A new synthetic protein, TAT-RH, inhibits tumor growth through the regulation of NFκB activity
Based on its role in angiogenesis and apoptosis, the inhibition of NFκB activity is considered an effective treatment for cancer, hampered by the lack of selective and safe inhibitors. We recently demonstrated that the RH domain of GRK5 (GRK5-RH) inhibits NFκB, thus we evaluated its effects on cancer growth.
The role of GRK5-RH on tumor growth was assessed in a human cancer cell line (KAT-4). RH overexpression was induced by adenovirus mediated gene transfer; alternatively we administered a synthetic protein reproducing the RH domain of GRK5 (TAT-RH), actively transported into the cells.
In vitro, adenovirus mediated GRK5-RH overexpression (AdGRK5-NT) in human tumor cells (KAT-4) induces IκB accumulation and inhibits NFκB transcriptional activity leading to apoptotic events. In BALB/c nude mice harboring KAT-4 induced neoplasias, intra-tumor delivery of AdGRK5-NT reduces in a dose-dependent fashion tumor growth, with the highest doses completely inhibiting it. This phenomenon is paralleled by a decrease of NFκB activity, an increase of IκB levels and apoptotic events. To move towards a pharmacological setup, we synthesized the TAT-RH protein. In cultured KAT-4 cells, different dosages of TAT-RH reduced cell survival and increased apoptosis. In BALB/c mice, the anti-proliferative effects of TAT-RH appear to be dose-dependent and highest dose completely inhibits tumor growth.
Our data suggest that GRK5-RH inhibition of NFκB is a novel and effective anti-tumoral strategy and TAT-RH could be an useful tool in the fighting of cancer.
KeywordsElectrophoretic Mobility Shift Assay Green Fluorescen Protein Fluorescence 293N3S Cell Nude Mouse Tumor SYBR Green Real Time Polymerase Chain Reaction
NFκB is a dimeric, ubiquitous transcription factor involved in cellular responses to stimuli such as stress, cytokines, free radicals, ultraviolet irradiation, oxidized LDL, and bacterial or viral antigens [1, 2, 3, 4, 5]. In basal conditions, NFκB dimers are sequestered in the cytoplasm by a family of inhibitors, called IκBs, that bind NFκB by means of ankyrin repeat domains masking its nuclear localization signals (NLS); such interaction blocks NFκB in an inactive form in the cytoplasm . Activation of NFκB is initiated by the signal-induced phosphorylation of IκB proteins by IκB kinase (IKK), thus inducing IκB ubiquitination and degradation by the proteasome. At this time, NFκB is active and stably localized within the nucleus where it induces the expression of specific genes. The activation of these genes by NFκB then leads to inflammation, immune response, cell survival, or cellular proliferation depending on cell type. The pathogenetic role of NFκB has been clarified in many diseases [7, 8, 9], such as type II diabetes and insulin-resistance [10, 11]; cardiac hypertrophy ; atherosclerosis ; chronic heart failure ; cancer and angiogenesis .
There are many evidences about the role of NFκB in cancer. Indeed, Hanahan and Weinberg identified the six hallmarks that characterized tumor cells (Self-Sufficiency in Growth Signals, Insensitivity to Anti-growth Signals, Evading Apoptosis, Limitless Replicative Potential Sustained Angiogenesis, Tissue Invasion and Metastasis) [16, 17] and most of the genes that regulate such effects are under the transcriptional control of NFκB. Moreover, NFκB transcription activity is constitutively increased in many tumors like melanoma [18, 19], thyroid [20, 21] and colon  carcinoma. The mechanisms determining persistent and de-regulated NFκB activity in cancer cells are not well understood but a major role is probably played by the cellular concentration of the inhibitory protein IκB. In particular, the constitutive activation of NFκB in solid tumors has been mainly attributed to decreased IκB levels due to defective IκBα activity, constitutive IKK activity, enhanced proteasome activity, etc.
To date, different approaches have been developed to block NFκB in several conditions. A successful one is using a proteasome inhibitor, PS-341, to treat patients with refractory or resistant multiple myeloma . A protein that disrupts the association of the IKK complex is used to prevent inflammatory bone destruction . The inhibition of IκBα phosphorylation by the Bay 11-7082 compound, has been successfully used to prevent tumor growth and leukemic infiltration in a mouse model of adult T cell leukemia . Furthermore, inhibition of NFκB activation by expression of a mutant IκBα, which is resistant to phosphorylation and degradation, increased NFκB dependent apoptosis to stimuli such as TNFα [26, 27]. All these approaches open new fields for the management of NFκB-associated diseases like cancer.
