The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2007, 178: 7450-7457.
Copyright © 2007 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Spohn, G.
Right arrow Articles by Bachmann, M. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Spohn, G.
Right arrow Articles by Bachmann, M. F.
Right arrowPubmed/NCBI databases
*Substance via MeSH
Medline Plus Health Information
*Joint Disorders
*Tuberculosis

A Virus-Like Particle-Based Vaccine Selectively Targeting Soluble TNF-{alpha} Protects from Arthritis without Inducing Reactivation of Latent Tuberculosis

Gunther Spohn1,*, Reto Guler2,{dagger}, Pål Johansen{ddagger}, Iris Keller*, Muazzam Jacobs{dagger}, Markus Beck*, Franziska Rohner*, Monika Bauer*, Klaus Dietmeier*, Thomas M. Kündig{ddagger}, Gary T. Jennings*, Frank Brombacher{dagger} and Martin F. Bachmann*

* Cytos Biotechnology AG, Zurich-Schlieren, Switzerland; {dagger} Institute of Infectious Disease and Molecular Medicine, Department of Immunology, Health Science Faculty, University of Cape Town, Cape Town, South Africa; and {ddagger} Unit for Experimental Immunotherapy, Department of Dermatology, University Hospital of Zurich, Zurich, Switzerland


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Neutralization of the proinflammatory cytokine TNF-{alpha} by mAbs or soluble receptors represents an effective treatment for chronic inflammatory disorders such as rheumatoid arthritis, psoriasis, or Crohn’s disease. In this study, we describe a novel active immunization approach against TNF-{alpha}, which results in the induction of high titers of therapeutically active autoantibodies. Immunization of mice with virus-like particles of the bacteriophage Qbeta covalently linked to either the entire soluble TNF-{alpha} protein (Qbeta-C-TNF1–156) or a 20-aa peptide derived from its N terminus (Qbeta-C-TNF4–23) yielded specific Abs, which protected from clinical signs of inflammation in a murine model of rheumatoid arthritis. Whereas mice immunized with Qbeta-C-TNF1–156 showed increased susceptibility to Listeria monocytogenes infection and enhanced reactivation of latent Mycobacterium tuberculosis, mice immunized with Qbeta-C-TNF4–23 were not immunocompromised with respect to infection with these pathogens. This difference was attributed to recognition of both transmembrane and soluble TNF-{alpha} by Abs elicited by Qbeta-C-TNF1–156, and a selective recognition of only soluble TNF-{alpha} by Abs raised by Qbeta-C-TNF4–23. Thus, by specifically targeting soluble TNF-{alpha}, Qbeta-C-TNF4–23 immunization has the potential to become an effective and safe therapy against inflammatory disorders, which might overcome the risk of opportunistic infections associated with the currently available TNF-{alpha} antagonists.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Local overproduction of the proinflammatory cytokine TNF-{alpha} is critically involved in the pathogenesis of several chronic inflammatory disorders, including rheumatoid arthritis, psoriasis, and Crohn’s disease. Neutralization of TNF-{alpha} by mAbs (infliximab, adalimumab) or chimeric soluble receptors (etanercept) is efficacious in the treatment of these conditions but has several potential drawbacks, including the high costs. In addition, induction of allotype- or Id-specific Abs is a frequent problem, which might limit long-term efficacy in many patients. Moreover, as the number of treated patients increases, it is becoming evident, that treatment with TNF-{alpha} antagonists, in particular mAbs, increases the risk of opportunistic infections, especially those caused by intracellular pathogens like Mycobacterium tuberculosis, Listeria monocytogenes, or Histoplasma capsulatum (1, 2, 3). Therefore, there is a need for safer and more cost-effective long-term TNF-{alpha}-neutralizing treatments.

Active immunization against TNF-{alpha} represents an alternative to administration of TNF-{alpha} antagonists. In addition to the likelihood of reducing costs, a therapeutic vaccine would also have the advantage of requiring less frequent administrations. Moreover, induction of anti-antagonist Abs, as is the case for certain mAbs, would not be an issue with active immunization. The challenge in developing such an approach is to balance efficacy with potential safety issues related to both the immunization procedure itself and to the neutralization of the central cytokine TNF-{alpha}. To be efficacious, an anti-TNF-{alpha} vaccine must overcome or bypass the natural tolerance of the immune system to self proteins and induce a robust Ab response, which should neutralize the harmful effects of TNF-{alpha}. In contrast, the beneficial effects of TNF-{alpha} in host defense against microorganisms should be affected as little as possible. Furthermore, the Ab response induced by the vaccine should be reversible and not be influenced by endogenously produced TNF-{alpha}.

We have recently developed a vaccination technology that renders Ags of choice highly repetitive by chemically cross-linking them to the surface of virus-like particles (VLPs)3 (4, 5). Self-Ags displayed in such an ordered and repetitive manner efficiently cross-link B cell receptors and elicit a strong and long-lasting autoantibody response (6). This approach has already been successfully used to produce a therapeutic vaccine against the TNF-superfamily member receptor activator of NF {kappa}B ligand (RANKL). Immunization with a conjugate vaccine consisting of the extracellular part of RANKL chemically cross-linked to VLP of the bacteriophage Qbeta elicited high titers of autoantibodies against RANKL, which protected from bone loss in a murine model of postmenopausal osteoporosis (7). Similarly, immunization with Qbeta VLPs coupled to the proinflammatory cytokine IL-17 protected mice from collagen-induced arthritis, experimental autoimmune encephalitis (8), and autoimmune myocarditis (9). Available clinical data furthermore suggest that autologous vaccines based on VLPs are highly immunogenic in humans. Vaccination with Qbeta VLPs conjugated to an angiotensin II peptide elicited a specific autoantibody response in 100% of injected individuals (10). The induced Ab response was reversible, with a half-life of 19 days, and was not associated with any systemic side effect, indicating good safety and tolerability of the vaccine (10).

The present study describes the production of two different autologous VLP-based vaccines against TNF-{alpha}, one comprising the entire soluble TNF-{alpha} molecule, the other comprising a 20-aa peptide derived from TNF-{alpha}. We show that both vaccines are effective in reducing the clinical signs of disease in a murine model of rheumatoid arthritis, whereas they differ strongly in their safety with respect to the susceptibility to infections with the intracellular bacteria M. tuberculosis and L. monocytogenes.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice

C57BL/6 and DBA/1 mice as well as TNF-{alpha}-deficient C57BL/6 mice (11) were purchased from Charles River Laboratories or the University of Cape Town breeding stock. All mice were maintained under specific pathogen-free conditions and were matched for age (8–12 wk) and sex in all experiments. For M. tuberculosis infections, mice were kept in individually ventilated cages in a biohazard level 3 physical containment facility. Mouse experiments were performed according to protocols approved by the Swiss Federal Veterinary Office or the University of Cape Town Animal Research Ethics Committee.

