Chimeric Form of Tumor Necrosis Factor-A Has Enhanced Surface

Home , Tumor necrosis factor

ORIGINAL ARTICLE Chimeric form of tumor necrosis factor-a has enhanced surface expression and antitumor activity R Rieger1, D Whitacre1, MJ Cantwell1,3, C Prussak2 and TJ Kipps1 1Department of Medicine, Moores Cancer Center, University of California, San Diego, La Jolla, CA, USA and 2Memgen, LLC, Solana Beach, CA, USA

Tumor necrosis factor (TNF)-a is a type-II transmembrane protein that is cleaved by TNF-a-converting enzyme (TACE/ADAM-17) to release soluble TNF, a cytokine with potent antitumor properties whose use in clinical applications is limited by its severe systemic toxicity. We found that human cells transfected with vectors encoding TNF without the TACE cleavage site (DTACE-TNF) still released functional cytokine at substantial levels that varied between transfected cell lines of different tissue types. Vectors encoding membrane-associated domains of CD154, another TNF-family protein, conjoined with the carboxyl-terminal domain of TNF, directed higher-level surface expression of a functional TNF that, in contrast to DTACE-TNF, was resistant to cleavage in all cell types. Furthermore, adenovirus vectors encoding CD154-TNF had significantly greater in vivo biological activity in inducing regression of established, syngeneic tumors in mice than adenovirus vectors encoding TNF, and lacked toxicity associated with soluble TNF. As such, CD154-TNF is a novel TNF that appears superior for treatment of tumors in which high-level local expression of TNF is desired. Cancer Gene Therapy (2009) 16, 53–64; doi:10.1038/cgt.2008.57; published online 25 July 2008 Keywords: tumor; necrosis; factor; immunotherapy; adenovirus; CD154

9–11 Introduction from that of secreted TNF in vitro and in vivo. However, cells expressing such truncated forms of TNF Tumor necrosis factor (TNF)-a was initially described lacking the TACE cleavage site still could release 9 as a 17-kDa soluble cytokine that could induce local active TNF. Moreover, enzymes other than TACE could and systemic tumor regression.1 However, clinical trials cleave the 26-kDa TNF proform at sites other than that 12 revealed that the soluble cytokine had a low therapeutic recognized by TACE, to release a soluble TNF. As such, index, owing to its toxicity when present at therapeutic the interpretation that membrane TNF has biological concentrations in the systemic circulation.2–4 activity distinct from that of soluble TNF has to be Soluble TNF is released from the surface of tempered in light of the fact that soluble TNF still might cells expressing the 26-kDa proform of TNF through be released in the microenvironment of cells expressing the TNF-a-converting enzyme (TACE/ADAM-17),5,6 a such modified forms of TNF. matrix metalloproteinase that recognizes a cleavage site To address the hypothesis that membrane-bound TNF in the extracellular domain of the full-length TNF elicits a biological response distinct from that of the molecule.7 Deletion of this cleavage site results in soluble molecule, we generated chimeric forms of TNF the generation of a TNF molecule that has enhanced that are highly resistant to release from the cellular membrane stability.7–9 Mice made transgenic for this membrane. This was accomplished by fusing the active membrane-stabilized form of TNF produced significantly ligand-binding portion of the TNF molecule with the higher levels of interleukin-12 than wild-type mice or transmembrane and proximal extracellular domains of TNF-deficient mice after administration of lipopoly- TNF family proteins that primarily are expressed as cell- saccharide, suggesting that the transmembrane form of surface molecules (for example, CD154 or CD70). We TNF might affect cellular immune responses in vivo and generated adenovirus vectors encoding wild-type TNF have quantitatively and qualitatively distinct functions (Ad-wtTNF), TNF lacking the defined TACE site (Ad- DTACE-TNF) and chimeric forms of TNF with CD154 (Ad-CD154-TNF) or CD70 (Ad-CD70-TNF), and ex- Correspondence: Dr TJ Kipps, Moores UCSD Cancer Center, amined cells infected with such constructs for expression University of California, San Diego School of Medicine, 3855 Health Sciences Drive no. 0820, La Jolla, CA 92093-0820, USA. of functional cell-surface TNF relative to that of the E-mail: [email protected] soluble cytokine. Furthermore, we compared the ability 3Current address: Adventrx Pharmaceuticals, San Diego, CA, USA of CD154-TNF versus wtTNF to induce an antitumor Received 20 January 2008; revised 22 May 2008; accepted 3 June response against syngeneic fibrosarcoma and lymphoma 2008; published online 25 July 2008 in vivo. Antitumor activity of membrane-anchored TNF R Rieger et al 54 Materials and methods ATTCACAGGGCGATGATTCCAAAG. The putative Cell lines and mice matrix metalloproteinase cleavage site of CD154 was then HeLa (human cervical carcinoma), HT1080 (human deleted from this construct using Stratagene’s Quick- fibrosarcoma), A549 (human lung), HCT-15 and COLO- Change site-directed mutagenesis kit according to the 205 (human colon), DU-145 (human prostate), RPMI- manufacturer’s instructions with the following primers: 50-CAAAGAAGAGAAAAAAGAAGATGAGGATCCT 8226 (human myeloma), HT1376 (human bladder), L929 0 (mouse fibroblast), WEHI-164 (mouse fibrosarcoma) and GTAGCCC and 5 -GGGCTACAGGATCCTCATCTTC A20 (mouse B cell lymphoma) were obtained from ATCC, TTTTTTCTCTTCTTTG. Manassas, VA. The cell line 293AC2, a subclone of 293, The DTACE-TNF constructs were created by deletion was obtained from Molecular Medicine Inc. (San Diego, of the nucleotides corresponding to the first 12 amino CA). All cell lines except for 293AC2 were cultured in acids of the mature TNF, that is, amino acids 77–88 of RPMI-1640 containing 10% fetal calf serum (FCS) and the full-length human and 80–91 of the mouse protein. Using the QuickChange site-directed mutagenesis kit 2mML-glutamine and maintained in a 5% CO incubator. 2 (Stratagene, San Diego, CA), we deleted the nucleotide 293AC2 cells were cultured in DMEM containing 10% 0 ML sequence 5 -GTCAGATCATCTTCTCGAACCCCGAG FCS and 2 m -glutamine and maintained in a 10% CO2 0 incubator. TGACAAGCCT or 5 -CTCAGATCATCTTCTCAAAA Female BALB/c and FVB/N mice, 6–8 weeks of TTCGAGTGACAAGCCT, from the human or mouse age, were purchased from Charles River Laboratories TNF coding sequences, respectively. All clones were (Wilmington, MA) and housed at the University of validated by nucleic acid sequence analysis. The proteins California, San Diego (UCSD) animal facility. All animal encoded by each contruct was tested for its ability to bind studies were approved by the Animal Subjects Committee specific anti-TNF monoclonal antibodies (mAbs), as and the Biosafety Committee of the UCSD School of assessed by flow cytometry. Medicine and were performed in accordance with institu- tional guidelines. Adenovirus synthesis pcDNA3 plasmids containing the transgene of interest Chimeric gene construction were digested with the restriction enzymes NruI and SmaI We generated chimeric genes using cDNA encoding either to release a DNA fragment containing the CMV the mouse or human protein. Experiments conducted with promoter from pcDNA3, the transgene and the poly- human cells used constructs encoding the human protein adenylation signal from pcDNA3. Following gel puri- and studies performed with mouse tumor-cell lines or fication of this fragment by electrophoretic separation in a in vivo used constructs encoding the murine protein. 1% agarose gel, the DNA fragment was ligated into the The human CD154-TNF chimera was designed as EcoRV site of the adenoviral shuttle vector MCS (SK) follows. DNA encoding a fragment of human CD154 pXCX2. This plasmid is a modification of the plasmid spanning the intracellular, transmembrane and partial pXCX2 such that the pBluescript polylinker sequence has extracellular region adjacent to the transmembrane region been cloned into the E1 region (JR Tozer, UCSD, of CD154 was amplified from the full-length CD154 unpublished data, September 1993). A quantity of 5 mg cDNA by PCR using the following primers (50-GACAA each of the transgene containing MCS (SK) pXCX2 GCTTATGATCGAAACATACAACC and 50-TCAGG plasmid and JM17 plasmid were then cotransfected into ATCCTCATCTTTCTTCG). EcoRI and BamHI restric- 293AC2 cells using the calcium phosphate Profection Kit tion enzyme sites were added at the 50 and 30 ends of this from Promega according to the manufacturer’s instruc- fragment. DNA encoding the extracellular region of tions. Following transfection, the cells were cultured for human TNF distal to the TACE cleavage site was also 5 days to allow for homologous recombination and viral PCR-amplified from the full-length TNF cDNA using the synthesis. Total cells and supernatant were then harvested following primers: 50-CCGGATCCTGTAGCCCATGTT and freeze–thawed thrice to release cell-associated adeno- GTAGCAAACC and 50-GTTTCTAGATTCACAGAG virus. Clonal isolates of the adenovirus were then CAATGATTCCAAAG. XbaI and BamHI restriction obtained by plaque purification as previously described. enzyme sites were added to the ends of this fragment. Large-scale adenovirus preparations were prepared by The CD154 and TNF DNA fragments were digested with successively infecting increasing quantities of 293AC2. the restriction enzymes described above followed by The large-scale crude adenovirus preparations were then coligation into the EcoRI and XbaI sites of pcDNA3 purified over cesium chloride step gradients. This method (Invitrogen, San Diego, CA). makes use of a cesium chloride gradient for concentrating The mouse CD154-TNF chimera was constructed virus particles through a step gradient, with the densities as follows. A mouse CD154 fragment homologous to of 1.45 and 1.20 g cmÀ3, in which 293AC2-expanded the human fragment described above was released from a virus samples are centrifuged for 2 h in a SW40 rotor plasmid containing full-length mouse CD15413 with (Beckman, Brea, CA) at 25 000 r.p.m. at 4 1C The virus EcoRI and BamHI. This was ligated into pcDNA3 with band was isolated using a 27-gauge needle and syringe the PCR-amplified extracellular domain of murine TNF and desalted using a Sephadex G-25 DNA grade column generated with the following primers: 50-CCGGATCCTG (Pharmacia, Piscataway, NJ). The virus was desalted TAGCCCATGTTGTAGCAAACC and 50-GTTTCTAG against phosphate-buffered saline containing 10% glycerol