G protein coupled receptor (GPCR) kinases (GRKs) regulates GPCRs signaling by inducing receptor desensitization. Recent findings unveil new cellular function for these kinases. Indeed, we have recently demonstrated that GRK5 regulates the activity of the transcription factor NFκB . In particular, in endothelial cells GRK5 is able to bind the inhibitory protein of NFκB, IκBα, by means of the RH domain (GRK5-RH) and stabilize the complex IκBα/NFκB in the nucleus, thus inhibiting NFκB transcriptional activity . Indeed, RH overexpression, which causes GRK5-IκBα interaction, inhibits the transcriptional activity and DNA binding of NFκB both in basal condition and after stimulation with LPS.
Given the notion that the modulation of NFκB transcriptional activity is an effective therapeutic strategy for cancer, we hypothesized that RH, being a potent inhibitor of NFκB, could reveal therapeutic potentialities in oncology. We therefore tested in a human tumor cell line, KAT-4, the effects of GRK5-RH on cell growth either in cultured cells or in tumors in BALB/c nude mice, grown after subcutaneous injection of KAT-4 cells.
Materials and methods
Human tumor cells (KAT-4) were a kind gift of Dr. Maddalena Illario (Federico II University, Dept. of Cellular and Molecular Pathology). This cell line has been recently authenticated by short tandem repeat (STR) profiling which demonstrated, despite the widespread knowledge, that KAT-4 are not of thyroid cancer origin but share the same STR profile with the HT-29 colon cancer cell line . Cells were cultured in Dulbecco's minimal essential medium (DMEM) supplemented with 10% foetal bovine serum (FBS) at 37°C in 95% air-5%CO2.
In order to synthesize a protein reproducing the RH domain of GRK5, engineered to be actively transported into the cells by means of TAT sequence, we designed the TAT-RH plasmid, using pcDNA3.1-GRK5-RH  as template to amplify RH sequence. The primers were designed with the addition of NCO I and KPN I restriction sites sequences: 5'CCCCCATGGCCCGAGATTACTGCAGTTTA3' and
Amplified sequences were purified by gel extraction kit (Invitrogen) and cloned into pTAT-HA vector a kind gift of Dr. Steven Dowdy (Washington University School of Medicine)  by means of T4 DNA ligase (Promega). The right frame and orientation were confirmed by restriction analysis and DNA sequencing (Avant 3100, Applied Biosystem). pTAT-HA vector comprises the 11 aminoacid TAT domain, that is the minimum sequence of HIV TAT able to transduce into cells, and histidine and HA as tags.
Production of adenoviral vectors
The adenoviral construct GRK5-NT (AdGRK5-NT) used for gene transfer in vivo has been previously described . AdGRK5-NT comprises the N terminal and RH domains of GRK5 conjugate to Green Fluorescen Protein gene (GFP) under the CMV promoter.
For Lac-z adenoviral construct, we subcloned Lac-Z gene, excised by digestion from pcDNA 3.1/Lac-Z (Invitrogen), into the AscI site of the pShuttle-linker plasmid, and the resulting plasmid was used to generate the first generation of adenoviruses (AdLac-Z).
First-generation of adenoviral vectors AdGRK5-NT and AdLac-Z was amplified in 293N3S cells in monolayer for small-scale amplification and in suspension for large scale amplification. Virus production was performed as follows: 3 liters of 293N3S cells at a concentration of 3-4 × 105 cells/ml were harvested by centrifugation and resuspended in 5% of the volume of conditioned medium and then infected with all the crude lysate obtained from the infection of two 150-mm dishes. Virus adsorption was performed at 37°C on a magnetic stir plate for 2 hours; after this step, medium (25% conditioned and 75% fresh Joklik's MEM supplemented with 5% FBS) was added to a final volume of 2 liters. Infected cells were harvested after 48 hours, lysed and resuspended in TM solution (10 mM Tris-HCl pH 8.0, 2 mM MgCl2). After three freeze-thaw cycles cells were incubated in 2 M MgCl2 in presence of DNase (10 mg/ml) for 1 hour at 37°C. After DNAase incubation, cellular debris were eliminated by centrifugation at 3500 rpm for 15 minutes and the remaining lysate was subjected to ultracentrifugation on a continuous CsCl2 gradient at 35,000 rpm for 2 hours at 4°C. The upper band, which contains empty particles, was eliminated and a second ultracentrifugation on a continuous CsCl gradient (35,000 rpm for 18 hr at 4°C) was performed. The harvested vector was dialyzed twice against TM with 4% sucrose and stored at -80°C until use. Vector concentration (particle number) was determined by UV spectrophotometric analysis at 260 nm.