Purification of C-TNF1–156

The nucleotide sequence encoding a hexahistidine-tagged version of the soluble form of murine TNF-{alpha} (corresponding to aa 80–235 of the transmembrane form) was amplified by PCR from plasmid pTNFa-cys-DsbC (gift from P. Sebbel, Cytos Biotechnology, Zurich-Schlieren, Switzerland) using the oligonucleotide pair TNF-2/TNF-REV (5'-ATATATCATATGGGTTGCGGCGGTGGCCACCATCACCATCATCACGGTAG-3', 5'-ATATATCTCGAGTTACAGAGCAATGACTCCAAAGTAG-3'; underlined nucleotides indicate NdeI and XhoI restriction sites, respectively) and cloned into the expression vector pModEC1 (gift from Y. Zou, Cytos Biotechnology). The resulting plasmid encodes the soluble form of murine TNF-{alpha} fused to a cysteine-containing linker and a hexahistidine tag at the N terminus (C-TNF1–156). An overnight culture of Escherichia coli strain BL21 harboring this plasmid was diluted 1/100 in 5 liters of Terrific Broth and grown at 37°C to an OD600 of 1.0. Expression of the recombinant protein was induced by addition of isopropyl-beta-D-thiogalactopyranoside to a final concentration of 1 mM. After overnight growth at 37°C, bacteria were harvested and resuspended in 100 ml of lysis buffer (50 mM Na2HPO4, 300 mM NaCl, 10 mM imidazole, pH 8.0). Cells were disrupted by sonication, cellular debris was removed by centrifugation, and the cleared lysate was applied to a Ni2+-nitrilotriacetic acid column. The column was washed with excess wash buffer (50 mM Na2HPO4, 300 mM NaCl, 20 mM imidazole, pH 8.0), and the bound C-TNF1–156 protein was eluted with elution buffer (50 mM Na2HPO4, 300 mM NaCl, 135 mM imidazole, pH 8.0) using a linear gradient. Imidazole was then removed by overnight dialysis against HEPES-buffered saline, pH 7.2.

Vaccine production and analysis

Qbeta VLPs were purified as described (4). For production of Qbeta-C-TNF4–23, purified Qbeta capsids (in 20 mM HEPES, 150 mM NaCl, pH 7.2) were derivatized by a 1-h incubation at room temperature with a 10-fold molar excess of succinimidyl-6-(beta-maleimidopropionamido)hexanoate. After removal of free cross-linker by extensive dialysis against 20 mM HEPES, 150 mM NaCl (pH 7.2), derivatized Qbeta capsids were incubated for 2 h at 15°C with a 5-fold molar excess of chemically synthesized C-TNF4–23 peptide (CGGSSQNSSDKPVAHVVANHQVE) to allow chemical cross-linking. Uncoupled peptide was removed by extensive dialysis against PBS. For production of Qbeta-C-TNF1–156, purified Qbeta capsids (in 20 mM HEPES, 150 mM NaCl, pH 7.2) were derivatized by a 1-h incubation at room temperature with a 2.5-fold molar excess of succinimidyl-6-(beta-maleimidopropionamido)hexanoate. Free cross-linker was removed by extensive dialysis against 20 mM HEPES, pH 7.2. The recombinant C-TNF1–156 protein was incubated for 1 h at room temperature with an equimolar amount of tris(2-carboxyethyl)phosphine hydrochloride for reduction of the N-terminal cysteine residue and then mixed with an equimolar amount of derivatized Qbeta capsids to allow chemical cross-linking. Uncoupled C-TNF1–156 protein was removed by dialysis against PBS using cellulose ester membranes with a molecular weight cutoff of 300,000 kDa (Spectrum Laboratories). Coupled products were analyzed on a 12% SDS-polyacrylamide gel under reducing conditions and by Western blot using Abs specific for murine TNF-{alpha} (R&D Systems) or Qbeta. After SDS-PAGE, intensities of Coomassie blue-stained bands were determined by densitometry and used to calculate the coupling efficiency and the number of Ag molecules displayed per VLP. Coupling efficiency was defined as the molar ratio of Qbeta monomers coupled to one or more TNF-{alpha} Ags to total (coupled + uncoupled) Qbeta monomers. In the case of Qbeta-C-TNF1–156, the intensity of the single discrete 33-kDa band corresponding to one Qbeta monomer coupled to one C-TNF1–156 monomer was used for calculation. In the case of Qbeta-C-TNF4–23, the intensities of the four discrete bands corresponding to one to four peptides coupled to one Qbeta monomer were used. Because the intensities of Coomassie blue-stained bands are proportional to mass, the estimate of coupling efficiency was corrected for the corresponding molecular weights. Epitope density was defined as the average number of C-TNF4–23 peptides linked to one Qbeta monomer and was calculated by averaging the relative intensities of the four discrete coupling bands after multiplication by the corresponding number of peptides.

Immunizations and ELISA

Qbeta-C-TNF4–23, Qbeta-C-TNF1–156, or Qbeta VLPs (50 µg each) were diluted in PBS to 200 µl and injected s.c. (100 µl on two ventral sites) in the absence of adjuvants. LPS (1 ng) was premixed with 20 mg of D-galactosamine (Sigma-Aldrich) and injected i.p. into immunized mice. Recombinant murine TNF-{alpha} (Peprotech) was injected i.v. Sera from immunized mice were serially diluted in PBS, 0.05% Tween 20, 2% BSA and applied to ELISA plates (Nunc), which had been coated with 1 µg/ml recombinant C-TNF1–156 protein. Reactivity of serum Abs with C-TNF1–156 was determined using HRP-conjugated goat anti-mouse IgG secondary Ab (Jackson ImmunoResearch Laboratories) at a dilution of 1/1000 in PBS, 0.05% Tween 20, 2% BSA. Plates were developed with 1,2-phenylenediamine dihydrochloride (0.4 mg/ml in 0.066 M Na2HPO4, 0.035 M citric acid, 0.01% H2O2, pH 5.0) and OD450 were determined using an ELISA reader (Bio-Rad). ELISA titers were expressed as those serum dilutions that lead to half-maximal OD450 (OD50%).

