutics improved tumor blood flow and accessibility to chemotherapeutics.1 Another approach is to enhance permeability of tumor blood vessels. Treating rats with vasoactive nitroglycerol improved accumulation of adenoviral vectors in liver tumors by permeabilizing the tumor vasculature.2 An important endogenous mediator of inflammatory processes and endothelial permeability is the cytokine tumor necrosis factor alpha (TNFα).3 Secreted TNFα binds to its receptors TNFR-1 and -2 inducing inflammation and upregulation of nuclear factor kappa B (NFκB), where the latter effect renders many tumors resistant against TNFα mediated apoptosis.4 Only in case of an inadequate NFκB response, double-stranded RNA-activated protein kinase PKR, e.g., can lead to tumor cell apoptosis.5 TNFα mediated inflammation can also activate tumor endothelium promoting extravasation of circulating macrophages and other immune cells into the tumor stroma.6 This could also allow enhanced permeation of chemotherapeutics into tumors. However, the clinical use of TNFα protein is limited to local treatments because of dose-limiting systemic toxicity.7 Therefore, only isolated limb perfusion in combination with the drug melphalan is approved for the treatment of sarcoma and successfully applied to prevent sarcoma-caused limb amputation.8 Applying gene vectors encoding for TNFα is one option to obtain controllable and tumor targeted action of TNFα. TNFerade, an adenoviral vector, is the currently most advanced gene vector for treating tumors with TNFα and also synergizes with gemcitabine pointing at the vasoactive role of TNFα.9 Although being highly effective in inducing transgene expression, viral gene vectors evoke inflammatory and immunological responses and are hence limited in their applicability for repeated systemic injection. In this study, we apply a plasmid vector optimized for in vivo transgene expression complexed with synthetic polymers without or with targeting to the epidermal growth factor receptor (EGFR). The local expression of TNFα within tumor cells should activate the tumor endothelium and promote accumulation of macromolecular drugs making this treatment also applicable for tumors otherwise resistant toward direct tumor cell killing by TNFα.

Results

The plasmid vector pCpG-hCMV-EF1α-TNFα (short: pEF1α-TNF) carries a synthetic, expression optimized TNFα cDNA driven by a human cytomegalovirus (CMV) enhancer and the human elongation factor 1α promoter. Transfection of Neuro 2A murine neuroblastoma cells with the biodegradable carrier G3-HD-OEI10 and 500 ng plasmid per well led to sustained and high TNFα transgene expression with up to 600 pg protein/ml/24 hours peaking between 48 and 72 hours after transfection (Figure 1a). Increasing the amount to 1,000 ng plasmid did not increase the TNFα expression, whereas with 250 or 500 ng plasmid expression was lower (eightfold and fourfold, respectively). No TNFα was detectable in the supernatant of cells transfected with control polyplexes containing plasmid pCpG-hCMV-EF1α-Luc (short: pEF1α-Luc).11 Transgenic TNFα was fully biofunctional as shown in a bioassay (Supplementary Figure S1 online) but significant reduction in cell viability was only seen on L929 murine fibroblasts (Figure 1b), a cell line used for the TNFα bioassay.12 No significant cell killing was observed in other cell lines as compared with treatment with pEF1α-Luc control polyplexes. The effect of transgenic TNFα on endothelial cell permeability was evaluated in an in vitro trans-well system. Supernatant from transfected N2A cells or recombinant TNFα was applied onto a confluent layer of porcine endothelial cells and the permeation of fluoresceinisothiocyanate (FITC) labeled dextran measured (Figure 1c). Both, recombinant TNFα and the supernatant of Neuro2A cells transfected with pEF1α-TNF increased the transfer of FITC-dextran through the endothelial cell layer. As a first in vivo model, A/J mice bearing subcutaneously implanted murine Neuro2A neuroblastoma were systemically treated with polyplexes. Repeated systemic application of pEF1α-Luc polyplexes based on G3-HD-OEI polymer at a dose of 2.5 mg/kg allowed tumor-restricted transgene expression for >1 week (Figure 2a). Luciferase expression in all other organs examined (heart, lung, liver, spleen, and kidney) was at least 100-fold lower than in the tumor, as quantified in tissue lysates 2 days after the last polyplex injection (Supplementary Figure S2 and Supplementary Materials and Methods online). In the same model, TNFα mRNA was quantified in liver and tumor tissue by quantitative PCR 48 hours after a single injection of pEF1α-TNF/G3-HD-OEI polyplexes (Figure 2b). With TNFα polyplexes, the glyceraldehyde-3-phosphate dehydrogenase-mRNA normalized TNFα-mRNA level was ~30-fold higher and 20-fold higher in tumor and liver, respectively, as compared with the control, but also with pEF1α-Luc/G3-HD-OEI control polyplexes elevated TNFα mRNA levels were found. In an attempt to distinguish between endogenous and transgenic TNFα, primers for quantitative PCR were designed in a way that allowed discrimination of the two types of TNFα mRNAs: the relative level of transgenic TNFα mRNA in tumor tissue was about sixfold higher than in liver showing tumor specific expression of transgenic TNFα. The effect of TNFα on tumor endothelium was evaluated by staining for the endothelial cell marker CD31 in tumor cryosections of animals treated with a similar schedule as in Figure 2a (Figure 2c). Only pEF1α-TNF polyplexes significantly reduced endothelial cell density (P < 0.05, pEF1α-TNF treatment versus pEF1α-Luc treatment and control, ANOVA (Duncan)), whereas in pEF1α-Luc treated animals no such reduction was observed as compared with the control.

Figure 1.

Figure 1

Open in a new tab

Polyplex mediated expression of bioactive Tumor necrosis factor alpha (TNFα) in vitro. Indicated tumor cell lines were transfected with polyplexes based on the biodegradable polymer G3-HD-OEI using plasmids pEF1α-TNF or pEF1α-Luc. (a) Neuro2A murine Neuroblastoma (7 ×104 seeded cells/well) were transfected with 1 µg pEF1α-TNF polyplex, the supernatant collected every 24 hours and the TNFα content measured by enzyme-linked immunosorbent assay (n = 3 + SD). (b) Indicated cell lines were transfected with 1µg polyplex formed either with pEF1α-TNF (full bars) or pEF1α-Luc (empty bars). After 24 hours cellular cytotoxicity was measured by MTT assay and expressed as percent reduction in viability as compared with untreated cells (n = 3 + SD). (c) A confluent layer of porcine endothelial cells (PEC) grown on a membrane in a trans-well system was treated with supernatant from either pEF1α-TNF or pEF1α-Luc transfected Neuro2A cells or with 1 ng TNFα protein together with 1 µg/ml Actinomycin D for 24 hours. Thereafter, the relative transfer of 40 kDa fluoresceinisothiocyanate-dextran through the cell layer occurring within 50 minutes was measured.

Figure 2.