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 55 and stored at À70 1C. The final titer of the virus was ELISA plates were measured using a multiwell plate determined by anion-exchange HPLC. reader.

Flow cytometry Bioassay for soluble TNF in culture supernatant Cells were washed and suspended in staining buffer, This assay was based on a previously described assay for measuring TNF activity.14 Briefly, L929 cells were plated composed of phosphate-buffered saline containing 3% 4 FCS, 0.05% sodium azide and 10 mgmlÀ1 propidium in a 96-well flat-bottomed tissue culture plate at 4 Â 10 iodide. Cells were incubated with saturating amounts of cells per well followed by addition of serial dilutions of fluorochrome-conjugated mAb for 30 min at 4 1C. Anti- cell supernatant previously cleared of dead cells and bodies utilized included phycoerythrin-conjugated anti- debris by microcentrifugation for 5 min at 400 g.In addition, actinomycin D (Sigma, St Louis, MO) was body specific for TNF (clone Mab11, BD Biosciences, À1 San Diego, CA) and phycoerythrin-conjugated isotype added to a final concentration of 1 mgml . The final control antibody (clone MOPC-21). Stained cells were culture volume was 100 ml per well. Following 18 h 1 washed twice with staining buffer and analyzed using incubation in a 5% CO2 37 C incubator, 50 ml per well a FACSCaliber flow cytometer (Becton Dickinson, of XTT reagent (Roche Applied Science, Indianapolis, San Jose, CA). Dead cells and debris were excluded from IN) was added followed by another hour incubation the analysis by light scatter and by gating out cells that period to allow color development. The 450 nm optical failed to exclude propidium iodide. We determined the densities of developed plates were measured using a percentages of cells that expressed the transgene product multiwell plate reader. and the mean fluorescence intensity ratio of transgene- expressing cells to assess the relative expression levels of Assay for bioactivity of cell-surface TNF the transgene. The mean fluorescence intensity ratio is The functional activity of membrane-bound TNF was calculated by dividing the mean fluorescence intensity of determined in a coculture assay using L929 cells and cells that were stained with the fluorochrome-conjugated HeLa cells infected with recombinant Ad virus encoding antigen-specific mAb by the mean fluorescence intensity TNF. HeLa cells were infected with Ad virus at a of cells that were stained with a control fluorochrome- multiplicity of infection (MOI) of 10 for 24 h, the infected conjugated mAb of the same isotype but of irrelevant cells were washed with PBS, detached from the culture specificity. plate with PBS containing 10 mM EDTA, and washed once with complete media. HeLa cells were labeled with CFSE (Vybrant CFDA SE Cell Tracer Kit, Molecular Transient transfection 5 Probes, Eugene, OR) according to the manufacturer’s A total of 5 Â 10 cells were transiently transfected with recommendations. A total of 5 Â 104 HeLa cells and an 2 mg of plasmid, previously purified using Qiagen’s equal number of L929 cells were then plated in a 12-well (Valencia, CA) Maxi DNA purification kit, using culture plate, either mixed together or, for transwell Lipofectamine 2000 (Invitrogen, San Diego, CA) accord- separation experiments, HeLa cells were added to the ing to the manufacturer’s instructions. upper chamber of a 0.4 mm transwell insert to prevent direct cell–cell contact with the L929 cells. Actinomycin D ELISA quantitation of soluble TNF (Sigma) was added to the culture medium for a final Two days following infection of cells with adenovirus, the concentration of 1 mgmlÀ1. Following 18 h culture, the cell supernatant was harvested and cleared of debris by cells were stained with 10 mgmlÀ1 propidium iodide and centrifugation for 10 min at 200 g. In some assays, cells the viability of L929 cells was determined by flow negative bright were infected with adenovirus in the presence of 100 mM of cytometry and gating on the CFSE /PI cells. either the generalized matrix metalloproteinase inhibitor The antibody neutralization experiments were similarly GM6001 (Calbiochem, San Diego, CA) or its negative performed with the following exceptions: 5 Â 103 HeLa control (Calbiochem). Soluble TNF was measured by a cells transduced at an MOI of 0.3 were plated with 5 Â 104 sandwich enzyme-linked immunosorbent assay (ELISA). L929 cells. To neutralize TNF activity, the purified mouse Capture (clone Mab1, unlabeled) and detection (clone IgG1 antihuman TNF mAb (clone MAb1, BD Bio- Mab11, biotinylated) antibodies for human TNF were sciences) was added to a final concentration of 10 mgmlÀ1. À1 both obtained from BD Biosciences. Capture (clone Replicate control wells had 10 mgml of mouse IgG1 of G281-2626, unlabeled) and detection (clone MP6-XT3, irrelevant specificity. biotinylated) antibodies for mouse TNF were also obtained from BD Biosciences. Specific quantities of ELISPOT-assay TNF were calculated on the basis of titrations of a known Ninety-six-well PVDF filter plates (Multiscreen IP, quantity of recombinant soluble TNF (Biosource Inter- Millipore, Bedford, MA) were coated overnight at 4 1C national). Horseradish peroxidase avidin-biotin conju- with an antimouse interferon (IFN)-g antibody (clone gated substrate (ABC kit, Vector Laboratories, R4-6A2, BD Biosciences, San Diego, CA) at 4 mgmlÀ1 in Burlingame, CA) and TMB peroxidase substrates sterile PBS. The plates were washed two times and then (Kirkegaard Perry Laboratories) were used for enzymatic blocked with RPMI (Roswell Park Memorial Institute) color development according to the manufacturers’ 1640 medium containing 10% FCS, for at least 1 h at instructions. The 450 nm optical densities of developed 37 1C. Splenocytes from treated mice were isolated. A