Binding of soluble TNF-{alpha} by immune sera

Female C57BL/6 mice (n = 4 per group) were immunized three times in biweekly intervals with either Qbeta-C-TNF4–23 or Qbeta-C-TNF1–156. Sera were collected after the last immunization, pooled, and mixed at fixed dilutions (1/100 for Qbeta-C-TNF1–156 and 1/2700 for Qbeta-C-TNF4–23 sera, respectively), with titrating amounts of soluble C-TNF1–156. After 1 h of incubation, sera were applied to ELISA plates that had been coated with 0.03 µg/ml C-TNF1–156. Plate-bound serum IgG Abs were detected with a HRP-labeled secondary Ab.

Binding of transmembrane TNF-{alpha} by immune sera

The DNA sequence encoding full length transmembrane murine TNF-{alpha} was amplified from cDNA obtained from activated macrophages and cloned into the viral expression vector pSinRep5 (Invitrogen Life Technologies). Baby hamster kidney (BHK) cells were infected with recombinant Sindbis virus expressing transmembrane TNF-{alpha} at a multiplicity of infection of 0.5. After overnight incubation, cells were harvested, washed with PBS containing 1% FCS, and incubated with serial dilutions of mouse sera, followed by Cy5-conjugated F(ab')2 goat anti-mouse IgG (Jackson ImmunoResearch Laboratories). Cells were analyzed by flow cytometry on a FACSCalibur using CellQuest software (BD Biosciences).

Collagen-induced arthritis model

Male DBA/1 mice were immunized s.c. on days 0, 14, and 28 with Qbeta-C-TNF4–23, Qbeta-C-TNF1–156, or Qbeta VLPs and then injected twice intradermally with 200 µg of bovine type II collagen (Chondrex) mixed with CFA (day 42) and IFA, respectively (day 63). After disease induction, mice were examined daily, and clinical scores ranging from 0 to 3 were assigned to each limb according to the degree of reddening and swelling observed, resulting in a maximal cumulative clinical score per mouse of 12. Scores were assigned according to the following scoring system: 0, normal, no swelling/reddening; 1, mild reddening and/or swelling of digits/paw; 2, reddening and swelling of entire paw/joint; 3, strong swelling including deformation of paw/joint, complete stiffness of joint. Ankle thickness of hind limbs was measured every 2–3 days with calipers. Blood samples were withdrawn on days 14, 28, 42, 63, and 85 and analyzed for TNF-{alpha}-specific Abs. Animals that lost >15% of their initial body weight or displayed a clinical score of 3 at one or more limbs for 3 consecutive days were euthanized.

L. monocytogenes infection

Female C57BL/6 mice were immunized s.c. on days 1 and 22 with Qbeta-C-TNF4–23, Qbeta-C-TNF1–156, or Qbeta VLPs. On day 46, mice were infected by i.p. injection of 10,000 CFU of L. monocytogenes (a gift from Reinhard Zbinden, Institute for Medical Microbiology, University of Zurich, Zurich, Switzerland). The inocula were prepared from overnight log phase cultures in trypticase soy broth and washed in PBS before injection. Four days later, spleens were isolated, and single-cell suspensions of spleen cells were homogenized and lysed in 0.5% saponin. Numbers of viable bacteria were determined by plating serial dilutions of homogenates on tryptic soy agar plates and counting bacterial colonies after overnight incubation at 37°C.

Culture of M. tuberculosis and preparation of glycerol stocks

M. tuberculosis (H37Rv) was grown in Middlebrook 7H9 broth (Difco) supplemented with Middlebrook oleic acid-albumin-dextrose-catalase enrichment medium (Invitrogen Life Technologies), 1% glycerol, and saline solution containing 0.05% Tween 80. Before use, bacterial clumping was disrupted by 30 aspirations through a 29-gauge needle. Midlogarithmic phase cultures were harvested, aliquoted, and frozen at –80°C. After thawing, viable cell counts were determined by plating serial dilutions of the cultures on Middlebrook 7H10 agar plates followed by incubation at 37°C.

In vivo aerosol infection with M. tuberculosis

All M. tuberculosis experiments were performed in the BSL 3 laboratories at the University of Cape Town. Stocks of M. tuberculosis were thawed and aspirated 30 times through a 29-gauge needle to disrupt clumping. Pulmonary infection was performed using an inhalation exposure system (system model A4224; Glas-Col). To infect mice with a natural dose of 50–100 CFU/lung, animals were exposed for 40 min to an aerosol generated by nebulizing ~5.5 ml of a suspension containing 107 live bacteria. Inoculum size was checked 24 h after infection by determining the bacterial load in the lung of infected mice. The body weights of infected mice were measured weekly.

Determination of bacterial load in organs of mice infected with M. tuberculosis

Bacterial loads in lung, liver, and spleen of infected mice were evaluated at different time points after infection with M. tuberculosis. Organs from sacrificed mice were removed aseptically, weighed, and homogenized in saline solution containing 0.04% Tween 80. Tenfold dilutions were plated in duplicate onto 7H10 agar supplemented with 10% oleic acid-albumin-dextrose-catalase enrichment medium and 0.5% glycerol. Plates were incubated at 37°C for 21 days, and the colonies were counted.

Antibiotic and reactivation treatment of mice

Two weeks after M. tuberculosis infection, mice were treated with 0.1 g/L rifampin and 0.1 g/L isoniazid, in drinking water, for a total of 6 wk. At week 15, mice were given 2.5% (w/v) aminoguanidine (AG) containing 10% glucose delivered in drinking water for 6 wk. Control groups received normal drinking water (12). Drinking water containing antibiotics or AG was changed weekly and volume measured to verify that mice consumed the correct dosage.

Statistics

Statistical analyses were performed using GraphPad Prism (GraphPad Software). The Mann-Whitney U test was used to determine the significance of group differences in bacterial loads after L. monocytogenes and M. tuberculosis infection. One-way ANOVA and the Newman-Keuls multiple comparison test were used to determine significance of group differences in clinical scores and ankle thickness at individual time points in the collagen-induced arthritis model. A p value of <0.05 was considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Production of two autologous VLP-based vaccines against TNF-{alpha}

We designed and produced two therapeutic vaccines against murine TNF-{alpha} by conjugating VLPs of the bacteriophage Qbeta either to the entire soluble murine TNF-{alpha} molecule (Qbeta-C-TNF1–156) or to a peptide corresponding to aa 4–23 of soluble TNF-{alpha} (Qbeta-C-TNF4–23) (13). This latter peptide was chosen because the corresponding amino acid stretch is predicted to be exposed preferably on soluble TNF-{alpha} but not on the membrane-bound form (14). It was reasoned that Abs raised against this peptide could specifically recognize the soluble form of TNF-{alpha}.