Figure 2

Open in a new tab

Polyplex mediated transgene expression in vivo. A/J mice bearing subcutaneously implanted Neuro2A tumors were treated intravenously with G3-HD-OEI polyplexes containing plasmids pEF1α-tumor necrosis factor (TNF) or pEF1α-Luc. (a) pEF1α-Luc polyplexes (2.5 mg/kg pDNA) were injected at day 8, 11, and 14 after tumor inoculation (arrow heads) and the BLI signal quantified in the tumor area after injection of luciferin. The insert shows a representative animal at 48 hours after polyplex injection. n = 3 + SD. (b) Mice were treated once intravenously with pEF1α-TNF or pEF1α-Luc polyplexes or buffer alone (ctrl), 48 hours thereafter killed and the mRNA level of total TNFα normalized to the housekeeper GAPDH determined in liver and tumor; *P < 0.05, t-test (independent) ctrl versus pEF1α-Luc or pEF1α-Luc versus pEF1α-TNF. (c) Mice were treated on day 9, 12, and 14 after tumor inoculation with buffer alone (ctrl), pEF1α-Luc or pEF1α-TNF polyplexes. On day 17 animals were killed and tumor cryosections stained for CD31 (orange), nuclei were counterstained with DAPI (blue). GAPDH, glyceraldehyde- 3-phosphate dehydrogenase.

To evaluate the effect of TNFα on the permeability of tumor endothelium for macromolecular drugs, A/J mice with subcutaneous Neuro2A tumors were first treated with EF1α-TNF polyplexes, 48 hours thereafter with 5 mg/kg Doxil, and doxorubicine concentration was quantified in tumor tissue after further 24 hours (Figure 3a). With TNFα pre-treatment, an on average >12-fold increased accumulation was observed in tumor lysates reaching a doxorubicine concentration >10 µg/g wet tumor tissue. In liver tissue only a 1.7-fold, although statistical significant, increase of doxorubicine concentration was observed (Doxil treatment: 0.41 ± 0.1 µg/g; pEF1α-Luc/Doxil treatment: 0.56 ± 0.1 µg/g; pEF1α-TNF/Doxil treatment: 0.71 ± 0.2 µg/ml; P < 0.05 pEF1α-TNF/Doxil versus all groups, ANOVA (Duncan)). Still, the absolute level remained very low as compared with the tumor tissue. With a similar schedule, nude mice bearing HUH7 hepatocellular carcinoma xenografts were treated with EGF receptor targeted polyplexes using the conjugate LPEI-PEG-GE11.13 Here a 3.5-fold increase in Doxil accumulation was achieved (Figure 3b). Within cryosections of Neuro2A tumors, we observed large areas of extravasated doxorubicine; its localization within tumor cell nuclei points at sufficient drug release from liposomes (Figure 3c, top right panel). Furthermore, no increased extravasation was found in liver indicating the absence of sufficient TNFα concentrations in circulation. To track the distribution of Doxil in real time, we labeled the liposomes with the near infrared emitting lipidic dye 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide (Supplementary Figure S3 and Supplementary Materials and Methods online). Fifteen minutes after administration, most of the material remained in circulation and no differences were seen between pEF1α-TNF and pEF1α-Luc polyplex treated animals. After 24 hours, pEF1α-Luc treated animals exhibited only weak accumulation in the tumor, whereas in pEF1α-TNF polyplex treated tumors fluorescence levels were distinctly increased.

Figure 3.

Figure 3

Open in a new tab

Tumor necrosis factor alpha (TNFα) transgene expression enhances accumulation of Doxil in subcutaneous tumors. A/J mice with subcutaneous Neuro2A tumors (a and c) or NMRI nu/nu mice with subcutaneous HUH7 hepatoma (b) were treated systemically with polyplexes (formed with plasmid pEF1α-Luc or pEF1α-TNF), either based on the biodegradable polymer G3-HD-OEI (a and c) or the EGF-receptor targeted conjugate LPEI-PEG-GE11 (b) and Doxil. (a and b) Mice bearing tumors with ~ 8 mm in diameter were treated intravenously with indicated polyplexes (2.5 mg/ml) and 48 hours thereafter with Doxil (5 mg/kg), control mice (ctrl) received Doxil only. After further 24 hours mice were killed and the tumor concentration of doxorubicine quantified (n = 4–6 + SD). (c) Cryosections from tumors (top row) and livers (bottom row) from A/J mice treated as in (a) are shown (fluorescent doxorubicine shows up in orange).

To evaluate the therapeutic potential of TNFα gene therapy, A/J mice were treated 9 days after Neuro2A tumor inoculation with three doses of TNFα polyplex (2.5 mg/kg plasmid per injection) every third day and tumor growth was monitored (Figure 4). Comparing pEF1α-TNF- and pEF1α-Luc polyplex treated animals, a moderate growth retardation was observed in the pEF1α-TNF treated group (Figure 4a). To unleash the full therapeutic potential of TNFα gene therapy, the treatment was combined with Doxil (Figure 4b) in three treatment cycles of polyplex injection followed by Doxil (1.5 mg/kg Doxil). Doxil alone or in combination with pEF1α-Luc polyplexes could not significantly reduce tumor growth. In contrast, the TNFα/Doxil combination stopped tumor growth already after the second treatment cycle and lasted until 8 days after the third treatment, resulting in a significantly reduced tumor size in the TNFα/Doxil treatment group compared to the Luc/Doxil group. Thereafter, re-growth of the tumor occurred. In a separate experiment, a fourth treatment cycle was conducted at onset of tumor re-growth following the first three treatment cycles. Again, tumor growth was significantly retarded.(Figure 4c). Quantitative analysis of CD31 staining on cryosections exhibited a significant decrease in signal intensity by 50 % in the pEF1α-TNF treated group.

Figure 4.

Figure 4

Open in a new tab

Tumor necrosis factor alpha (TNFα) gene therapy synergizes with Doxil in tumor growth reduction. A/J mice bearing subcutaneous Neuro2A tumors were treated with G3-HD-OEI polyplexes (2.5 mg/kg pDNA per dose) and Doxil (1.5 mg/kg per dose) and the tumor volume followed over time by caliper measurements. Control animals (ctrl) received buffer alone. (a) Animals were treated on day 9, 12, and 15 after tumor inoculation with pEF1α-Luc (n = 5) or pEF1α-TNF polyplexes (n = 6). (b) Combined treatment with three cycles of polyplexes (full arrow heads) and Doxil (empty arrow heads) started at day 10 after tumor setting. Animals treated with Doxil alone were treated on day 12, 16, and 20 after tumor setting; n = 3. (c) treatment as in b (ctrl: n = 4, Doxil 3×: n = 5, pEF1α-Luc + Doxil 3×: n = 3, pEF1α-TNF + Doxil 3×: n = 4), but with an additional treatment group (pEF1α-TNF + Doxil 4× n = 7), where a fourth treatment cycle was applied on day 26 and 28 after tumor inoculation; §P < 0.05, t-test (independent) pEF1αLuc 3× versus pEF1αTNF 3× *P < 0.05, **P < 0.01, ANOVA (Duncan), pEF1α-Luc + Doxil 3× versus pEF1α-TNF + Doxil 4× #P < 0.05, t-test (independent), pEF1α-TNF + Doxil 3× versus pEF1α-TNF + Doxil 4×.