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 56 total of 2 Â 105 splenocytes were then plated out into separate wells of an enzyme-linked immunosorbent spot assay (ELISPOT) plate and then either cultured alone or cocultured with mitomycin C (Sigma Chemical Co.) treated A20 stimulator cells in a 1:1 cell ratio at a total culture volume of 200 ml. The plates were incubated for 24 h at 37 1C in a humidified atmosphere of 5% CO2. The plates were then washed three times with PBS, followed by three washes with PBS containing 0.05% Tween-20. To each well, 100 ml of biotinylated anti-IFN-g mAb (clone XMG1.2, BD Biosciences, San Diego, CA) was added at a final concentration of 2 mgmlÀ1 in PBS/10% FCS, and the plates were incubated for 2 h at room temperature. Following three washes with PBS/0.05% Tween-20, 100 ml of a 1:500 Figure 1 Chimeric TNF constructs. Scale representation of the dilution of horseradish peroxidase-conjugated strepta- chimeric constructs of CD154, TNF and CD70 indicating the vidin in PBS/10% FCS was added to each well. The plates membrane anchoring domains (MAD) and ligand binding domains were incubated for 60 min at room temperature and then (LBD). The boundaries of TNF-binding domain are identical in the washed three times with PBS/0.05% Tween-20 followed chimeric constructs. The human and murine sequences display a by three washes with PBS. Fresh 3-amino-9-ethyl- high degree of similarity and differ in length by no more than two carbazole (AEC) substrate was prepared by dissolving amino acids for homologous sequences. 4 mg of 3-amino-9-ethylcarbazol (Sigma Chemical Co.) in 1 ml of dimethylformamide and then diluting this into 9 ml of sodium acetate buffer, pH5.0, that contained 5 ml of freshly added 30% hydrogen peroxide. A total of 100 ml of the sterile-filtered AEC substrate was P-values of o0.05 were considered significant. Analyses then added to each well and the plates were incubated were performed with Prism software (GraphPad, San Diego, CA). For in vivo toxicity studies, mice were for 5 min or more at room temperature. The sub- 9 strate solution was discarded and the plates were injected twice intraperitoneally with 10 PFU of adeno- rinsed with water and air-dried. After the plates were virus at a 48 h interval. Blood was collected from animals completely dry, spots were counted using the Immuno- 24 h after the second injection. Spot image analyzer and software (Cellular Technology Ltd, Cleveland, OH), and triplicate wells were used to calculate the average number of spots±the standard error (s.e.). Results Generation of Soluble TNF Animal and tumor studies We generated expression vectors encoding wtTNF WEHI-164 and A20 cell doses have been titrated in (pwtTNF) or DTACE-TNF (pDTACE-TNF), TNF lack- preliminary experiments such that subcutaneous implan- ing the putative TACE cleavage site (Figure 1). Cells tation of 3 Â 106 WEHI-164 or 1 Â 105 A20 cells, transfected with pDTACE-TNF unexpectedly released respectively, would induce palpable tumors in BALB/c substantial amounts of functional TNF, although at levels mice in approximately 1 week in 100% of the injected that were significantly less than that of cells transfected animals. For WEHI-164 tumor studies, mice were injected with pwtTNF. The culture supernatants of HT1080 cells subcutaneously into the left, shaved flank with 3 Â 106 transfected with either pwtTNF or pDTACE-TNF, but cells in 100 ml of PBS. Approximately 8 days later, when not control pcDNA3, induced apoptosis of L929 cells, tumors reached a mean diameter of 7 mm, tumors were a TNF-sensitive mouse fibroblast cell line14 (Figure 2a). injected three times in 4-day intervals with 109 plaque Upon further investigation into the expression of forming units (PFU) adenovirus (in 100 ml). For A20 DTACE-TNF, we found that soluble TNF was released tumor studies, mice were injected subcutaneously into the from various human cell lines infected with adenovirus left, shaved flank with 1 Â 105 A20 cells in 100 ml of PBS. encoding DTACE-TNF (Ad-DTACE-TNF) (Table 1). When tumors reached a mean diameter of 4–5 mm, mice Regardless of the cell line used, we found that soluble received a one-time intratumoral injection of 5 Â 108 PFU TNF was released by Ad-DTACE-TNF-infected cells, adenovirus in 100 ml of PBS. Tumor progression was although at significantly lower levels (Po0.05, Student’s monitored biweekly over the course of the experiment by t-test) than that released from Ad-wtTNF-infected cells measuring the length and width using calipers. Tumor (Table 1). No detectable TNF was produced from any of volumes were calculated by the following formula: the cell lines before transduction or after transduction volume ¼ 0.5 Â length Â (width)2. Mice were sacrificed with a control adenovirus vector. Interestingly, the differ- when tumor length exceeded 20 mm. The log-rank test ence in the amounts of soluble TNF released by cells was performed for survival analysis and differences expressing comparable levels of wtTNF versus DTACE- in tumor volume were compared using Student’s t-test. TNF varied for the different cell lines tested and ranged

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 57

Figure 2 Expression of TNF by transfected cells. (a) Secretion of soluble TNF from DTACE-TNF-transfected cells. HT1080 cells were transfected with pcDNA3 plasmids encoding vector alone (square), wtTNF (circle), or DTACE-TNF (triangle) and incubated for 2 days. Serial dilutions of cell supernatants were then incubated with L929 cells and soluble TNF bioactivity determined by the measurement of L929 apoptosis using the XTT colorimetric assay. (b) HT1080 cells were infected with adenovirus as indicated for two days. We examined the culture supernatants for soluble TNF by ELISA. (c) Cell-surface expression of TNF by Ad-infected HeLa cells. HeLa cells were infected with the indicated adenovirus and then analyzed for TNF surface expression by flow cytometry (open histograms). Shaded histograms represent cells stained with isotype control antibody. The percentage of surface-TNF-positive cells is indicated for each histogram. (d) The relationship of soluble TNF versus the mean fluorescence intensity ratio of TNF surface expression was plotted for HT1080 cells (left graph) and HeLa cells (right graph) infected at increasing MOI with adenovirus encoding wtTNF (closed squares), DTACE-TNF (open squares), CD70-TNF (open triangles) and CD154-TNF (closed circles). Data points from four different MOI for each TNF construct were used to determine the linear functions plotted on the graph.