Protein synthesis and purification
TAT and TAT-RH plasmids were transformed into a BL21 (DE3) pLysS (Invitrogen) bacterial strain. 2 L of Luria Broth (LB) culture was grown overnight and then Isopropylthiogalactoside (IPTG, 100 μM, 3 hrs) was added to induce protein expression. For protein purification we used denaturing conditions to recover all the recombinant proteins from bacterial inclusion bodies (Lysis buffer: 8 M Urea, 100 mM NaCl, 20 mM Hepes pH 8). Ni-NTA columns (GE Healthcare) were used for protein purification. Columns were pre-equilibrated with 10 mM imidazole; lysates were clarified by sonication and applied to the columns. After extensive washing with lysis buffer plus 20 mM imidazole, recombinant protein were eluted with increasing amounts of imidazole (100, 200 and 500 mM). Protein refolding was realized by buffer exchanging into Tris 1 M pH 7.5 using Amicon Ultra-4 Centrifugal Filters (Millipore). The recombinant protein TAT-RH was added to the culture medium in KAT-4 cells for the in vitro study and injected intra-tumor for the in vivo study. TAT protein was used as control.
Nuclear extracts preparation, Immunoprecipitation and western blot and Apoptosis analysis
The experiments were performed as previously described .
Cells were transfected with plasmid expression vectors containing the luciferase reporter gene linked to five repeats of a NFκB binding site (κB-Luc) or β-galactosidase (β-Gal) and infected with AdGRK5-NT. Transient transfection was performed using the Lipofectamine 2000 (Invitrogen) according to manufacturer's instruction. Lysates were analysed using the luciferase assay system with reporter lysis buffer from Promega and measured in a β-counter. Relative luciferase activity was normalized against the co-expressed β-galactosidase activity to overcome variations in transfection efficiency between samples.
Electrophoretic mobility shift assay (EMSA)
EMSA was performed using nuclear extracts. Double stranded NFκB oligonucleotide (5' AGTTGAGGGGACTTTCCCAGGC 3') was end-labelled using [32P]-γ ATP (GE Healthcare) and T4 polynucleotide kinase (Roche). Samples were subject to electrophoresis in 8% non denaturing polyacrilamide gels with 0.5% TBE buffer (0,09 M Tris, 0,09 M boric acid, 0,02 M EDTA). Digitalized gels autoradiographies were then quantified (Image Quant).
In vivo Study Design
Experiments were carried out, in accordance to NIH guidelines for Animal Investigation, in 6-weeks-old BALB/c immunoincompetent nude mice (Charles River), which had access to food and water ad libitum. For tumor formation, a suspension containing 2 × 106 KAT-4 cells in 200 μl of DMEM were injected subcutaneously in the dorsal side of nude mice. Animals were anesthetized using isofluorane 2%. We used mice that developed tumors of approximately 6 mm in 2 weeks. Mice were divided into 9 groups (5 mice/group) and administered twice a week for 17 days (AdGRK5-NT) or 4 weeks (TAT-RH) with intra-tumor injections of the specific treatment.
In particular, two groups received AdGRK5-NT at either low (108 pfu/ml) or high dosage (1011 pfu/ml) while another group received the higher dose of AdLACZ and was used as negative control. Three more groups of mice were treated with TAT-RH protein at the dosage of 8, 12 or 16 mg/Kg. These mice were compared to two control groups treated with either saline solution or the pTAT-HA protein lacking the RH sequence. Tumor growth was measured by caliper twice a week and by Ultrasound (VeVo 770, Visualsonics) once a week. At the end of the treatment, mice were sacrificed by cervical dislocation and tumors processed for biochemical or histological analysis. All in vivo experimental protocols were approved by the Federico II University Ethical Committed for Animal Studies.
Real Time PCR
Total RNA was isolated using Trizol reagent (Invitrogen) and cDNA was synthetized by means of Thermo-Script RT-PCR System (Invitrogen), following the manufacturer instruction. After reverse transcription reaction, real-time quantitative polymerase chain reaction (PCR) was performed with the SYBR Green real time PCR master mix kit (Applied Biosystems).