To better reflect the clinical situation of metastasized malignancies, we established a disseminated neuroblastoma model by intravenous injection of Neuro2A cells stably expressing luciferase (Figure 5). Two weeks after tumor inoculation, luciferase expressing tumor metastases were detectable by bioluminescence imaging (BLI) measurement and exhibited an exponential increase in BLI signal up to day 22 (Figure 5a). Thereafter animals became morbid and had to be killed. Combined TNFα/Doxil chemo–gene therapy started on day 6 after tumor injection; three treatment cycles were applied similar as in Figure 4b and the BLI signal measured (Figure 5a,b). Treating animals with control polyplexes (promoterless plasmid pΔEF1α-TNF) and Doxil, only a moderate signal reduction was observed, whereas with pEF1α-TNF/Doxil treatment the total BLI signal, but also the number and BLI intensity of tumor nodules was highly significant decreased (P values ranging from <0.01 to <0.001). After killing animals at day 22, several large (>5mm in diameter) and well vascularized tumor nodules were found in the abdominal region of untreated or pΔEF1α-TNF polyplex/Doxil treated animals (Figure 5d). In sharp contrast, after pEF1α-TNF polyplex/Doxil treatment only few small tumors remained. Total tumor burden in the visceral organs was measured by deducting the average visceral weight from similarly treated, but tumor-free animals (Figure 5c), where TNFα/Doxil treatment significantly reduced tumor burden as compared with the controls.

Figure 5.

Figure 5

Open in a new tab

Tumor necrosis factor alpha (TNFα)/Doxil chemo–gene therapy for the treatment of metastatic murine neuroblastoma. A/J mice were injected intravenously with luciferase marked Neuro2A cells. Control animals (ctrl) received buffer solution i.v. Treated animals received G3-HD-OEI polyplexes with plasmid pEF1α-TNF (pEF1α-TNF) or the promoterless plasmid pΔEF1α-TNF (pΔEF1α-TNF) (2.5 mg/kg) on day 6, 10 and 14, Doxil (1.5 mg/kg) was applied on day 8, 12, and 16 (n = 11), animals were sacrificed on day 22. (a) Whole body BLI signal as a measure for total tumor burden; *P < 0.05 ctrl versus pΔEF1α-TNF treated, **P < 0.01, ***P < 0.001 ctrl versus pEF1α -TNF treated; (ANOVA; Duncan). (b) Luciferase activity in selected tumor bearing animals; a reflected light image (black–white) is overlaid by the color codes BLI signal. (c) Average tumor weight located in the visceral of animals + SD;**P < 0.01 ctrl versus pEF1α -TNF treated; (ANOVA; Duncan). (d) Representative pictures of animals from (a) with opened body cavities, yellow arrows point at tumor lesions. n.s., non significant.

In addition to the syngeneic murine tumor model, we treated a disseminated liver xenograft model of human LS174T colon cancer in nude mice with the combined TNFα/Doxil chemo–gene therapy starting 3 days after tumor inoculation (Figure 6). To monitor tumor growth, LS174T cell were marked with luciferase as described above. One day after the third treatment cycle, animals were killed, liver and spleen explanted, weighted and luciferase activity in liver lysates quantified. Untreated animals developed a high tumor burden, which increased the liver weight by up to 50%, and multifocal tumors grew in all liver lobes. In the spleen, a large primary tumor developed. Although pEF1α-Luc polyplex/Doxil treatment only moderately decreased tumor burden, combined TNFα/Doxil therapy significantly reduced the tumor load with the average liver and spleen weight reaching levels of tumor free animals (tumor-free liver: 1520 ± 100 mg, tumor-free spleen: 200 ± 34 mg). Only few small nodes (1 mm and below) remained, which exhibited necrotic cores (Supplementary Figure S4 and Supplementary Materials and Methods online).

Figure 6.

Figure 6

Open in a new tab

Tumor necrosis factor alpha (TNFα)/Doxil chemo-gene therapy for the treatment of human colon carcinoma liver metastases. NMRI nu/nu mice were injected intrasplenically with luciferase marked human LS174T cells. Control animals (ctrl) received buffer solution i.v. Treated animals received LPEI-PEG-GE11 polyplexes with plasmid pEF1α-TNF or pΔEF1α-TNF (2.5 mg/kg) on day 3, 7, and 11, Doxil (1.5 mg/kg) was applied on day 5, 9, and 13 (n = 11), mice were sacrificed on day 15. *P < 0.05, **P < 0.01, and ***P < 0.001; AVOVA (Duncan). Average liver (a) and spleen (b) weight of NMRI nu/nu mice bearing LS174T tumor metastases after treatment. (c) luciferase activity in liver lysates as a measure of luciferase expressing LS174T tumor metastases. (d) Explanted livers and spleens of representative animals from a–c.

Discussion

TNFα treatment of cancer is still controversially discussed and can be seen as a double-edged sword, as the cytokine can in principle induce both, promotion of cancer growth and killing of cancer cells.14 Still, when delivered at appropriate concentration to the tumor site, efficient eradication of malignancies can be achieved.8 Here, we applied synthetic gene vectors to selectively express TNFα protein in tumor tissues. As nonviral gene delivery systems often suffer from inadequate transgene expression levels, we used an optimized plasmid to achieve high TNFα levels in tumor cells. We could recently show that combining the human EF1α promoter with the hCMV enhancer element leads to high and sustained transgene expression in liver15 or tumor.11 With TNFα as a transgene, its secretion peaked 2–3 days after transfection being far superior as compared to a conventional CMV driven plasmid pGSmuTNFα.16 Tumor cells strongly vary in their response to TNFα, and upregulation of NFκB,4,17 but also other factors play an important role in resistance toward TNFα induced cell killing.18 Figure 1b clearly points at the importance of this resistance, as all cancer cell lines tested did not exhibit a reduced viability after TNFα treatment. In contrast, for endothelial cells such resistance mechanisms have not been described yet, and TNFα can lead to both opening of the tight junctions,19 but also to apoptosis of endothelial cells after caspase activation.20 This should in principle allow also successful treatment of solid tumors, even where tumor cells per se are resistant towards TNFα treatment. Both, our in vitro permeability assay (Figure 1c) and histological studies on TNFα polyplex treated animals (Figure 2c) clearly show that both effects are also induced with transgenic TNFα. Polyplexes based on the potentially biodegradable polymer G3-HD-OEI have recently been shown to accumulate in subcutaneous tumors leading to tumor selective transgene expression.10 G3-HD-OEI bears a core of generation 3 polypropylenimine surface modified with low molecular weight branched polyethylenimine (800 Da) interlinked by the homo-bifunctional hydrophobic linker hexanedioldiacrylate. Such polyplexes are well tolerated and do not induce any increase in liver enzyme levels or erythrocyte aggregation up to a dose of at least 5 mg/kg.10 Selective transgene expression in tumor in this model is due to passive accumulation of the polyplexes in the hypervascularized Neuro2A tumor, an effect also observed with similar gene carriers.21 Depending on the organ, transgene expression is either nondetectable (brain and kidney) or >95% lower than in tumor (liver, heart, and lung; see ref. 10). In related studies, luciferase expression was found to last for at least 1 week after a single polyplex administration.11 Here we confirmed transgene activity in tumor tissue only for at least 1 week after pEF1α-Luc transfer (Figure 2a); thereafter animals had to be killed due to tumor burden. We also used these polyplexes for tumor selective TNFα expression (Figure 2b). Here, already control polyplexes alone led to elevated TNFα mRNA levels in liver and tumor. Nevertheless, when applying pEF1α-TNF polyplexes, TNFα mRNA levels in tumor were clearly higher as compared with all controls. Of note, also in the liver TNFα mRNA was increased pointing at the possibility that the polyplexes transfected liver tissue. However, quantification of mRNA for transgenic TNFα rather implies that most of the TNFα transgene expression occurs in the tumor. Apparently, in pEF1α-TNF polyplex treated animals only minute amounts of TNF can stimulate endogenous TNF expression in liver, a well-known positive feedback loop.3