Table 1 Soluble TNF generation following Ad-TNF infection

Name Type Soluble TNFa

wtTNF DTACE-TNF CD154-TNF

HT1080 Fibrosarcoma 43 100±3000 4000±100 o40 A549 Lung 48 200±2500 14 200±900 ND HeLa Cervical 640 000±600 52 000±500 o40 RPMI-8226 Myeloma 21 400±1600 4500±200 o40 HT1376 Bladder 60 200±300 300±10 o40 Abbreviation: ND, not determined. aExpressed as pg mlÀ1±s.d. Cell lines were infected at an MOI of 10 for 2 days followed by analysis of soluble TNF by ELISA.

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 58 from an approximate 3-fold difference for the lung of soluble cytokine relative to that of cell-surface TNF. cell line A549 to greater than a 200-fold difference for We found that the relative steepness of slopes for the the bladder cell line HT1376. Furthermore, addition cells infected with each of the Ad-TNF constructs had of the broadly active matrix metalloproteinase inhibitor the following relationship: Ad-wtTNF4Ad-DTACE- GM6001 to HeLa cells transfected with Ad-DTACE-TNF TNF4Ad-CD70-TNF4Ad-CD154-TNF. For example, did not prevent these cells from releasing soluble TNF the slopes for HT1080 cells infected with the Ad-TNF (11 000 pg mlÀ1±100 with inhibitor versus 52 000 pg mlÀ1± constructs were as follows: Ad-wtTNF (slope ¼ 1300; 500 without inhibitor, mean±s.e., n ¼ 3). Collectively, these R2 ¼ 0.8) 4Ad-DTACE-TNF (slope ¼ 290; R2 ¼ 0.8) studies indicate that DTACE-TNF still has a cleavage 4Ad-CD70-TNF (slope 415; R2 ¼ 0.9) 4Ad-CD154- site(s) that allows for release of soluble TNF from cells TNF (slope ¼ 0; R240.9) (Figure 2d). Likewise, the slopes transduced with this truncated form of TNF. for infected HeLa cells were as follows: Ad-wtTNF To explore this possibility, we generated a chimeric (slope ¼ 4600; R240.9) 4Ad-DTACE-TNF (slope ¼ 46; construct that fuses the extracellular carboxyl-terminal R240.9) 4Ad-CD70-TNF (slope 410; R240.9) 4Ad- domain of TNF with the membrane spanning portion CD154-TNF (slope ¼ 0; R240.9) (Figure 2d). Further- of CD154 that is devoid of putative matrix metallo- more, these relationships are consistent with the values proteinase cleavage sites.15 We compared the generation reported for the fold difference in amount of soluble of soluble TNF by cells infected with Ad-CD154-TNF to TNF released from cell lines infected with Ad-wtTNF cells infected with either Ad-wtTNF or Ad-DTACE-TNF compared with cells infected with either Ad-DTACE-TNF (Figure 2b). To control for nonspecific TNF secretion or Ad-CD154-TNF (Table 1). by cells, we also examined cells that either were infected with control adenovirus encoding the irrelevant transgene In vitro activity of CD154-TNF b-galactosidase (Ad-LacZ) or that were not infected To examine whether the chimeric CD154-TNF construct (noninfected). We found that Ad-CD154-TNF-infected was functional, we cocultured TNF-sensitive L929 cells HT1080 cells produced significantly less soluble TNF with HeLa cells that were infected with adenovirus (o40 pg mlÀ1, below detection limit of ELISA assay, encoding either wtTNF or CD154-TNF (Figure 3a). We Po0.05, Student’s t-test) than cells infected with adeno- separated the L929 and HeLa cells from direct cell–cell virus encoding either wtTNF (43 100±3000 pg mlÀ1)or contact using a 0.4 mm transwell insert. As such, apoptosis DTACE-TNF (4000±100 pg mlÀ1) (Table 1). Infection of of L929 should result only from diffusion of soluble other cell lines with Ad-CD154-TNF, including HCT-15 TNF from the HeLa cells across the membrane to the and COLO-205 (human colon), DU-145 (human pros- L929 cells. As expected, Ad-wtTNF-infected HeLa cells tate) and 293AC2 (human kidney) generated cells that induced apoptosis of L929 even when separated by had high-level cell-surface expression with negligible a transwell membrane. In contrast, Ad-CD154-TNF- release of soluble TNF (data not shown). infected HeLa cells did not induce apoptosis of L929 cells more than control-infected HeLa cells when sepa- Expression of surface membrane TNF rated from the L929 cells across the transwell membrane We also examined for the cell-surface expression of TNF (Figure 3a). The observed activity was due to TNF on cells transduced with Ad-wtTNF, Ad-DTACE-TNF or signaling because the killing of L929 cells could be Ad-CD154-TNF (Figure 2c). In addition, we analyzed blocked specifically by adding a neutralizing mouse cells infected with adenovirus encoding a different antihuman TNF mAb to the culture medium before the chimeric molecule composed of the TNF extracellular assay (data not shown). domain fused to the amino-terminal portion of CD70 Although the portion of CD154 used for constructing (Ad-CD70-TNF), comprised of a shorter extracellular CD154-TNF is not known to be responsible for signaling, stalk region (5 amino acids) than CD154-TNF (65 amino we specifically tested for the ability of CD154-TNF to acids), to determine if expression results were specific for make a functional interaction with CD40. We infected CD154-TNF. Each gene construct shares the same ligand- HeLa cells with Ad-CD154 or Ad-CD154-TNF, con- binding portion of the TNF molecule. At MOI ratios of firmed high-level expression of each transgene by flow 1–10, Ad-CD154-TNF-infected cells expressed higher cytometry 24 h post-infection and subsequently cocul- cell-surface levels of TNF than cells infected with tured BALB/c splenocytes with the HeLa cells overnight. adenovirus encoding any one of the other Ad-TNF Measuring CD40 surface expression on splenocytes by constructs (Figure 2c, and data not shown). flow cytometry revealed that CD154, but not CD154- We performed linear regression analyses on the TNF, decreased the amount of detectable CD40 expres- relationship between released TNF and cell-surface sion (data not shown), indicating that the CD154 portion TNF observed for cells infected at various MOI with of the TNF fusion protein does not functionally interact each of the Ad-TNF constructs. In each case, we could with CD40. define a linear relationship between the relative expression levels of soluble TNF versus cell-surface TNF for each In vivo toxicity of Ad-CD154-TNF versus Ad-wtTNF cell type following infection at various MOIs. We We examined the local and systemic toxicity of compared the different Ad-TNF constructs by calculating Ad-CD154-TNF relative to that of Ad-wtTNF. First, the slope of the line that defined each relationship, with we injected 109 PFU Ad-wtTNF, Ad-CD154-TNF or steeper slopes indicating production of greater amounts Ad-blank into WEHI-164 fibrosarcoma tumors 5 mm in