The reaction was visualized by SYBR Green Analysis (Applied Biosystem) on StepOne instrument (Applied Biosystem).
Primers for cytokines gene analysis were as follows:
TNFα: forward, 5'CCAGGAGAAAGTCAGCCTCCT3'; reverse, 5'CGATAAAGGGGTCAGAGTAAT3'; VEGF: forward, 5'CAGGCTGTCGTAACGATGAA3', reverse 5'TTTCTTGCGCTTTCGTTTTT3', GAPDH: forward, 5'AGTATGTCGTGGAGTCTACT3', reverse 5'TGTGG TCATGAGCCCTTCCAC3'.
All values obtained were normalized to the values obtained with the GAPDH primers. The results are expressed as the relative integrated intensity.
Histology and Immunocytochemistry
Paraffin embedded sections were processed for the triple layered immunocytochemical peroxidase anti-peroxidase (PAP) method. PCNA (Cell signaling), Cleaved caspase 3 (Cell signalling) and Lectin (Sigma) antiserum were used to analyze cell proliferation, death and neo-angiogenesis, respectively. The peroxidase was revealed in presence of 0,03% hydrogen peroxide and of an electron donor, 2,5% diaminobenzidine, which becomes visible as a brown precipitate. For negative controls, the primary antiserum was omitted. Sections were then viewed with an Eclipse E1000 Fluorescence Microscope (Nikon) and acquired using Sigma Scan Pro software (Jandel). For X-GAL staining and GFP visualization, cryostat sections were incubated with β-GAL solution (PBS, 0.02 mM K3Fe(CN)6, 0.02 mM K4Fe(CN)6, 0.02 mM MgCl2, 0.002% NP40, 0.05 mg/ml X-GAL stain) for 90 min at 37°C and counterstained with Eosin. Adobe Photoshop was used for final assembly of the images.
All values are presented as mean ± SEM. Two-way ANOVA was performed to compare the different parameters among the different groups. A significance level of P < 0.05 was assumed for all statistical evaluations. Statistics were computed with GraphPad Prism Software (San Diego, California).
1) Adenoviral mediated overexpression of GRK5-RH
a) Cell studies
b) BALB/c nude mice tumors
In nude mice, the injection of 2 × 106 KAT-4 cells in the dorsal lateral region results in the development of a ~6 mm diameter tumor in 2 weeks, in about 70% of mice. Tumors were treated with different doses of AdGRK5-NT. One high dose of AdLac-Z was used as control.
After 17 days of treatment, mice were sacrificed and tumors were taken for biochemical and histological analysis. Expression of AdGRK5-NT transgene is found by GFP fluorescence in the majority of cells within tumor when observed under green light (Figure 2B). In control tumors, LAC-Z expression is confirmed by blue X-GAL staining at direct light (Figure 2B). In order to evaluate the effect of GRK5-RH, we performed a western blot in homogenized tumors to analyze IκBα expression and apoptosis. Figure 2C shows that adenoviral mediated overexpression of GRK5-RH induces an increase of IκBα and cleaved caspase 3 levels in treated tumors compared to controls. NFκB also controls tumor angiogenesis. Therefore, we evaluated VEGF expression and found it significantly decreased in treated tumors compared to controls (Figure 2C). Results suggest that adenoviral mediated GRK5-RH expression inhibits NFκB activity in tumor cells. Indeed, EMSA analysis of tumors confirms NFκB inhibition: less NFκB is bound to genomic DNA in treated tumors compared to controls (Figure 2D).
2) Effects of administration of TAT-RH protein
a) Cell studies
b) BALB/c nude mice tumors
Mice body weight during TAT-RH treatment (gr)
Days of tretament
TAT RH 16 mg/kg
26,2 ± 0,10
25,6 ± 0,28
25,8 ± 0,41
24,7 ± 0,23
22,7 ± 0,32
24,0 ± 0,35
22,7 ± 0,32
23,4 ± 0,35
The effects of TAT-RH treatment on IκBα levels, apoptosis and angiogenesis, were assessed in control, 12 mg/kg and 8 mg/kg treated mice since at higher doses (16 mg/kg) of protein the tumor had a too small size for biochemical assessments. Figure 6B shows that TAT-RH treatment leads to a dose-dependent increase of IκBα and cleaved caspase 3 and a reduction of VEGF levels. Such effects is associated with NFκB activity inhibition, as evaluated by EMSA (Figure 6C).