For production of Qbeta-C-TNF1–156, soluble murine TNF-{alpha} was engineered to comprise a cysteine-containing amino acid linker at its N terminus. This TNF-{alpha} derivative (C-TNF1–156) was expressed in E. coli, purified to homogeneity, and chemically cross-linked, via the introduced cysteine residue, to lysine residues on the surface of Qbeta VLPs. For production of Qbeta-C-TNF4–23, a peptide corresponding to aa 4–23 of soluble murine TNF-{alpha} (C-TNF4–23) was chemically synthesized and covalently attached via an additional cysteine-containing linker at its N terminus to Qbeta VLPs. The products of both coupling reactions were analyzed by SDS-PAGE. Cross-linking C-TNF1–156 to Qbeta VLPs yielded a prominent product corresponding to one Qbeta monomer covalently linked to one C-TNF1–156 monomer (Fig. 1a). Western blot analysis with Abs specific for Qbeta or murine TNF-{alpha} confirmed covalent attachment of C-TNF1–156 to Qbeta (not shown). The degree of coupling was calculated by densitometry to be 33%, indicating that every third Qbeta monomer was covalently linked to one C-TNF1–156 molecule. Given that each Qbeta VLP is composed of 180 subunits, it was estimated that 60 C-TNF1–156 molecules are displayed on the surface of 1 VLP. Covalent attachment of C-TNF4–23 peptides to Qbeta VLPs yielded four prominent bands with molecular masses corresponding to one to four peptides coupled per Qbeta monomer (Fig. 1b). Coupling efficiency for this vaccine was calculated to 95%, with an epitope density of 1.9, indicating that ~340 C-TNF4–23 peptides are displayed on each Qbeta particle.


Figure 1
View larger version (31K):
[in this window]
[in a new window]

  		FIGURE 1. a and b, Vaccine production. a, Cross-linking of C-TNF1–156 to Qbeta VLPs derivatized with a 2.5-fold molar excess of cross-linker (2.5xdQbeta). The conjugate vaccine (Qbeta-C-TNF1–156) displays a prominent band of ~33 kDa which corresponds to one Qbeta monomer coupled to one C-TNF1–156 monomer (black arrow). b, Cross-linking of the C-TNF4–23 peptide to Qbeta VLPs derivatized with a 10-fold molar excess of cross-linker (10xdQbeta). The conjugate vaccine (Qbeta-C-TNF4–23) displays four prominent bands (black arrows) corresponding to 1 to 4 peptides coupled to one Qbeta monomer. Higher m.w. bands (gray arrows) represent peptides coupled to Qbeta dimers. c–f, Characterization of the Ab response to Qbeta-C-TNF1–156 and Qbeta-C-TNF4–23. c, Induction of TNF-{alpha}-specific autoantibodies. Female C57BL/6 mice (n = 5 per group) were immunized s.c. on days 0, 14, and 28 (black arrows) with either vaccine, and sera were analyzed for TNF-{alpha}-specific IgG by ELISA at the indicated time points. Shown are mean titers ± SEM. d, Absence of endogenous boosting of anti-TNF-{alpha} Ab titers. Female C57BL/6 mice (n = 5 per group) were immunized s.c. on days 0 and 14 (black arrows) with either vaccine and injected i.p. on day 21 with a mixture of LPS and D-galactosamine (gray arrow). e, Binding of soluble TNF-{alpha}. Qbeta-C-TNF1–156- and Qbeta-C-TNF4–23-immune sera were incubated with titrating amounts of soluble TNF-{alpha} and applied to TNF-{alpha}-coated ELISA plates. Bound serum IgG was detected with a HRP-labeled secondary Ab. f, Binding of membrane TNF-{alpha}. BHK cells expressing transmembrane TNF-{alpha} were incubated with different dilutions of the same immune sera as used in e, or preimmune sera. Bound serum IgG was detected by FACS. The degree of binding is expressed as mean fluorescence intensity (MFI). Results shown in a, b, c, e, and f are representative of at least three repeat experiments.

 
Immunogenicity of Qbeta-C-TNF1–156 and Qbeta-C-TNF4–23

Mice were immunized s.c. with either Qbeta-C-TNF1–156 or Qbeta-C-TNF4–23 in the absence of any adjuvant. Both vaccines could overcome self-tolerance and induce TNF-{alpha}-specific autoantibodies (Fig. 1c), confirming previous results (13). In the case of Qbeta-C-TNF1–156, an ELISA titer (OD50%) of ~1/900 was reached, whereas immunization with Qbeta-C-TNF4–23 yielded an ELISA titer of ~1/2400 (not shown). A booster injection on day 14 increased the Ab titers to 1/37,000 and 1/7,500, respectively. An additional injection of Qbeta-C-TNF4–23 on day 28 further increased Ab titers to 1/14,900, but an injection of Qbeta-C-TNF1–156 had no effect (not shown). In the absence of further booster injections, titers declined with a half-life of ~26 days for Qbeta-C-TNF4–23 and 60 days for Qbeta-C-TNF1–156 (Fig. 1c). Ten months after the last immunization, titers had dropped by 98% in the case of Qbeta-C-TNF4–23 and 94% in the case of Qbeta-C-TNF1–156.

To rule out the possibility that endogenously produced TNF-{alpha} had the ability to boost the Ab response elicited by either Qbeta-C-TNF1–156 or Qbeta-C-TNF4–23, mice were immunized twice with either vaccine and then injected with a mixture of LPS and D-galactosamine, which causes a strong transient increase in endogenous TNF-{alpha} (15). Endogenously produced TNF-{alpha} did not boost the anti-TNF-{alpha} Ab titers elicited by either vaccine (Fig. 1d). Similarly, i.v. injection of 4 ng of recombinant TNF-{alpha} did not boost anti-TNF-{alpha} Ab levels (not shown).

Recognition of soluble and transmembrane TNF-{alpha} by Abs induced by Qbeta-C-TNF1–156 and Qbeta-C-TNF4–23

Competition ELISAs were performed to examine whether Abs induced by Qbeta-C-TNF1–156 and Qbeta-C-TNF4–23 recognized TNF-{alpha} in solution. Sera from immunized mice were first incubated with serial dilutions of soluble TNF-{alpha} and then applied to TNF-{alpha}-coated ELISA plates. Binding of both Qbeta-C-TNF1–156- and Qbeta-C-TNF4–23-immune sera to immobilized TNF-{alpha} was similarly inhibited by the addition of soluble TNF-{alpha} (Fig. 1e), indicating that Abs raised by both vaccines bound efficiently to soluble TNF-{alpha}.