The synergistic effect of TNFα protein and chemotherapy was observed with several chemotherapeutics in preclinical models and in tumor patients.8,9 Improved Doxil accumulation and extravasation led to a better therapeutic response.22 Hence, we investigated to which extent TNFα gene delivery augments Doxil accumulation in tumors (Figure 3). In the Neuro2A model, TNFα expression was sufficient to achieve up to 35-fold higher accumulation of doxorubicine, but with high variation between individual animals due to the heterogeneity in transgene expression, but also in vascularization. The relative increase in tumor accumulation is comparable to data obtained after pretreatment of mice with 50 µg/kg TNFα protein.22 Doxorubicine accumulation in cell nucleic within large areas of the tumor section points at its potential antitumoral activity, as not only the absolute amount of chemotherapeutic in the tumor, but also its distribution within the tissue is important for therapeutic efficacy.22 Of note, only in pEF1α-TNF treated animals doxorubicine accumulation was increased and not in pEF1α-Luc treated ones. This clearly demonstrates that the induction of endogenous TNFα expression in this setting is not sufficient to induce a significant effect. There was a slight increase in liver concentration of TNFα/Doxil treated animals, which could be due to minute amounts of TNFα being released into circulation. In the HUH7 hepatoma model, passive accumulation of G3-HD-OEI TNFα gene carriers were not able to induce TNFα expression levels sufficient for improved Doxil accumulation. This is in line with our previous observations, where in contrast to the Neuro2A model, luciferase transgene levels were significantly lower in HUH7 tumors using untargeted polyplexes.11 Therefore we applied EGFR targeted polyplexes, as EGFR is overexpressed in hepatocellular carcinoma and other tumors.23 The synthetic gene carrier based on linear PEI with the peptide GE11 as EGFR ligand (LPEI-PEG-GE11) was already successfully applied in this model for a combined radio-gene therapy expressing the sodium iodide symporter (NIS) and treatment with radioiodine.24 Unlike recombinant EGF, this synthetic peptide GE11 evades activation of the EGFR-cascade upon EGFR interaction.13,25

Our combination therapy approach aimed at maximizing the specific TNFα effect and its synergy with Doxil. For this purpose, we only applied three polyplex doses at 2.5 mg/kg containing plasmids completely devoid of potentially immune stimulatory, unmethylated CpG sequences. Previous studies already demonstrated that nonviral gene carriers delivering TNFα plasmid are able to induce growth retardation of subcutaneous tumors on their own.16,26 In these studies polyplexes were repeatedly applied (8–10 times) with cumulative dosage reaching up to 25 mg/kg of plasmid and up to 250 mg/kg of polymeric carrier.26 Similar observations have been made with gene carriers using noncoding plasmid or polycations alone leading to intrinsic antitumoral activity by inducing the expression of endogenous TNFα expression.27,28 Hence, the antitumoral effect can be attributed to the combined action of transgenic and endogenous TNFα. Furthermore, in humans already a total dose of 0.03 mg/kg of plasmid containing unmethylated CpG sequences induced elevated levels of inflammatory cytokines and dose limiting side effects, like fever and hypotension.29 Therefore it is of utmost importance that a systemically applied plasmid vector should be applied in CpG reduced form at the lowest possible dosage to reduce inflammatory responses due to the activation of the innate immune response.30

As expected, only moderate growth retardation was seen with TNFα encoding polyplex alone compared to the luciferase control group (Figure 4a). In sharp contrast, the combination with Doxil almost stopped tumor growth for ~2 weeks when combined with TNFα polyplex, whereas the luciferase control was almost unaffected. Also after tumor re-growth pEF1α-TNF/Doxil treatment was effective and significantly retarded tumor growth again, which points at the absence of resistance mechanisms. With the same schedule, Neuro2A tumors were also successfully treated in an experimental metastasis model (Figure 5). Although described to develop mainly liver metastases after intravenous injection,31 we also observed tumor nodules in various places of the abdominal region, but also in the head. Tumors nodules were well vascularized, similar as in the subcutaneous setting and were accessible for G3-HD-OEI polyplexes mediated transfection with luciferase (Farkasova et al, manuscript in preparation). This, and the fact that transgenic TNFα is specifically expressed in tumor tissue (Figure 2) leads us to the conclusion that the therapeutic effect on reduced tumor burden is due to TNFα transgene expression within tumor nodules tissue synergizing with Doxil.

The EGFR, overexpressed in 21–77% of colorectal cancers, represents a valid target for targeted gene delivery.23 EGFR overexpressing LS174T human Colorectal cancer cells develop highly aggressive liver metastases several days after intrasplenic injection, and only 25% of metastases respond to doxorubicine treatment at a dose of 10 mg/kg.32 Access for macromolecular drugs to LS174T metastases is also limited: with a CEA antibody, diffusion into larger tumor nodules was hampered, and the maximal diffusion path length achieved after 24 hours was on average 400 µm.33 Hence, diffusion of 100 nm liposomes like Doxil is expected to be inefficient. Our TNFα/Doxil combination therapy applied here demonstrated that TNFα enables efficient treatment with low Doxil doses allowing the liposomes to access the tumor metastases. Of interest, liver and spleen weight in pΔEF1α-TNF/Doxil treated, LS174 tumor bearing animals was significantly decreased, but there was no reduction observed in terms of luciferase signal in the liver. We conclude from these data that the Doxil treatment alone was not sufficient to reduce the viable tumor mass in liver (e.g., due to restricted access), but rather caused toxic side effects in the liver leading to reduced liver weight. This is a known, dose dependent side effect of doxorubicine treatment in rodent models,34 which could be even potentiated in case the liver is already burdened by tumor lesions.