The major finding of our manuscript is that GRK5-RH causes NFκB inhibition, through a novel mechanism of IκB stabilization, and leads to inhibition of growth and regression of size of tumors both in vitro and in vivo. We provide evidence using two sets of experiment: the first set, gathered through the overexpression of GRK5-RH by AdGRK5-RH, provides the proof of concept that GRK5-RH is able to inhibit cancer growth. It also represents the background for the synthesis and administration of a recombinant protein that resembles the inhibitory features of GRK5-RH, which was used in the second set of data, showing that TAT-RH can be used to obtain pharmacological inhibition of tumor growth.
It is known that a sustained, constitutive activation of NFκB contributes to malignant progression and therapeutic resistance in most of the major forms of human cancer, such as human lymphomas , carcinomas of the breast , prostate , lung , colon , pancreas , thyroid , head and neck  and cervix . Thus, the modulation of NFκB activity represents an useful therapeutic strategy for cancer, since NFκB inhibition promotes apoptotic events induced by chemotherapy, reduces the high proliferative rate that characterizes tumor cells and inhibits metastasis . We have recently demonstrated that the RH domain of GRK5 is an effective inhibitor of NFκB transcriptional activity in endothelial cells with implications in wound healing and tissue regeneration. In the present study we evaluated the effects of GRK5-RH in cancer. Our study was performed on an aggressive human carcinoma, the KAT-4 cell line. This tumor shows high proliferative rates and is NFkB sensitive . Several strategies have been developed for blocking NFκB in tumors that include the inhibition of NFκB signaling pathway by proteasome inhibitors, IKK inhibitors, antioxidants or recombinant adenovirus-mediated overexpression of the IκBα gene, interfering with NFκB mRNA using specific anti-sense oligonucleotide [39, 40, 41]. We synthesized a protein reproducing the RH domain of GRK5, engineered to be actively transported into the cells by means of TAT domain without the support of other vehicles and tested its anti-cancer property. This approach has been already successfully used for protein transport in the treatment of several mouse model of cancer, inflammation and other diseases [42, 43]. Our data suggest that TAT-RH protein enters into tumor cells, inhibits NFκB transcription activity and induces apoptosis, reduces tumor angiogenesis, blocks cell proliferation and consequently tumor growth in a dose dependent manner.
A clear advantage of our strategy of inhibition of NFκB is the fact that there is no overexpression of a transgene, but rather, the pharmacological inhibition of a mechanism of degradation of IκB which is both efficient and selective. One concern about inhibiting the NFκB pathway is the specificity. In particular, the proteasome which is responsible for IκB degradation has many other important functions. Thus, inhibition of proteasome activity could potentially cause severe side effects . Since our strategy is based on the sterical interaction of TAT-RH and IκB and therefore does not require the inhibition of a general cellular mechanism, such as the proteasome, we hypothesize that the side effects of chronic treatment with TAT-RH, or a small molecule resembling it, will be of minimal intensity. Indeed, TAT-RH protein is not toxic when delivered through a systemic route, as it did not induce organ damage in mice. Thus, our data suggest that TAT-RH protein is a potent inhibitor of tumor growth both in vitro and in vivo, that is safe and well tolerated when administered systemically.
We believe that the anti-tumoral properties of GRK5-RH rely at least on two possible mechanisms. It is well known that apoptosis induced by chemotherapies is attenuated in tumor cells showing enhanced NFκB activity [1, 45, 46, 47]. Here we demonstrated that GRK5-RH is able to induce apoptotic events in our model of cancer as evidenced by increased levels of the active form of caspase 3 (both by western blot and histological analysis) and Annexin V staining. Another feature that characterizes aggressive tumors is the ability to generate new vessels. Tumor angiogenesis has shown to be dependent on angiogenic factors, like chemokines and growth factors produced by macrophages, neutrophils and other inflammatory cells, all factors that have been shown to be regulated by NFκB [48, 49]. GRK5-RH treatment reduces VEGF expression and production compared to controls and inhibits angiogenesis, thus providing another mechanism for retarding the in vivo growth of KAT-4 cells.
This study demonstrates the ability of the RH domain of GRK5 to inhibit tumor growth through NFκB antagonism. This effect is achieved both in vitro and in vivo either through gene therapy or injection of TAT-RH protein. Thus, our data propose GRK5-RH as an useful therapeutic tool for cancer.