Next we compared the ability of both sera to recognize the membrane-anchored form of TNF-{alpha}. BHK cells were infected with recombinant Sindbis virus expressing transmembrane TNF-{alpha}. These cells were incubated with different dilutions of sera from mice immunized with either vaccine, and the binding of serum Abs was detected by FACS. Abs from Qbeta-C-TNF1–156-immunized mice efficiently bound to TNF-{alpha} expressed on the membrane of BHK cells, whereas almost no binding was observed with sera from Qbeta-C-TNF4–23-immunized mice (Fig. 1f).

Efficacy of anti-TNF-{alpha} immunization in the collagen-induced arthritis model

To test the therapeutic potential of Qbeta-C-TNF1–156 and Qbeta-C-TNF4–23, groups of mice were immunized three times with either vaccine or Qbeta VLPs as control and then injected with collagen in adjuvant to induce arthritis. Qbeta-immunized mice developed severe arthritis, peaking with an average clinical score of 4.4 (Fig. 2a). In contrast, Qbeta-C-TNF4–23-immunized mice showed significantly milder symptoms, with a maximum score of only 3.5 (p < 0.05 vs Qbeta between day 68 and day 75). Mice immunized with Qbeta-C-TNF1–156 showed an even stronger reduction in the clinical score, reaching a maximum of only 2.3 (p < 0.05 vs Qbeta between day 68 and day 82). In parallel, as shown in Fig. 2b, the increase in ankle thickness was markedly delayed in both the Qbeta-C-TNF4–23- (p < 0.01 vs Qbeta on day 70, p < 0.05 vs Qbeta on day 72) and the Qbeta-C-TNF1–156-immunized group (p < 0.01 vs Qbeta on days 70 and 72; p < 0.05 vs Qbeta on day 75). Monitoring of TNF-{alpha}-specific autoantibody titers revealed that immunization with both vaccines yielded peak titers of 1/25,000 at the time the disease model was initiated. Titers declined during the disease phase to a final level of ~1/5000 (Fig. 2c). The inflammatory response induced by injection of collagen/adjuvant did not boost the anti-TNF-{alpha} Ab titers, further supporting the observation that endogenously produced TNF-{alpha} does not influence the anti-TNF-{alpha} Ab response induced by the vaccines. In a repeat experiment, kidneys of arthritic Qbeta-C-TNF4–23 and Qbeta-C-TNF1–156-immunized mice were examined histologically for signs of inflammation or immune complex pathology. All kidneys appeared normal, indicating that immune complex deposition did not occur in these mice (not shown).


Figure 2
View larger version (13K):
[in this window]
[in a new window]

  		FIGURE 2. Immunization with Qbeta-C-TNF1–156 and Qbeta-C-TNF4–23 protects from collagen-induced arthritis. Male DBA/1 mice (n = 8 per group) were immunized with Qbeta-C-TNF1–156, Qbeta-C-TNF4–23, or Qbeta VLPs on days 0, 14, and 28 and then injected with bovine type II collagen in CFA (day 42) and IFA (day 63), respectively, to induce arthritis. Data are representative of one of two repeat experiments. a, Clinical scores. Averages of each mouse group are given with SEM. *, p < 0.05 vs Qbeta between day 68 and day 75; **, p < 0.05 vs Qbeta between day 68 and day 82. b, Ankle thickness. Averages of mouse groups are given with SEM. *, p < 0.05; **, p < 0.01 vs Qbeta. c, Sera were analyzed for TNF-{alpha}-specific IgG by ELISA at the indicated time points. Values are mean titers ± SEM.

 
Enhanced susceptibility of Qbeta-C-TNF1–156-immunized mice to L. monocytogenes infection

TNF-{alpha} is critically involved in the protective immune response against the intracellular bacterium L. monocytogenes (16). To test the safety of both vaccines in this regard, groups of mice were immunized twice with Qbeta-C-TNF1–156, Qbeta-C-TNF4–23, or nonconjugated Qbeta VLPs and challenged with L. monocytogenes. Four days after infection, the bacterial load in spleens of Qbeta-C-TNF1–156-immunized mice was increased 100-1000-fold compared with mice that had been immunized with Qbeta VLPs alone (p < 0.05; Fig. 3a). In contrast, Qbeta-C-TNF4–23-immunized mice exhibited similar bacterial loads as Qbeta-immunized controls (p = 0.34; Fig. 3a). TNF-{alpha}-specific IgG titers were very similar at the time of infection in both immunized groups (Fig. 3b), ruling out the possibility that the missing interference with bacterial clearance after Qbeta-C-TNF4–23 vaccination was due to lower Ab titers. We conclude that immunization with Qbeta-C-TNF1–156 but not Qbeta-C-TNF4–23 increases the susceptibility to infection with L. monocytogenes.


Figure 3
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 3. Qbeta-C-TNF1–156-immunized mice show increased susceptibility to L. monocytogenes infection. Female C57BL/6 mice (n = 4 per group) were immunized twice with Qbeta-C-TNF1–156, Qbeta-C-TNF4–23, or Qbeta VLPs, and challenged with L. monocytogenes. a, Bacterial titers in spleens 4 days after infection. Individual bacterial titers (CFU) are given together with group medians. *, p < 0.05. b, Sera were analyzed for TNF-{alpha}-specific IgG by ELISA at the indicated time points. Values are mean titers ± SEM. Black arrows, vaccine injections; gray arrow, time point of L. monocytogenes challenge.

 
Increased reactivation of latent tuberculosis in Qbeta-C-TNF1–156- immunized mice

Blocking both soluble and membrane-bound forms of TNF-{alpha} or knocking out TNF-{alpha} in mice results in increased susceptibility to mycobacterial infection (17, 18). Expression of functional transmembrane TNF-{alpha} in the absence of soluble TNF-{alpha} results in enhanced susceptibility, but only during chronic infection (19, 20, 21). We tested the effect of anti-TNF-{alpha} immunization in a model of spontaneous and drug-induced reactivation of latent tuberculosis. Groups of mice were immunized with either Qbeta-C-TNF4–23, Qbeta-C-TNF1–156 or Qbeta VLPs, and infected with M. tuberculosis by aerosol. A group of naive TNF-{alpha}-deficient mice was infected at the same time as control. Two weeks after infection, mice were treated with a combination of antibiotics (rifampin and isoniazid) for 6 wk, which reduced the bacterial burden to undetectable levels (not shown). In this setting, a low number of latent bacilli survive and can cause a flare-up of disease either by spontaneous reactivation or after inducible NO synthase (iNOS) inhibition in macrophages (22). To test for disease reactivation in both situations, infected antibiotic-treated mice were left either untreated or treated with the iNOS inhibitor AG for 6 wk, starting at 15 wk after infection. As expected from previous studies (23), all antibiotic-treated TNF-{alpha}-deficient mice died within 28 wk of infection, indicating spontaneous reactivation in the absence of TNF-{alpha} (Fig. 4a). All Qbeta- and Qbeta-C-TNF4–23-immunized mice survived the infection and gained weight, irrespective of reactivation treatment (Fig. 4, b and c). Reactivation with AG resulted in rapid weight loss in all Qbeta-C-TNF1–156-immunized mice (Fig. 4d). One mouse of this group succumbed to the infection at week 18 with 1000-fold higher bacilli counts in lungs as compared with Qbeta- and Qbeta-C-TNF4–23-immunized mice (not shown), whereas the other three mice regained weight and recovered. Furthermore, one of the four non-antibiotic-treated Qbeta-C-TNF1–156-immunized mice died at 41 wk after infection, indicating increased susceptibility to M. tuberculosis.