Taken together, this study demonstrates for the first time enhanced tumoral accumulation and antitumoral effects of macromolecular chemotherapeutics after systemic pre-treatment with a tumor targeted, synthetic TNFα gene carrier. Both, in a murine neuroblastoma model and xenografted human colon carcinoma a clear beneficial effect of pEF1α-TNF is observed. In contrast to TNFα protein therapy, this treatment can be applied systemically without side effects reaching effective TNFα doses in tumors. Thus, the concept may be applied for other macromolecular drugs and gene carriers, setting ground for new anticancer therapies.

Materials and Methods

Conjugates, plasmids. The synthesis of polymers G2-HD-OEI and G3-HD-OEI has been described previously.10 In brief, polypropylenimine dendrimer generation 2 (G2) or generation 3 (G3) were modified via 1,6-hexandioldiacrylate with branched oligoethylenimine 800 Da (OEI). The EGFR targeting conjugate LPEI-PEG-GE11 was synthesized by coupling the peptide GE11 (CYHWYGYTPQNVI) via a heterobifunctional, 2 kDa PEG linker NHS-PEG-OPSS as recently described.13 Murine TNFα cDNA based on the published sequence (nucleotide ID NM_013693.2) was synthesized by Geneart (Regensburg, Germany). The sequence was designed for optimized codon usage in mouse; cryptic splicing sites were removed. TNFα cDNA was obtained within the vector pMA flanked by restriction sites for BglII and NheI. The luciferase gene was excised from pCpG-hCMV-EF1α-Luc plasmid11 (short: pEF1α-Luc) by BglII-NheI double restriction enzymes digestion and the vector ligated with the TNFα cDNA excised from pMA-TNFα plasmid with the same enzymes to construct pCpG-hCMV-EF1α-TNFα (short: pEF1α-TNF). pΔEF1α-TNF was generated by deleting the EF1α promoter of pEF1α-TNF.

Cell lines. Murine Neuro2A neuroblastoma cells (ATCC CCL-131) were propagated in Dulbecco's modified Eagle medium, murine L929 fibroblasts (ATCC CCL-1), U87MG human glioblastoma (ATCC HTB 14) and LS174T human colon adenocarcinoma (ATCC CCL188) were cultured in RPMI-1640, human hepatocellular carcinoma HUH7 (JCRB 0403, Tokyo, Japan) in Ham's F-12 complete medium, all supplemented with 10% fetal bovine serum (FBS), 2 mmol/l L-glutamine, 100 U/ml penicillin and 100 mg/ml streptomycin. BT549 human breast carcinoma (CLS, #300132) were grown in RPMI 1640 with 0.023 IU/ml insulin and 10% FBS, MDA-MB-231 human breast carcinoma (kindly provided by Prof. A. Ullrich, Max-Planck Institute, Munich, Germany) in Dulbecco's modified Eagle medium: Ham's F12 (1:1) supplemented with 2 mmol/l L-glutamine and 5% FBS, MDA-MB-453 human breast carcinoma (kindly provided by Prof. A. Ullrich) in Dulbecco's modified Eagle medium/Ham's F-12 (1:1); all culture media were supplemented with 10% FBS, 2 mmol/l stable glutamine and 1% penicillin and streptomycin. Primary porcine endothelial cells were isolated from explants of porcine aortic vessels and cultured in M199 medium supplemented with 10% FBS, 10 ng/ml basic fibroblast growth factor, 100 U/ml penicillin and 100 mg/ml streptomycin.

Lentiviral transductions. Neuro2A or LS174T were seeded in T25 cell culture flasks at a density of 100,000 cells 24 hours before transduction with a lentiviral vector encoding a PGK-EGFPLuc cassette. The EGFP-Luc cDNA was excised from the vector pEGFPLuc (Clontech, Mountain View, CA) and cloned into the vector pHIV7/SF-GFP to replace EGFP.35 Lentiviral vectors were produced as described.36 Transduction was performed with 1 ml virus supernatant in the presence of 8 µg/ml polybrene, 24 hours thereafter cells were transferred into T75 flask. After two rounds of splitting, a p24 enzyme-linked immunosorbent assay was performed to ensure the absence of any lentiviral particles. Transduced cells were sorted with a MoFlo cell sorter (Beckman Coulter, Krefeld, Germany) to obtain a homogenous cell population of EGFP-positive cells. Resulting populations are named Neuro2A-lenti-luc and LS174T-lenti-Luc, respectively.

Other materials and kits. Doxil (Liposomal Doxorubicin) was purchased from Schering-Plough (Kenilworth, NJ), transwell polycarbonate membrane inserts (pore size 0.4 µm, membrane diameter 6.5 mm) were purchased from Corning B.V. Life Sciences (Amsterdam, The Netherlands), mouse TNFα enzyme-linked immunosorbent assay Kit from eBioscience (San Diego, CA). All other materials were purchased from Sigma (Munich, Germany).

Preparation of transfection polyplexes. G3-HD-OEI/pDNA and G2-HD-OEI/pDNA polyplexes were prepared as described.10 For in vitro, pDNA (40 ug/ml in 20 mmol/l HEPES pH 7.4, 5% glucose w/v, termed HEPES buffered glucose) was flash-mixed with an equal volume of polymer (40 µg/ml in HEPES buffered glucose). For in vivo application, plasmid and polymer concentrations were tenfold higher. LPEI-PEG-GE11 polyplexes were mixed in a similar way, but with an N/P ratio (molar ratio nitrogen in LPEI to phosphate in pDNA) of 6.13

In vitro cell transfections. Indicated tumor cells were seeded at a density of 7 × 104 cells/well in a 24-well tissue culture plate and after overnight culture transfected with G3-HD-OEI/DNA polyplexes containing 1µg pDNA/well in 500 µl fetal calf serum supplemented medium, which was replaced after 4 hours by fresh medium. Supernatants were collected every 24 hours and stored frozen at −80°C until further analysis.

L929 bioassay. L929 cytotoxicity was analyzed by modifying the method from Shiau et al.37 In brief, 1.25 × 104 L929 murine fibroblasts were seeded per well in a 96-well tissue culture plate and after overnight culture the medium replaced by 100 µl of supernatant collected from transfected tumor cells at different time points as described above. After exposure for 24 hours in the presence of 1µg/ml Actinomycin D, mitochondrial activity as a measure for cell viability was measured by MTT assay. Relative cytotoxicity was calculated by the formula: % cytotoxicity = 100x (Absorbance of untreated cells − absorbance of treated cells) / absorbance of untreated cells.

Endothelial permeability assay. 2 × 104 primary porcine endothelial cells were seeded into the collagen A coated inner chamber of a transwell system and cultivated for 72 hours until complete confluency was reached. Then, the medium in the inner chamber was exchanged for cell free supernatant of transfected Neuro2A cells or medium containing 1 ng/ml TNFα protein together with 1 µg/ml Actinomycin D. After further incubation for 24 hours, FITC-dextran (40 kDa) was added at a final concentration of 1mg/ml to the inner chamber. After 50 min of incubation at 37 °C the medium in both chambers was collected and fluorescence measured in a Cary Eclipse fluorescence spectrometer (Agilent, München, Germany) at Ex = 490 nm, Em = 530 nm and compared with a standard curve prepared with defined concentrations of FITC-dextran diluted in medium. The relative transfer was calculated by the formula % transfer = 100× concentration FITC-dextran in outer chamber/concentration of FITC-dextran added.