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 59

Figure 3 In vitro and in vivo toxicity of soluble and cell-surface TNF. (a) Contact-dependent killing of L929 cells by Ad-TNF-infected HeLa cells. HeLa cells were infected with adenovirus encoding the indicated transgene (GFP, wtTNF or CD154-TNF) and then either stained with CFSE and plated out with an equal number of L929 cells, to allow cell–cell contact (coculture, filled bars) or HeLa cells were placed in the upper well of a transwell filter insert (transwell, empty bars) to prevent direct cell–cell contact with L929 cells. Following 18 h coculture, L929 cells were analyzed for viability by flow cytometry as described in Material and methods. (b) Hemorrhagic tumor necrosis after injection of Ad-TNF. BALB/c mice with WEHI-164 tumors were injected intratumorally with 109 PFU Ad-blank, Ad-wtTNF or Ad-CD154-TNF. Pictures were taken 1 week after injection and are representative for each group. (c) Toxicity after systemic injection of Ad-TNF. Groups of eight FVB/N mice were injected intraperitoneally twice in a 48 h interval with 109 PFU adenovirus encoding wtTNF (closed triangle), CD154-TNF (closed diamond) or adenovirus containing no transgene (open square). Survival of animals is plotted over time. diameter that developed from subcutaneous injections of tions of Ad-blank or Ad-CD154-TNF did not develop tumor cells into the hind flanks of BALB/c mice. Injection any cutaneous ulcerations (Figure 3b). with Ad-wtTNF resulted in the formation of hemorrhagic To examine the relative toxicity of each of the ulcers surrounding the tumor within 2 days after the adenovirus vectors, we injected 109 PFU of either virus injection in all the treated animals (Figure 3b). The vector into the peritoneal cavity of FVB/N mice. Injection cutaneous ulcers of Ad-wtTNF-treated mice were most of mice with Ad-wtTNF resulted in systemic reactions pronounced around the injection site and regressed 12 h after the injection, including lethargy, ruffled fur and approximately 2–3 weeks after the injection. Similar inflammation around the eyes. Death occurred in 25% of cutaneous ulcers were observed when Ad-wtTNF was the treated animals within 48 h of injection (Figure 3c). In injected subcutaneously into the flank of mice that were contrast, mice injected with Ad-CD154-TNF or Ad-blank not bearing tumors (data not shown). However, mice that did not demonstrate any signs of toxicity. We injected the received intratumoral injections or subcutaneous injec- surviving mice in each group with the same vector

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 60 green fluorescence protein. In contrast to animals treated with Ad-GFP, the animals treated with injections of either Ad-wtTNF or Ad-CD154-TNF experienced an antitumor effect (Figure 4a). However, mice injected with Ad-wtTNF developed hemorrhagic ulcers around the injection site (Figure 3b). In spite of this, the injected tumors continued to grow in most of the Ad-wtTNF- treated animals. Injection with Ad-CD154-TNF, however, caused the tumors to regress in most animals. Moreover, the tumors of Ad-CD154-TNF-treated animals had a significantly smaller average size than the tumors of animals treated with Ad-wtTNF at 5 weeks (89±80 mm3 or 710±280 mm3, respectively) and 6 weeks (220±190 or 1230±390 mm3, respectively) after tumor challenge (Po0.05, Student’s t-test, n ¼ 8 for all groups) (Figure 4a). Sixteen weeks following the initial tumor challenge, a significantly greater proportion of the mice injected with Ad-CD154-TNF survived (80%) than did mice treated with Ad-wtTNF (47%) (P ¼ 0.02, log rank test) (Figure 4b). We also examined the activity of these vectors using the murine A20 lymphoma tumor-model system. A20 cells do not express TNF receptor 1 (TNFR1) and are not directly sensitive to TNF-induced apoptosis (data not shown). For this, BALB/c mice were each injected subcutaneously with 1 Â 105 syngeneic A20 lymphoma B cells, resulting in the development of subcutaneous lymphoma nodules of 5 mm in diameter within 10 days in each animal. These nodules were each injected with 5 Â 108 PFU Ad-blank, Ad-wtTNF or Ad-CD154-TNF, or were left untreated. Figure 4 Intratumoral injection of WEHI-164 tumor nodules. BALB/ Again, subcutaneous injection of Ad-wtTNF caused c mice were injected subcutaneously with 3 Â 106 WEHI-164 tumor cutaneous hemorrhagic ulcers around the injection site. cells at day 0. On days 8, 12 and 16, tumor nodules were injected However, in contrast to the WEHI-164 model system, the with 109 PFU adenovirus encoding GFP (open triangles, n ¼ 15), animals injected with Ad-wtTNF did not experience an wtTNF (closed triangles, n ¼ 28), CD154-TNF (closed diamonds, antitumor effect relative to that of mice injected with the n ¼ 27) or saline (open squares, n ¼ 8). (a) Mean tumor size and (b) control adenovirus vector, Ad-blank. In contrast, the animal survival are plotted over time. Asterisks indicate statistically A20 tumors regressed in most animals treated with significant differences between CD154-TNF and wtTNF treatment Ad-CD154-TNF and/or had significantly slower tumor (Student’s t-test or log-rank test, respectively). growth compared with that of tumors injected with Ad-wtTNF (44±21 versus 362±121 mm3 at week 4.5, 48 h after the first injection. All the mice that received a P ¼ 0.012, Student’s t-test, Figure 5a). Furthermore, second injection of Ad-wtTNF died, whereas none of the a significantly higher percentage of mice treated with animals injected twice with Ad-CD154-TNF or Ad-blank Ad-CD154-TNF survived than did animals treated with displayed any signs of systemic toxicity (Figure 3c). Ad-wtTNF (75 versus 33%, P ¼ 0.019, log rank test, Twenty-four hours after the second injection, we detected Figure 5b). TNF in the sera of mice treated with Ad-wtTNF We hypothesized that injection of A20 tumors with (13 300±800 pg mlÀ1) but not in the sera of mice injected the various adenovirus vectors enhanced development of with Ad-CD154-TNF or Ad-blank (data not shown). an A20-specific cellular immune response. To test this, we isolated splenocytes from tumor-bearing mice 2 weeks Antitumor activity of Ad-CD154-TNF versus Ad-wtTNF after they were injected with adenovirus vector. Spleno- Having established the decreased systemic toxicity result- cytes were cocultured with mitomycin-C-treated A20 cells ing from the elimination of soluble TNF, we tested and the numbers of IFN-g-secreting splenocytes were whether CD154-TNF could induce an antitumor response enumerated in a 24-h ELISPOT assay. We found that in vivo. In one model system, BALB/c mice were each mice injected with Ad-CD154-TNF had a significantly injected subcutaneously with a sufficient number (3 Â 106) higher number of IFN-g-secreting, A20-specific spleno- of syngeneic WEHI-164 fibrosarcoma cells to form cytes (P ¼ 0.016, Bonferroni t-test) than mice injected with palpable tumor nodules within 1 week after injection. Ad-wtTNF or Ad-blank, or mice that did not receive any These tumor nodules each were injected three times at injections with adenovirus (Figure 5c). 4-day intervals with 109 PFU of Ad-CD154-TNF, Eleven weeks following the initial injection of A20 cells, Ad-wtTNF or Ad-GFP, a control adenovirus encoding mice that had rejected the tumor following injection of