Figure 4
View larger version (27K):
[in this window]
[in a new window]

  		FIGURE 4. Immunization with Qbeta-C-TNF1–156 enhances reactivation of latent M. tuberculosis infection. a, TNF-{alpha}-deficient naive mice (n = 25), or C57BL/6 mice (n = 20 per group), which were immunized four times (weeks –5, –3, –1, and 10) with either b, Qbeta, c, Qbeta-C-TNF4–23, or d, Qbeta-C-TNF1–156, were infected at wk 0 with M. tuberculosis. Mice were either left untreated for the rest of the experiment (n = 8 per group) or were treated with antibiotics for 6 wk starting at wk 2 after infection (n = 12 per group). From wk 15 on, a subgroup of antibiotic-treated mice was treated with AG for 6 wk to reactivate latent tuberculosis (n = 4 per group). Weight was measured weekly. Mice that lost >20% of their original body weight were considered moribund (cross) and were sacrificed. Data are presented as the mean of body weight change ± SEM.

 
These data indicate that immunization with Qbeta-C-TNF1–156 but not Qbeta-C-TNF4–23 increases the risk for drug-induced reactivation of latent tuberculosis. This was further confirmed by bacterial counts from lungs, liver, and spleens of immunized mice. Qbeta-C-TNF1–156-immunized mice had 10- to 100-fold increased bacterial titers in all three organs when compared with Qbeta-immunized mice (p < 0.05 for all organs; Fig. 5a) at 16 wk postinfection. At wk 43, Qbeta-C-TNF1–156-immunized mice had still higher bacterial numbers in all three organs with respect to the other groups (not shown). In contrast, Qbeta-C-TNF4–23-immunized mice had bacilli counts similar to those of Qbeta-immunized control mice at all time points. TNF-{alpha}-specific IgG titers were comparable between both immunized groups at all time points tested (wks 0, 16, and 43; Fig. 5b), ruling out the possibility that differences in the Ab titers were responsible for the different safety profiles. Long-term infection with M. tuberculosis (week 60) resulted in similar bacterial burden in all three groups, most probably as a result of the reduced TNF-{alpha}-specific autoantibody titers at this time point (not shown).


Figure 5
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 5. a, Increased bacterial burden in Qbeta-C-TNF1–156-immunized mice after M. tuberculosis infection. Bacterial load was determined in mice immunized with Qbeta, Qbeta-C-TNF1–156 or Qbeta-C-TNF4–23 at 16 wk postinfection (p.i.) with M. tuberculosis (n = 4 per group). Individual bacterial titers (CFU/organ) are given together with group medians. *, p < 0.05. b, Time course of TNF-{alpha}-specific IgG titers. Values are mean titers ± SEM. Black arrows, vaccine injections; gray arrow, time point of M. tuberculosis infection.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
The treatment of inflammatory autoimmune diseases with TNF-{alpha} antagonists has proven efficacious in reducing disease symptoms and is becoming increasingly important as a therapeutic alternative to classical immunosuppressive regimens. However, surveillance of adverse advents after treatment with these antagonists has shown an unusually high incidence of granulomatous infections, especially tuberculosis (2, 3, 24). Recent case reports have also highlighted an increased risk for infection with the intracellular bacterium L. monocytogenes (1) and the fungus H. capsulatum (25). Treatment with infliximab appeared to be associated with a significantly higher risk of infection than treatment with the recombinant soluble TNF-{alpha} receptor etanercept (2, 3, 26). Although the mechanism is not yet fully understood, it is likely that differences in the type of TNF-{alpha} binding account for these different safety profiles. Infliximab forms stable complexes with soluble TNF-{alpha} and binds with high avidity to transmembrane TNF-{alpha}, whereas etanercept forms relatively unstable complexes with soluble TNF-{alpha} and interacts only weakly and reversibly with the transmembrane form (27). TNF-{alpha} expressed on the surface of macrophages and T cells is particularly important in host defense against intracellular bacteria, principally by mediating the formation of granulomas, which are essential for containment of the infection. It has been shown that TNF-{alpha}-deficient mice readily succumb to M. tuberculosis (17), whereas mice lacking the soluble form of TNF-{alpha} but expressing a noncleavable transmembrane form are able to control the infection, at least during the acute phase (19, 20, 21). These latter mice can still form compact granulomas, which prevent bacterial spreading, whereas mice completely lacking TNF-{alpha} are severely impaired in granuloma formation. Expression of a functional transmembrane form of TNF-{alpha} also confers partial protection to L. monocytogenes infection (28, 29), whereas TNF-{alpha}- or TNF-RI-deficient mice are highly susceptible (16, 29, 30, 31).

In an attempt to develop an efficacious and safe active vaccination approach against TNF-{alpha}, we designed and produced two different VLP-based vaccines, one comprising the entire soluble molecule (Qbeta-C-TNF1–156), the other comprising a 20-aa peptide derived from the N-terminal region (Qbeta-C-TNF4–23). Both vaccines raised high titers of TNF-{alpha}-specific autoantibodies, which could be boosted by multiple injections (Fig. 1c). In the absence of booster injections, titers constantly declined with half-lives of 1–2 mo, indicating reversibility of the induced Ab response (Fig. 1c). Importantly, TNF-{alpha}-specific autoantibody titers were not increased after stimulation of TNF-{alpha} production by either LPS/D-galactosamine injection (Fig. 1d), arthritis induction (Fig. 2c), or mycobacterial infection (Fig. 5b), showing that endogenous TNF-{alpha} does not have the ability to boost the Ab responses induced by either vaccine. This is consistent with an intact Th cell tolerance in the host, which renders it impossible to activate TNF-{alpha}-specific memory B cells with TNF-{alpha} alone (32).