Tumor models. All animal procedures were approved and controlled by the local ethics committee and carried out according to the guidelines of the German law of protection of animal life. Animals were kept in individually vented cages with a 12-hour day/night cycle; water and chow were provided ad libitum. For subcutaneous Neuro2A tumors, 6-week-old female A/J mice (Harlan-Winkelmann, Borchen, Germany) were injected subcutaneously into the flank with 1 × 106 Neuro2A cells resuspended in 100 µl PBS. For subcutaneous HUH7 tumors, 6-week-old female NMRI nu/nu mice (Janvier, Le Genest-Saint-Isle, France) were injected subcutaneously with 3 × 106 HUH7 cells resuspended in 100 µl PBS. Size of subcutaneous tumors was monitored by caliper measurements in three perpendicular directions (a,b, and c) and the tumor volume (in mm3) calculated with the formula a × b × c × 0.4. Tumors were clearly palpable with a minimal size of 2.5 mm, which then reliably led to a growing tumor mass (100% take rate). For the Neuro2A metastasis model, A/J mice were injected intravenously with 1 × 106 Neuro2A-lenti-luc cells in 100 µl PBS. For the liver metastasis model of human LS174T-lenti-luc human colon carcinoma, 7.5 × 105 cells resuspended in 50 µl PBS were injected into the spleen of 6-week-old female NMRI nu/nu mice analogously as described.32 Growth of metastases was analyzed by BLI (see below).

BLI and quantification of luciferase activity. BLI measurement of luciferase activity was carried out as recently described.11 In brief, mice were anaesthetized with isofluorane in oxygen and intraperitoneally injected with 6 mg Na-Luciferin (Promega, Hilden, Germany) dissolved in 100 µl of PBS. Ten minutes thereafter BLI was carried out in an IVIS lumina imaging system (Caliper Life Sciences GmbH, Rüsselsheim Germany). For quantification of the BLI signal, a region of interest was laid covering the analyzed area and the signal measured as emitted photons/second. Background values of similar sized region of interest, but on luciferase negative areas never exceeded 4% as compared with the BLI signal in the tumor area. To quantify luciferase activity in tissue, indicated organs were snap frozen on dry ice, homogenized with mortar and pestle cooled with liquid nitrogen and aliquots of the powder lysed in Promega cell lysis buffer (Promega). Further signal quantification was carried out as described.11 Background signal (100–200 RLU) was subtracted from each value and transfection efficacy expressed as relative light units (RLU) per mg protein. Protein content was determined by a modified BCA assay as described.10 One million RLU correspond to 2 ng of recombinant luciferase (Promega).

Determination of TNF mRNA by qPCR. For RNA isolation the high pure RNA Tissue Kit (Roche, Mannheim, Germany) was used according to the manufacturer's instructions; cDNA was obtained using the Transcriptor High Fidelity cDNA synthesis kit (Roche). Samples were measured in parallel from the same reverse transcriptase reaction, each done in duplicate per sample on a Roche LightCycler 480 system. Total TNFα was quantified using the LightCycler 480 Probes Master with UPL Probes (Roche) and normalized to the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase. TNFα forward primer: GCTTCCAGAACAGCAGACG, reverse primer: AGGCAGAACAGGGTGGTG, probe # 98; glyceraldehyde-3-phosphate dehydrogenase forward primer: AGCTTGTCATCAACGGGAAG, reverse primer: TTTGATGTTAGTGGGGTCTCG, probe # 9. To distinguish between endogenous and transgenic TNFα mRNA, primer were designed which can only amplify the exogenous pEF1α-TNF mRNA transcript. Analysis was carried out with the SYBR green kit according to the manufacturer's instructions. Forward primer: AAGCTCTAGCCAGAATAGCTCC; reverse primer: ACAGCCACTCCAGCTGCT. cDNA from untreated samples was used as calibrator.

Immunohistology. Five µm cryosections were fixed in ice-cold methanol for 5 min, blocked with 1% BSA in PBS for 15 minutes, with 10% goat serum in PBS for 15 minutes and subsequently incubated with rat antimouse CD31 mAb (eBioscience, San Diego, CA, dilution 1:200) and Alexa Fluor 647 goat anti-rat immunoglobulin G (eBioscience, dilution 1:200). Nuclei were counterstained with DAPI and slides mounted with Vectashield mounting media (Vector Laboratories, Burlingame, CA). Slides were evaluated on a Zeiss Axiovert 200 microscope (Carl Zeiss, Jena, Germany) using a 10× objective. CD31 fluorescence was quantified by image analysis using Image J software (http://rsbweb.nih.gov/ij/) calculating the relative CD31 positive area (n = 2). The parameter “total number of positive pixels per field” was analyzed for each image. For studying Doxil distribution, 5 µm cryosectionswere fixed in ice-cold methanol for 5 minutes, covered with a cover slide and fluorescence observed using a 10× objective and the Cy3 filter set (Carl Zeiss).

In vivo application of polyplexes and Doxil. Polyplexes with a final DNA concentration of 200 µg/ml in HEPES buffered glucose buffer were injected into the tail vein at a pDNA dose of 2.5 mg/kg (corresponding to 250 µl polyplex solution for a 20 g mouse). For accumulation and imaging experiments, Doxil was diluted in sterile PBS to a concentration of 400 µg/ml and applied intravenously at a dose of 5 mg/kg per application; for therapeutic studies Doxil (120 µg/ml) was applied at a dose of 1.5 mg/kg per application.

Quantification of doxorubicine in tissue. Doxorubicine was quantified by a fluorescence based assay as described.38 In brief, explanted tissues were incubated in acidified isopropanol (0.075 mol/l HCl in 90% isopropanol) for 24 hours at 4 °C, thereafter homogenized, centrifuged for 30 minutes at 4000 rpm and fluorescence in supernatants measured with a Varian Cary Eclipse fluorescence spectrometer (Agilent, Böblingen, Germany), Ex = 480 nm, Em = 590 nm. A standard curve was prepared with defined concentrations of Doxil diluted in acidified isopropanol.

Statistical methods. Data were analyzed with WinSTAT for Microsoft Excel (R. Fitch Software, Bad Krozingen, Germany) or GraphPad Prism (San Diego, CA) using Analysis of Variance (ANOVA, method according to Duncan) or t-test (independent). P values <0.05 were considered as significant.

SUPPLEMENTARY MATERIAL Figure S1. Polyplex mediated expression of TNFα in vitro with different plasmids. Figure S2. Luciferase activity in tissue lysates after G3-HD-OEI polyplex treatment. Figure S3. Influence of TNFα transgene expression on body distribution of Doxil. Figure S4. Histology of LS174T tumors after polyplex/Doxil treatment. Supplementary Materials and Methods.