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 61

Figure 5 Intratumoral injection of A20 lymphoma nodules. BALB/c mice were injected subcutaneously with 1 Â 105 A20 tumor cells at day 0. On day 10, tumor nodules remained untreated (closed square, n ¼ 10) or were injected with 5 Â 108 PFU adenovirus carrying no transgene (Ad- blank, open square, n ¼ 16), Ad-wtTNF (open triangle, n ¼ 15) or Ad-CD154-TNF (closed diamond, n ¼ 16). (a) The mean tumor size (±s.e.) and (b) animal survival are plotted over time. Asterisks indicate statistically significant difference between CD154-TNF and wtTNF treatment (Student’s t-test or log-rank test, respectively). (c) Immune response to A20 cells in treated mice analyzed by IFN-g ELISPOT assay. Splenocytes were isolated from three mice in each treatment group approximately 2 weeks after injection of Ad. Splenocytes were then cocultured with an equal number of mitomycin C-treated A20 cells in a 96-well filter plate and the number of IFN-g-secreting spleen cells was determined by ELISPOT assay. Bars depict the mean number (±s.d.) of IFN-g-producing cells per 2 Â 105 splenocytes. The asterisk indicates a statistically significant difference (Po0.05, Bonferroni t-test) in the mean number of spots between the Ad-CD154-TNF group and all other treatment groups. (d) Mice with complete tumor regression after treatment with Ad-wt-TNF (open circle, n ¼ 4) or Ad-CD154-TNF (cross, n ¼ 7) were rechallenged with 2 Â 105 A20 cells in the contralateral flank. A group of naı¨ve mice (closed squares, n ¼ 5) also received 2 Â 105 A20 cells. The mean tumor size (±s.e.) for each group is depicted over time.

Ad-wtTNF or Ad-CD154-TNF were challenged with isolated limb perfusion of cells transduced in situ,23,24 lethal inoculums of A20 tumor cells. In contrast to control physically linking soluble TNF to the tumor-cell mem- mice, the animals that previously had rejected A20 tumors brane,25 using TNF in the form of a fusion molecule,26,27 never developed tumor nodules and remained free of or by driving expression of TNF in transduced cells with tumor (Figure 5d), demonstrating that such animals weak, tissue-specific or radiation-inducible promoters.27–30 had protective immunity against subsequent challenge In addition, others have used vectors encoding a truncated with A20. TNF that lacks the TACE cleavage site.7–10 However, such strategies still have the potential to release soluble TNF into the systemic circulation. Discussion The problem associated with use of the truncated TNF is highlighted by the data presented in this study. Tumor necrosis factor is an attractive candidate molecule Indeed, although cells transduced with Ad-DTACE-TNF for cancer therapy.16,17 TNF can directly induce apoptosis released significantly less soluble TNF than did cells of tumor cells by interaction with TNFR1 (p55).18 In transduced with Ad-wtTNF, some transduced cells still addition, TNF is a potent immune-stimulatory molecule released significant amounts of soluble cytokine (for capable of inducing antigen-presenting cells to express example, A549, Table 1). Because the TNF released high levels of immune costimulatory molecules and by Ad-DTACE-TNF-transduced cells could induce of enhancing both cellular and humoral immune apoptosis of cells sensitive to TNF-mediated apoptosis responses.19–21 However, the use of TNF has been limited (Figure 2a), we conclude that TNF does have sites secondary to the serious toxicity of TNF in the systemic other than the TACE cleavage site that can be cleaved to circulation.22 release a functional soluble cytokine. This notion is Some investigators have devised strategies to mitigate supported by data from Marr et al.,31 who observed the problems associated with the systemic release of systemic toxicity in mice after injection of tumors with soluble TNF through techniques such as hypothermic Ad-DTACE-TNF.