- 12.Freund C, Schmidt-Ullrich R, Baurand A, Dunger S, Schneider W, Loser P, El-Jamali A, Dietz R, Scheidereit C, Bergmann MW: Requirement of nuclear factor-kappaB in angiotensin II- and isoproterenol-induced cardiac hypertrophy in vivo. Circulation. 2005, 111: 2319-2325. 10.1161/01.CIR.0000164237.58200.5ACrossRefPubMedGoogle Scholar
- 20.Visconti R, Cerutti J, Battista S, Fedele M, Trapasso F, Zeki K, Miano MP, de Nigris F, Casalino L, Curcio F: Expression of the neoplastic phenotype by human thyroid carcinoma cell lines requires NFkappaB p65 protein expression. Oncogene. 1997, 15: 1987-1994. 10.1038/sj.onc.1201373CrossRefPubMedGoogle Scholar
- 25.Dewan MZ, Terashima K, Taruishi M, Hasegawa H, Ito M, Tanaka Y, Mori N, Sata T, Koyanagi Y, Maeda M: Rapid tumor formation of human T-cell leukemia virus type 1-infected cell lines in novel NOD-SCID/gammac(null) mice: suppression by an inhibitor against NF-kappaB. J Virol. 2003, 77: 5286-5294. 10.1128/JVI.77.9.5286-5294.2003PubMedCentralCrossRefPubMedGoogle Scholar
- 28.Sorriento D, Ciccarelli M, Santulli G, Campanile A, Altobelli GG, Cimini V, Galasso G, Astone D, Piscione F, Pastore L: The G-protein-coupled receptor kinase 5 inhibits NFkappaB transcriptional activity by inducing nuclear accumulation of IkappaB alpha. Proc Natl Acad Sci USA. 2008, 105: 17818-17823. 10.1073/pnas.0804446105PubMedCentralCrossRefPubMedGoogle Scholar
- 29.Schweppe RE, Klopper JP, Korch C, Pugazhenthi U, Benezra M, Knauf JA, Fagin JA, Marlow LA, Copland JA, Smallridge RC, Haugen BR: Deoxyribonucleic acid profiling analysis of 40 human thyroid cancer cell lines reveals cross-contamination resulting in cell line redundancy and misidentification. J Clin Endocrinol Metab. 2008, 93: 4331-4341. 10.1210/jc.2008-1102PubMedCentralCrossRefPubMedGoogle Scholar
- 31.Bargou RC, Emmerich F, Krappmann D, Bommert K, Mapara MY, Arnold W, Royer HD, Grinstein E, Greiner A, Scheidereit C, Dorken B: Constitutive nuclear factor-kappaB-RelA activation is required for proliferation and survival of Hodgkin's disease tumor cells. J Clin Invest. 1997, 100: 2961-2969. 10.1172/JCI119849PubMedCentralCrossRefPubMedGoogle Scholar
- 36.Ondrey FG, Dong G, Sunwoo J, Chen Z, Wolf JS, Crowl-Bancroft CV, Mukaida N, Van Waes C: Constitutive activation of transcription factors NF-(kappa)B, AP-1, and NF-IL6 in human head and neck squamous cell carcinoma cell lines that express pro-inflammatory and pro-angiogenic cytokines. Mol Carcinog. 1999, 26: 119-129. 10.1002/(SICI)1098-2744(199910)26:2<119::AID-MC6>3.0.CO;2-NCrossRefPubMedGoogle Scholar
- 39.Cho S, Urata Y, Iida T, Goto S, Yamaguchi M, Sumikawa K, Kondo T: Glutathione downregulates the phosphorylation of I kappa B: autoloop regulation of the NF-kappa B-mediated expression of NF-kappa B subunits by TNF-alpha in mouse vascular endothelial cells. Biochem Biophys Res Commun. 1998, 253: 104-108. 10.1006/bbrc.1998.9697CrossRefPubMedGoogle Scholar
- 41.Higgins KA, Perez JR, Coleman TA, Dorshkind K, McComas WA, Sarmiento UM, Rosen CA, Narayanan R: Antisense inhibition of the p65 subunit of NF-kappa B blocks tumorigenicity and causes tumor regression. Proc Natl Acad Sci USA. 1993, 90: 9901-9905. 10.1073/pnas.90.21.9901PubMedCentralCrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.