The most notable difference between the vaccines is the almost complete lack of recognition of transmembrane TNF-{alpha} by Abs raised by Qbeta-C-TNF4–23 (Fig. 1f). One possible explanation for this difference might be the particular topology of the region corresponding to the peptide sequence in the globular TNF-{alpha} molecule. The amino acid stretch corresponding to the N-terminal part of peptide C-TNF4–23 (SSQNSSDKP) is exposed on the surface of soluble TNF-{alpha}, but, due to its location between the transmembrane and the extracellular domains, is predicted to be at least partially hidden in the membrane-bound form. The central part of the peptide (VAHVVA) corresponds to an internal beta-sheet, which is involved in intermolecular interactions of the individual TNF-{alpha} monomers and is expected to become accessible only upon dissociation of the TNF-{alpha} trimer. Monomerization of soluble trimeric TNF-{alpha} can be induced by certain small molecules (33, 34), and exchange of monomers between TNF-{alpha} trimers has been described (35), indicating that the trimer interface of soluble TNF-{alpha} can become at least transiently accessible to Ab binding. Few data are available concerning possible monomer-trimer transitions of transmembrane TNF-{alpha}. However, it is reasonable to speculate that such transitions are less likely to occur than for the soluble form, due to the higher local concentration of TNF-{alpha} on the cell membrane, which favors trimerization (36). Although the relative contributions of the different epitopes to the overall Ab response are not known, it is plausible that a large proportion of the Abs elicited by Qbeta-C-TNF4–23 bind their cognate epitope on the soluble and not on the transmembrane form.

Both Qbeta-C-TNF1–156 and Qbeta-C-TNF4–23 induced similar anti-TNF-{alpha} Ab titers (Fig. 2c), but Qbeta-C-TNF1–156 was more efficient than Qbeta-C-TNF4–23 in reducing disease symptoms in a murine model of rheumatoid arthritis (Fig. 2, a and b). This apparent discrepancy might be explained by the poor recognition of transmembrane TNF-{alpha} by Abs induced by Qbeta-C-TNF4–23. Although the importance of soluble vs membrane TNF-{alpha} has not been investigated in detail in the collagen-induced arthritis model, experiments with transgenic mouse models imply an important role for the transmembrane form in the pathogenic process (37, 38). Interestingly, the protective effect of both vaccines seemed to be most pronounced during the early symptomatic phase of disease and appeared to be partially lost during the late phase (Fig. 2, a and b). This might be explained by the constant drop in TNF-{alpha}-specific Ab titers in both groups, resulting in a 5-fold decrease from the time of disease induction (day 42) to the time of sacrifice (day 86; Fig. 2c). Preliminary experiments have shown that by increasing the dose from 50 to 500 µg, the Ab response to the Qbeta-C-TNF4–23 vaccine could be increase by a factor of 3 (not shown). Although not tested in the collagen-induced arthritis model, it can be anticipated that an increase in the anti-TNF-{alpha} titer by injection of higher doses and/or additional booster injections would result in a further delay of the onset of clinical symptoms or even in complete protection. Such an almost complete protection has been observed in the same model after immunization with a Qbeta-IL-17 vaccine (8). It should, however, be noted that this latter experiment was terminated at an earlier time point (day 81) than the experiment addressing the TNF-{alpha} vaccines. When comparing Qbeta-IL-17- and Qbeta-C-TNF1–156-immunized mice at this time point, no striking difference in the average clinical scores can be detected, indicating similar efficacy of both vaccines.

Immunization of mice with Qbeta-C-TNF4–23 did not affect immunity to intracellular bacteria, while immunization with Qbeta-C-TNF1–156 resulted in enhanced susceptibility. Consistent with the high susceptibility of TNF-{alpha}-deficient mice to L. monocytogenes infection and the reduced susceptibility of transgenic mice expressing only transmembrane TNF-{alpha}, complete neutralization of TNF-{alpha} by immunization with Qbeta-C-TNF1–156 impaired clearance of the bacteria, whereas neutralization of only the soluble form after immunization with Qbeta-C-TNF4–23 had little or no effect (Fig. 3a). Similarly, the complete neutralization of soluble and membrane TNF-{alpha} by Qbeta-C-TNF1–156 might be responsible for the increased susceptibility of Qbeta-C-TNF1–156-immunized mice to primary M. tuberculosis infection and especially to drug-induced reactivation of latent infection (Figs. 4 and 5). The poor neutralization of membrane TNF-{alpha} by Qbeta-C-TNF4–23 in contrast might ensure containment of the infection, despite the neutralization of the soluble form. Indeed, Qbeta-C-TNF4–23-immunized mice were able to mount an efficient cell-mediated immune response to M. tuberculosis infection. The maturation of lung granulomas and bactericidal mechanisms such as the activation of iNOS was similar between Qbeta-C-TNF4–23- and Qbeta-immunized groups at 11 wk postinfection (not shown). In contrast, Qbeta-C-TNF1–156-immunized mice displayed focal necrotic lesions, increased lung weights, and reduced iNOS induction when compared with the other groups (not shown).

In summary, immunization with Qbeta-C-TNF4–23 has the potential to become a novel efficient therapy for rheumatoid arthritis and other autoimmune disorders, which might add a new level of safety to the existing anti-TNF-{alpha} therapies. By selectively targeting only the soluble form of TNF-{alpha} and sparing the transmembrane form, pathogenic effects of TNF-{alpha} could be neutralized by the vaccine, while important functions in the host response to intracellular pathogens remain intact.


   Acknowledgments
 
We are grateful to Faried Abbass, Roanne Keeton, Anita Schwegmann, Joni Mitchell, Berenice Arendse, and Mona Afshar for excellent technical assistance. We thank Tania Botha (Cape Technikon, Cape Town, South Africa), for constructive advice on the tuberculosis reactivation model. We also thank Hiram Arendse and Noel Markgraaf (Animal Unit, University of Cape Town, Cape Town, South Africa) for their help in s.c. vaccine injections and P. Maurer and H. Stocker (Cytos Biotechnology) for their advice on data analysis.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
G. S., I. K., Ma. B., F. R., Mo. B., K. D., G. T. J., and M. F. B. are employees and hold stocks and/or stock options of Cytos Biotechnology.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Address correspondence and reprint requests to Dr. Gunther Spohn, Cytos Biotechnology AG, Wagistrasse 25, Zurich-Schlieren, Switzerland. E-mail address: gunther.spohn{at}cytos.com Back

2 R.G. was supported by grants from the Swiss National Foundation and the Swiss Foundation for Fellowships in Biology and Medicine. Back

3 Abbreviations used in this paper: VLP, virus-like particle; RANKL, receptor activator of NF-{kappa}B ligand; AG, aminoguanidine; BHK, baby hamster kidney; iNOS, inducible nitric-oxide synthase. Back