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 62 In light of the results with Ad-DTACE-TNF, we The observation that the ligand specificity of TNF was hypothesized that global modifications of TNF would maintained when the carboxyl-terminal portion of TNF be necessary to anchor TNF to the plasma membrane. was conjoined with the transmembrane and proximal Unlike TNF, many members of the TNF family are cell- membrane domains of another member of the TNF surface molecules. Among these is CD154, a molecule family supports the notion that the intracellular, trans- expressed on the plasma membrane of activated membrane and extracellular stalk region of TNF family T cells.20,32 Although CD154 can be cleaved by metallo- members do not directly interact with their cognate proteinases into a soluble molecule,33 deletion of this receptors. However, such domains still might play a role site generates a stable cell-surface protein15 (and data not in the metabolic fates of the expressed protein. This is shown). As such, we predicted that the membrane- supported by our studies using CD70-TNF, a construct proximal domain of CD154 lacking this solitary cleavage that we hypothesized should also anchor TNF to the site could provide a stable base for the presentation of the plasma membrane. Although the short stalk domain of functional carboxyl terminal domain of TNF. In addition, CD70 apparently resisted proteolytic cleavage, we could structural analysis of TNF family members indicate not effect high-level surface expression of CD70-TNF. a highly conserved tertiary structure of trimeric mole- Even for cells infected with Ad-CD70-TNF at high MOI, cules.34,35 Structural studies indicate that receptor-ligand the TNF surface expression was much lower than cells interactions occur in the extracellular portion correspond- similarly infected with Ad-CD154-TNF. This indicates ing to the soluble form of TNF. In contrast, the that the composition rather than the length of the stalk intracellular, transmembrane and proximal extracellular region may be important in governing the relative level of domains of TNF family-member proteins are assumed cell-surface expression. not to participate in receptor binding, although direct Tumor necrosis factor can exert its antitumor effects in structural evidence for this does not formally exist. This several ways. First, TNF can have a direct cytotoxic effect being the case, receptor binding and specificity might not on TNF-sensitive tumor cells, which involves signaling be dependent on the amino-terminal domain. As such, we through TNFR1 (p55).18 However, the antitumor activity predicted that a chimeric TNF molecule should preserve of Ad-wtTNF or Ad-TNF-CD154 that we observed in TNF receptor binding, although at the same time our study cannot be only because of the direct cytotoxic potentially acquiring the more stable cell-surface expres- effects of TNF on tumor cells, as A20 cells do not express sion properties of CD154. TNFR1 and do not undergo apoptosis in response to Indeed, we found that a chimeric TNF construct soluble TNF in vitro (data not shown). On the other hand, composed of the extracellular, TNF receptor-binding TNF can induce apoptosis of TNFR1-expressing vascular portion of TNF fused to the amino-terminal region of endothelial cells.36,37 Conceivably, some of the antitumor CD154 enables stabile, high-level cell-surface expression effect(s) of local TNF expression could be due to of TNF. Importantly, high-level surface expression with- disruption of the tumor’s microvasculature. TNF also out production of soluble cytokine was observed for this could act as a factor in the development of an antitumor chimeric molecule in all cell types tested. This is unlike immune response. TNF can stimulate maturation and what we observed for DTACE-TNF. activation of antigen-presenting cells, such as dendritic The differences in the amount of TNF released by cells cells.38 Moreover, TNF coupled with the membranes of infected with Ad-CD154-TNF versus those infected apoptotic tumor cells appears more effective than soluble with Ad-wtTNF or Ad-DTACE-TNF are not due to TNF in cross-priming T cells to tumor-associated differences in susceptibility to infection by each of the antigens.39 As such, tumor cells transduced to express adenovirus vector preparations. Indeed, we observed that CD154-TNF could be more efficient than tumor cells levels of surface TNF expression by cells infected with transduced to express wtTNF in inducing antitumor Ad-CD154-TNF at an MOI was lower than that required immunity. to effect even low-level surface expression of TNF In our studies, we observed a modest attenuation of with Ad-wtTNF or Ad-DTACE-TNF. Moreover, linear tumor growth in the groups of animals injected with regression analyses comparing the levels of soluble versus control adenovirus vectors lacking a relevant transgene. cell-surface TNF revealed different slopes for each The effect was observed with both WEHI-164 and A20 TNF construct between the different cell lines, as well tumors (see Figures 4 and 5). This could reflect inflamma- as differences in the slopes between different TNF tory and innate immune responses to infection with constructs. In all cases, infection with Ad-CD154-TNF adenovirus, which could be deleterious to tumor cell resulted in high-level surface expression of TNF with growth in vivo.40,41 The tumor growth kinetics observed negligible release of soluble TNF, resulting in a slope of in animals injected with control adenovirus, however, were zero (Figure 2d). Others have reported that removal of the significantly greater than those observed in animals injected TACE site is sufficient to abrogate detectable release of with adenovirus encoding TNF, particularly CD154-TNF. soluble TNF from murine hepatic carcinoma cells and Nevertheless, it is possible that effects on tumor cell growth that soluble and membrane-bound forms elicit distinct observed with adenovirus encoding TNF or CD154-TNF antitumor effects.10 The CD154-TNF represents a novel might be less dramatic if we employed other methods to type of TNF that, in contrast to DTACE-TNF, can be effect tumor-cell expression of TNF or CD154-TNF. expressed on the cell surface of a variety of different cell Finally, Ad-CD154-TNF was less toxic than Ad- types without detectable release of TNF. wtTNF. Injection of the latter into the skin or tumor

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 63 nodules of experimental animals induced hemorrhagic 9 Mueller C, Corazza N, Trachsel-Loseth S, Eugster HP, ulcers that were not observed in animals injected with Bu¨ hler-Jungo M, Brunner T et al. Noncleavable transmem- Ad-CD154-TNF, presumably due to the release of soluble brane mouse tumor necrosis factor-alpha (TNFalpha) TNF that can cause substantial destruction of endothelial mediates effects distinct from those of wild-type TNFalpha cells, ultimately resulting in hemorrhagic tissue necro- in vitro and in vivo. J Biol Chem 1999; 274: 38112–38118. sis.42,43 Moreover, even when delivered systemically by 10 Li Q, Li L, Shi W, Jiang X, Xu Y, Gong F et al. Mechanism injection into the peritoneum, Ad-CD154-TNF appeared of action differences in the antitumor effects of transmembrane and secretory tumor necrosis factor-alpha in vitro and well tolerated, in contrast to Ad-wtTNF. These results are in vivo. Cancer Immunol Immunother 2006; 55: 1470–1479. encouraging because CD154-TNF, in addition to enhanced 11 Grell M, Douni E, Wajant H, Lo¨ hden M, Clauss M, biological activity over that of soluble TNF, could make Maxeiner B et al. The transmembrane form of tumor TNF a viable candidate for use in clinical oncology. The necrosis factor is the prime activating ligand of the 80 kDa use of TNF for human cancer treatment has been limited tumor necrosis factor receptor. Cell 1995; 83: 793–802. due to its poor systemic tolerance, resulting in an 12 Killar L, White J, Black R, Peschon J. Adamalysins. A unacceptably low therapeutic index. However, the results family of metzincins including TNF-alpha converting reported here suggest that CD154-TNF could have a much enzyme (TACE). Ann NY Acad Sci 1999; 878: 442–452. higher therapeutic index in tumor-bearing patients, allow- 13 Mendoza RB, Cantwell MJ, Kipps TJ. Immunostimulatory ing TNF to play an effective role in anticancer therapy. effects of a plasmid expressing CD40 ligand (CD154) on gene immunization. J Immunol 1997; 159: 5777–5781. 14 Flick DA, Gifford GE. Comparison of in vitro cell cytotoxic assays for tumor necrosis factor. J Immunol Methods 1984; Acknowledgements 68: 167–175. 15 Pietravalle F, Lecoanet-Henchoz S, Aubry JP, Elson G, We would like to thank Mei Li for her help in generating Bonnefoy JY, Gauchat JF. Cleavage of membrane-bound the plasmid and adenovirus constructs. This work was CD40 ligand is not required for inducing B cell proliferation supported in part by a grant (to TJK) from the Alliance and differentiation. Eur J Immunol 1996; 26: 725–728. for Cancer Gene Therapy and an Award from the 16 Szlosarek PW, Balkwill FR. Tumour necrosis factor alpha: a National Institutes of Health, PO1-CA081534 for the potential target for the therapy of solid tumours. Lancet Oncol 2003; 4: 565–573. CLL Research Consortium (CRC). This work was 17 Mocellin S, Rossi CR, Pilati P, Nitti D. Tumor necrosis supported in part by 5R37 CA049870 grant from NIH factor, cancer and anticancer therapy. Cytokine Growth and the Alliance for Cancer Gene Therapy grant. Factor Rev 2005; 16: 35–53. 18 Chen G, Goeddel DV. TNF-R1 signaling: a beautiful pathway. Science 2002; 296: 1634–1635. References 19 Ranheim EA, Kipps TJ. Tumor necrosis factor-alpha facilitates induction of CD80 (B7-1) and CD54 on human 1 Carswell EA, Old LJ, Kassel RL, Green S, Fiore N, B cells by activated T cells: complex regulation by IL-4, Williamson B. An endotoxin-induced serum factor that IL-10, and CD40 L. Cell Immunol 1995; 161: 226–235. causes necrosis of tumors. Proc Natl Acad Sci USA 1975; 72: 20 Armitage RJ. Tumor necrosis factor receptor superfamily 3666–3670. members and their ligands. Curr Opin Immunol 1994; 6: 407–413. 2 Spriggs DR, Sherman ML, Frei III E, Kufe DW. Clinical 21 Ryffel B, Mihatsch MJ. TNF receptor distribution in human studies with tumour necrosis factor. Ciba Found Symp 1987; tissues. Int Rev Exp Pathol 1993; 34(Part B): 149–156. 131: 206–227. 22 Vrouenraets BC, Klaase JM, Nieweg OE, Kroon BB. 3 Blick M, Sherwin SA, Rosenblum M, Gutterman J. Phase I Toxicity and morbidity of isolated limb perfusion. Semin study of recombinant tumor necrosis factor in cancer Surg Oncol 1998; 14: 224–231. patients. Cancer Res 1987; 47: 2986–2989. 23 Lienard D, Ewalenko P, Delmotte JJ, Renard N, Lejeune FJ. 4 Selby P, Hobbs S, Viner C, Jackson E, Jones A, Newell D High-dose recombinant tumor necrosis factor alpha in et al. Tumour necrosis factor in man: clinical and biological combination with interferon gamma and melphalan in observations. Br J Cancer 1987; 56: 803–808. isolation perfusion of the limbs for melanoma and sarcoma. 5 Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, J Clin Oncol 1992; 10: 52–60. Wolfson MF et al. A metalloproteinase disintegrin that 24 Vrouenraets BC, Kroon BB, Ogilvie AC, van Geel AN, releases tumour-necrosis factor-alpha from cells. Nature Nieweg OE, Swaak AJ et al. Absence of severe systemic 1997; 385: 729–733. toxicity after leakage-controlled isolated limb perfusion with 6 Moss ML, Jin SL, Milla ME, Bickett DM, Burkhart W, tumor necrosis factor-alpha and melphalan. Ann Surg Oncol Carter HL et al. Cloning of a disintegrin metalloproteinase 1999; 6: 405–412. that processes precursor tumour-necrosis factor-alpha. 25 Zimmermann VS, Bondanza A, Rovere-Querini P, Colombo Nature 1997; 385: 733–736. B, Sacchi A, Fascio U et al. Characterisation of functional 7 Decoster E, Vanhaesebroeck B, Vandenabeele P, Grooten J, biotinylated TNF-alpha targeted to the membrane of apopto- Fiers W. Generation and biological characterization of tic melanoma cells. J Immunol Methods 2003; 276:79–87. membrane-bound, uncleavable murine tumor necrosis 26 Halin C, Gafner V, Villani ME, Borsi L, Berndt A, Kosmehl factor. J Biol Chem 1995; 270: 18473–18478. H et al. Synergistic therapeutic effects of a tumor targeting 8 Perez C, Albert I, DeFay K, Zachariades N, Gooding L, antibody fragment, fused to interleukin 12 and to tumor Kriegler M. A nonsecretable cell surface mutant of tumor necrosis factor alpha. Cancer Res 2003; 63: 3202–3210. necrosis factor (TNF) kills by cell-to-cell contact. Cell 1990; 27 Bauer S, Adrian N, Williamson B, Panousis C, Fadle N, 63: 251–258. Smerd J et al. Targeted bioactivity of membrane-anchored