Acknowledgments

This work was supported by the DFG with grant no. OG 63/4-1 (M.O.) and the Nanosystems Initiatve Munich (E.W.). B.S. received a personal scholarship from the China Scholarship Council (CSC) program. J.V. received a postdoctoral fellowship from the Bavarian Research Foundation (PDOC-78-11).

Supplementary Material

Figure S1.

Polyplex mediated expression of TNFα in vitro with different plasmids.

Click here for additional data file. (91.8KB, pdf)
Figure S2.

Luciferase activity in tissue lysates after G3-HD-OEI polyplex treatment.

Click here for additional data file. (108KB, pdf)
Figure S3.

Influence of TNFα transgene expression on body distribution of Doxil.

Click here for additional data file. (650.5KB, pdf)
Figure S4.

Histology of LS174T tumors after polyplex/Doxil treatment.

Click here for additional data file. (5.3MB, pdf)
Supplementary Materials and Methods.
Click here for additional data file. (173.6KB, pdf)

REFERENCES

  1. Dickson PV, Hamner JB, Sims TL, Fraga CH, Ng CY, Rajasekeran S.et al. (2007Bevacizumab-induced transient remodeling of the vasculature in neuroblastoma xenografts results in improved delivery and efficacy of systemically administered chemotherapy Clin Cancer Res 133942–3950. [DOI] [PubMed] [Google Scholar]
  2. Bilbao R, Bustos M, Alzuguren P, Pajares MJ, Drozdzik M, Qian C.et al. (2000A blood-tumor barrier limits gene transfer to experimental liver cancer: the effect of vasoactive compounds Gene Ther 71824–1832. [DOI] [PubMed] [Google Scholar]
  3. Balkwill F. Tumour necrosis factor and cancer. Nat Rev Cancer. 2009;9:361–371. doi: 10.1038/nrc2628. [DOI] [PubMed] [Google Scholar]
  4. Giri DK., and, Aggarwal BB. Constitutive activation of NF-kappaB causes resistance to apoptosis in human cutaneous T cell lymphoma HuT-78 cells. Autocrine role of tumor necrosis factor and reactive oxygen intermediates. J Biol Chem. 1998;273:14008–14014. doi: 10.1074/jbc.273.22.14008. [DOI] [PubMed] [Google Scholar]
  5. von Holzen U, Bocangel D, Pataer A, Vorburger SA, Liu Y, Lu X.et al. (2005Role for the double-stranded RNA-activated protein kinase PKR in Ad-TNF-alpha gene therapy in esophageal cancer Surgery 138261–268. [DOI] [PubMed] [Google Scholar]
  6. Calcinotto A, Grioni M, Jachetti E, Curnis F, Mondino A, Parmiani G.et al. (2012Targeting TNF-α to neoangiogenic vessels enhances lymphocyte infiltration in tumors and increases the therapeutic potential of immunotherapy J Immunol 1882687–2694. [DOI] [PubMed] [Google Scholar]
  7. Selby P, Hobbs S, Viner C, Jackson E, Jones A, Newell D.et al. (1987Tumour necrosis factor in man: clinical and biological observations Br J Cancer 56803–808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Eggermont AM, Schraffordt Koops H, Liénard D, Kroon BB, van Geel AN, Hoekstra HJ.et al. (1996Isolated limb perfusion with high-dose tumor necrosis factor-alpha in combination with interferon-gamma and melphalan for nonresectable extremity soft tissue sarcomas: a multicenter trial J Clin Oncol 142653–2665. [DOI] [PubMed] [Google Scholar]
  9. Murugesan SR, King CR, Osborn R, Fairweather WR, O'Reilly EM, Thornton MO.et al. (2009Combination of human tumor necrosis factor-alpha (hTNF-alpha) gene delivery with gemcitabine is effective in models of pancreatic cancer Cancer Gene Ther 16841–847. [DOI] [PubMed] [Google Scholar]
  10. Russ V, Günther M, Halama A, Ogris M., and, Wagner E. Oligoethylenimine-grafted polypropylenimine dendrimers as degradable and biocompatible synthetic vectors for gene delivery. J Control Release. 2008;132:131–140. doi: 10.1016/j.jconrel.2008.09.003. [DOI] [PubMed] [Google Scholar]
  11. Navarro G, Maiwald G, Haase R, Rogach AL, Wagner E, de Ilarduya CT.et al. (2010Low generation PAMAM dendrimer and CpG free plasmids allow targeted and extended transgene expression in tumors after systemic delivery J Control Release 14699–105. [DOI] [PubMed] [Google Scholar]
  12. Branch DR, Shah A., and, Guilbert LJ. A specific and reliable bioassay for the detection of femtomolar levels of human and murine tumor necrosis factors. J Immunol Methods. 1991;143:251–261. doi: 10.1016/0022-1759(91)90050-p. [DOI] [PubMed] [Google Scholar]
  13. Schäfer A, Pahnke A, Schaffert D, van Weerden WM, de Ridder CM, Rödl W.et al. (2011Disconnecting the yin and yang relation of epidermal growth factor receptor (EGFR)-mediated delivery: a fully synthetic, EGFR-targeted gene transfer system avoiding receptor activation Hum Gene Ther 221463–1473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Aggarwal BB. Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol. 2003;3:745–756. doi: 10.1038/nri1184. [DOI] [PubMed] [Google Scholar]
  15. Magnusson T, Haase R, Schleef M, Wagner E., and, Ogris M. Sustained, high transgene expression in liver with plasmid vectors using optimized promoter-enhancer combinations. J Gene Med. 2011;13:382–391. doi: 10.1002/jgm.1585. [DOI] [PubMed] [Google Scholar]
  16. Kircheis R, Ostermann E, Wolschek MF, Lichtenberger C, Magin-Lachmann C, Wightman L.et al. (2002Tumor-targeted gene delivery of tumor necrosis factor-alpha induces tumor necrosis and tumor regression without systemic toxicity Cancer Gene Ther 9673–680. [DOI] [PubMed] [Google Scholar]
  17. Sumitomo M, Tachibana M, Nakashima J, Murai M, Miyajima A, Kimura F.et al. (1999An essential role for nuclear factor kappa B in preventing TNF-alpha-induced cell death in prostate cancer cells J Urol 161674–679. [PubMed] [Google Scholar]
  18. Baritaki S., and, Bonavida B. Viral infection and cancer: the NF-kappaB/Snail/RKIP loop regulates target cell sensitivity to apoptosis by cytotoxic lymphocytes. Crit Rev Immunol. 2010;30:31–46. doi: 10.1615/critrevimmunol.v30.i1.20. [DOI] [PubMed] [Google Scholar]
  19. Vandenbroucke E, Mehta D, Minshall R., and, Malik AB. Regulation of endothelial junctional permeability. Ann N Y Acad Sci. 2008;1123:134–145. doi: 10.1196/annals.1420.016. [DOI] [PubMed] [Google Scholar]