Cancer Gene Therapy Antitumor activity of membrane-anchored TNF R Rieger et al 64 TNF by an antibody-derived TNF fusion protein. J Immunol 36 Paleolog EM, Delasalle SA, Buurman WA, Feldmann M. 2004; 172: 3930–3939. Functional activities of receptors for tumor necrosis factor- 28 Borsi L, Balza E, Carnemolla B, Sassi F, Castellani P, Berndt alpha on human vascular endothelial cells. Blood 1994; 84: A et al. Selective targeted delivery of TNFalpha to tumor 2578–2590. blood vessels. Blood 2003; 102: 4384–4392. 37 Stoelcker B, Ruhland B, Hehlgans T, Bluethmann H, 29 Kircheis R, Ostermann E, Wolschek MF, Lichtenberger C, Luther T, Mannel DN. Tumor necrosis factor induces tumor Magin-Lachmann C, Wightman L et al. Tumor-targeted necrosis via tumor necrosis factor receptor type 1-expressing gene delivery of tumor necrosis factor-alpha induces tumor endothelial cells of the tumor vasculature. Am J Pathol 2000; necrosis and tumor regression without systemic toxicity. 156: 1171–1176. Cancer Gene Ther 2002; 9: 673–680. 38 Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, Liu 30 Rasmussen H, Rasmussen C, Lempicki M, Durham R, YJ et al. Immunobiology of dendritic cells. Annu Rev Brough D, King CR et al. TNFerade biologic: preclinical Immunol 2000; 18: 767–811. toxicology of a novel adenovector with a radiation-inducible 39 Zimmermann VS, Bondanza A, Monno A, Rovere-Querini promoter, carrying the human tumor necrosis factor alpha P, Corti A, Manfredi AA. TNF-alpha coupled to membrane gene. Cancer Gene Ther 2002; 9: 951–957. of apoptotic cells favors the cross-priming to melanoma 31 Marr RA, Addison CL, Snider D, Muller WJ, Gauldie J, antigens. J Immunol 2004; 172: 2643–2650. Graham FL. Tumour immunotherapy using an adenoviral 40 Huebner RJ, Rowe WP, Schatten WE, Smith RR, vector expressing a membrane-bound mutant of murine Thomas LB. Studies on the use of viruses in the TNF alpha. Gene Ther 1997; 4: 1181–1188. treatment of carcinoma of the cervix. Cancer 1956; 9: 32 Ranheim EA, Kipps TJ. Activated T cells induce expression 1211–1218. of B7/BB1 on normal or leukemic B cells through a 41 Myers AL, Krewet JA, Shah MR. Tumor immunity and CD40-dependent signal. J Exp Med 1993; 177: 925–935. prolonged survival following combined adenovirus- 33 Graf D, Muller S, Korthauer U, van Kooten C, Weise C, HSP72 and CEA-plasmid vaccination. Vaccine 2005; 23: Kroczek RA. A soluble form of TRAP (CD40 ligand) is 3565–3571. rapidly released after T cell activation. Eur J Immunol 1995; 42 Shimomura K, Manda T, Mukumoto S, Kobayashi K, 25: 1749–1754. Nakano K, Mori J. Recombinant human tumor necrosis 34 Peitsch MC, Jongeneel CV. A 3-D model for the CD40 factor-alpha: thrombus formation is a cause of anti-tumor ligand predicts that it is a compact trimer similar to the activity. Int J Cancer 1988; 41: 243–247. tumor necrosis factors. Int Immunol 1993; 5: 233–238. 43 Watanabe N, Niitsu Y, Umeno H, Kuriyama H, 35 Idriss HT, Naismith JH. TNF alpha and the TNF receptor Neda H, Yamauchi N et al. Toxic effect of tumor necrosis superfamily: structure–function relationship(s). Microsc Res factor on tumor vasculature in mice. Cancer Res 1988; 48: Tech 2000; 50: 184–195. 2179–2183.

Cancer Gene Therapy