  20. Madge LA., and, Pober JS. TNF signaling in vascular endothelial cells. Exp Mol Pathol. 2001;70:317–325. doi: 10.1006/exmp.2001.2368. [DOI] [PubMed] [Google Scholar]
  21. Klutz K, Russ V, Willhauck MJ, Wunderlich N, Zach C, Gildehaus FJ.et al. (2009Targeted radioiodine therapy of neuroblastoma tumors following systemic nonviral delivery of the sodium iodide symporter gene Clin Cancer Res 156079–6086. [DOI] [PubMed] [Google Scholar]
  22. Seynhaeve AL, Hoving S, Schipper D, Vermeulen CE, de Wiel-Ambagtsheer G, van Tiel ST.et al. (2007Tumor necrosis factor alpha mediates homogeneous distribution of liposomes in murine melanoma that contributes to a better tumor response Cancer Res 679455–9462. [DOI] [PubMed] [Google Scholar]
  23. Khalil MY, Grandis JR., and, Shin DM. Targeting epidermal growth factor receptor: novel therapeutics in the management of cancer. Expert Rev Anticancer Ther. 2003;3:367–380. doi: 10.1586/14737140.3.3.367. [DOI] [PubMed] [Google Scholar]
  24. Klutz K, Schaffert D, Willhauck MJ, Grünwald GK, Haase R, Wunderlich N.et al. (2011Epidermal growth factor receptor-targeted (131)I-therapy of liver cancer following systemic delivery of the sodium iodide symporter gene Mol Ther 19676–685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Li Z, Zhao R, Wu X, Sun Y, Yao M, Li J.et al. (2005Identification and characterization of a novel peptide ligand of epidermal growth factor receptor for targeted delivery of therapeutics FASEB J 191978–1985. [DOI] [PubMed] [Google Scholar]
  26. Koppu S, Oh YJ, Edrada-Ebel R, Blatchford DR, Tetley L, Tate RJ.et al. (2010Tumor regression after systemic administration of a novel tumor-targeted gene delivery system carrying a therapeutic plasmid DNA J Control Release 143215–221. [DOI] [PubMed] [Google Scholar]
  27. Dufès C, Keith WN, Bilsland A, Proutski I, Uchegbu IF., and, Schätzlein AG. Synthetic anticancer gene medicine exploits intrinsic antitumor activity of cationic vector to cure established tumors. Cancer Res. 2005;65:8079–8084. doi: 10.1158/0008-5472.CAN-04-4402. [DOI] [PubMed] [Google Scholar]
  28. Whitmore M, Li S., and, Huang L. LPD lipopolyplex initiates a potent cytokine response and inhibits tumor growth. Gene Ther. 1999;6:1867–1875. doi: 10.1038/sj.gt.3301026. [DOI] [PubMed] [Google Scholar]
  29. Maclachlan I., and, Judge A. Protiva Biotherapeutics: Burnaby, CA; 2007. Glucocorticoid Modulation of Nucleic Acid-Mediated Immune Stimulation. [Google Scholar]
  30. Yew NS, Zhao H, Wu IH, Song A, Tousignant JD, Przybylska M.et al. (2000Reduced inflammatory response to plasmid DNA vectors by elimination and inhibition of immunostimulatory CpG motifs Mol Ther 1255–262. [DOI] [PubMed] [Google Scholar]
  31. Amirkhosravi A, Warnes G, Biggerstaff J, Malik Z, May K., and, Francis JL. The effect of pentoxifylline on spontaneous and experimental metastasis of the mouse Neuro2a neuroblastoma. Clin Exp Metastasis. 1997;15:453–461. doi: 10.1023/a:1018414624544. [DOI] [PubMed] [Google Scholar]
  32. Zirvi KA, Najjar TA., and, Slomiany BL. Sensitivity of human colon tumor metastases to anticancer drugs in athymic (nude) mice. Cancer Lett. 1993;72:39–44. doi: 10.1016/0304-3835(93)90008-w. [DOI] [PubMed] [Google Scholar]
  33. Fidarova EF, El-Emir E, Boxer GM, Qureshi U, Dearling JL, Robson MP.et al. (2008Microdistribution of targeted, fluorescently labeled anti-carcinoembryonic antigen antibody in metastatic colorectal cancer: implications for radioimmunotherapy Clin Cancer Res 142639–2646. [DOI] [PubMed] [Google Scholar]
  34. Zhao X, Zhang J, Tong N, Chen Y., and, Luo Y. Protective effects of berberine on doxorubicin-induced hepatotoxicity in mice. Biol Pharm Bull. 2012;35:796–800. doi: 10.1248/bpb.35.796. [DOI] [PubMed] [Google Scholar]
  35. Yam PY, Li S, Wu J, Hu J, Zaia JA., and, Yee JK. Design of HIV vectors for efficient gene delivery into human hematopoietic cells. Mol Ther. 2002;5:479–484. doi: 10.1006/mthe.2002.0558. [DOI] [PubMed] [Google Scholar]
  36. Wübbenhorst D, Dumler K, Wagner B, Wexel G, Imhoff A, Gansbacher B.et al. (2010Tetracycline-regulated bone morphogenetic protein 2 gene expression in lentivirally transduced primary rabbit chondrocytes for treatment of cartilage defects Arthritis Rheum 622037–2046. [DOI] [PubMed] [Google Scholar]
  37. Shiau MY, Chiou HL, Lee YL, Kuo TM., and, Chang YH. Establishment of a consistent L929 bioassay system for TNF-alpha quantitation to evaluate the effect of lipopolysaccharide, phytomitogens and cytodifferentiation agents on cytotoxicity of TNF-alpha secreted by adherent human mononuclear cells. Mediators Inflamm. 2001;10:199–208. doi: 10.1080/09629350123139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ten Hagen TL, Van Der Veen AH, Nooijen PT, Van Tiel ST, Seynhaeve AL., and, Eggermont AM. Low-dose tumor necrosis factor-alpha augments antitumor activity of stealth liposomal doxorubicin (DOXIL) in soft tissue sarcoma-bearing rats. Int J Cancer. 2000;87:829–837. doi: 10.1002/1097-0215(20000915)87:6<829::aid-ijc12>3.0.co;2-c. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1.

Polyplex mediated expression of TNFα in vitro with different plasmids.

Click here for additional data file. (91.8KB, pdf)
Figure S2.

Luciferase activity in tissue lysates after G3-HD-OEI polyplex treatment.

Click here for additional data file. (108KB, pdf)
Figure S3.

Influence of TNFα transgene expression on body distribution of Doxil.

Click here for additional data file. (650.5KB, pdf)
Figure S4.

Histology of LS174T tumors after polyplex/Doxil treatment.

Click here for additional data file. (5.3MB, pdf)
Supplementary Materials and Methods.
Click here for additional data file. (173.6KB, pdf)

Articles from Molecular Therapy are provided here courtesy of The American Society of Gene & Cell Therapy

RESOURCES