Received for publication January 25, 2007. Accepted for publication March 23, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Slifman, N. R., S. K. Gershon, J. H. Lee, E. T. Edwards, M. M. Braun. 2003. Listeria monocytogenes infection as a complication of treatment with tumor necrosis factor {alpha}-neutralizing agents. Arthritis Rheum. 48: 319-324. [Medline]
  2. Wallis, R. S., M. Broder, J. Wong, D. Beenhouwer. 2004. Granulomatous infections due to tumor necrosis factor blockade: correction. Clin. Infect. Dis. 39: 1254-1255. [Medline]
  3. Wallis, R. S., M. S. Broder, J. Y. Wong, M. E. Hanson, D. O. Beenhouwer. 2004. Granulomatous infectious diseases associated with tumor necrosis factor antagonists. Clin. Infect. Dis. 38: 1261-1265. [Medline]
  4. Jegerlehner, A., T. Storni, G. Lipowsky, M. Schmid, P. Pumpens, M. F. Bachmann. 2002. Regulation of IgG antibody responses by epitope density and CD21-mediated costimulation. Eur. J. Immunol. 32: 3305-3314. [Medline]
  5. Jegerlehner, A., A. Tissot, F. Lechner, P. Sebbel, I. Erdmann, T. Kundig, T. Bachi, T. Storni, G. Jennings, P. Pumpens, et al 2002. A molecular assembly system that renders antigens of choice highly repetitive for induction of protective B cell responses. Vaccine 20: 3104-3112. [Medline]
  6. Bachmann, M. F., U. H. Rohrer, T. M. Kundig, K. Burki, H. Hengartner, R. M. Zinkernagel. 1993. The influence of antigen organization on B cell responsiveness. Science 262: 1448-1451. [Abstract/Free Full Text]
  7. Spohn, G., K. Schwarz, P. Maurer, H. Illges, N. Rajasekaran, Y. Choi, G. T. Jennings, M. F. Bachmann. 2005. Protection against osteoporosis by active immunization with TRANCE/RANKL displayed on virus-like particles. J. Immunol. 175: 6211-6218. [Abstract/Free Full Text]
  8. Rohn, T. A., G. T. Jennings, M. Hernandez, P. Grest, M. Beck, Y. Zou, M. Kopf, M. F. Bachmann. 2006. Vaccination against IL-17 suppresses autoimmune arthritis and encephalomyelitis. Eur. J. Immunol. 36: 2857-2867. [Medline]
  9. Sonderegger, I., T. A. Rohn, M. O. Kurrer, G. Iezzi, Y. Zou, R. A. Kastelein, M. F. Bachmann, M. Kopf. 2006. Neutralization of IL-17 by active vaccination inhibits IL-23-dependent autoimmune myocarditis. Eur. J. Immunol. 36: 2849-2856. [Medline]
  10. Ambuhl, P. M., A. C. Tissot, A. Fulurija, P. Maurer, J. Nussberger, R. Sabat, V. Nief, C. Schellekens, K. Sladko, K. Roubicek, et al 2007. A vaccine for hypertension based on virus-like particles: preclinical efficacy and phase I safety and immunogenicity. J. Hypertens. 25: 63-72. [Medline]
  11. Marino, M. W., A. Dunn, D. Grail, M. Inglese, Y. Noguchi, E. Richards, A. Jungbluth, H. Wada, M. Moore, B. Williamson, et al 1997. Characterization of tumor necrosis factor-deficient mice. Proc. Natl. Acad. Sci. USA 94: 8093-8098. [Abstract/Free Full Text]
  12. Botha, T., B. Ryffel. 2002. Reactivation of latent tuberculosis by an inhibitor of inducible nitric oxide synthase in an aerosol murine model. Immunology 107: 350-357. [Medline]
  13. Chackerian, B., D. R. Lowy, J. T. Schiller. 2001. Conjugation of a self-antigen to papillomavirus-like particles allows for efficient induction of protective autoantibodies. J. Clin. Invest. 108: 415-423. [Medline]
  14. Jones, E. Y., D. I. Stuart, N. P. Walker. 1989. Structure of tumour necrosis factor. Nature 338: 225-228. [Medline]
  15. Car, B. D., V. M. Eng, B. Schnyder, L. Ozmen, S. Huang, P. Gallay, D. Heumann, M. Aguet, B. Ryffel. 1994. Interferon {gamma} receptor deficient mice are resistant to endotoxic shock. J. Exp. Med. 179: 1437-1444. [Abstract/Free Full Text]
  16. Pfeffer, K., T. Matsuyama, T. M. Kundig, A. Wakeham, K. Kishihara, A. Shahinian, K. Wiegmann, P. S. Ohashi, M. Kronke, T. W. Mak. 1993. Mice deficient for the 55 kd tumor necrosis factor receptor are resistant to endotoxic shock, yet succumb to L. monocytogenes infection. Cell 73: 457-467. [Medline]
  17. Bean, A. G., D. R. Roach, H. Briscoe, M. P. France, H. Korner, J. D. Sedgwick, W. J. Britton. 1999. Structural deficiencies in granuloma formation in TNF gene-targeted mice underlie the heightened susceptibility to aerosol Mycobacterium tuberculosis infection, which is not compensated for by lymphotoxin. J. Immunol. 162: 3504-3511. [Abstract/Free Full Text]
  18. Flynn, J. L., M. M. Goldstein, J. Chan, K. J. Triebold, K. Pfeffer, C. J. Lowenstein, R. Schreiber, T. W. Mak, B. R. Bloom. 1995. Tumor necrosis factor-{alpha} is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 2: 561-572. [Medline]
  19. Olleros, M. L., R. Guler, N. Corazza, D. Vesin, H. P. Eugster, G. Marchal, P. Chavarot, C. Mueller, I. Garcia. 2002. Transmembrane TNF induces an efficient cell-mediated immunity and resistance to Mycobacterium bovis Bacillus Calmette-Guérin infection in the absence of secreted TNF and lymphotoxin-{alpha}. J. Immunol. 168: 3394-3401. [Abstract/Free Full Text]
  20. Olleros, M. L., R. Guler, D. Vesin, R. Parapanov, G. Marchal, E. Martinez-Soria, N. Corazza, J. C. Pache, C. Mueller, I. Garcia. 2005. Contribution of transmembrane tumor necrosis factor to host defense against Mycobacterium bovis Bacillus Calmette-Guérin and Mycobacterium tuberculosis infections. Am. J. Pathol. 166: 1109-1120. [Abstract/Free Full Text]
  21. Saunders, B. M., S. Tran, S. Ruuls, J. D. Sedgwick, H. Briscoe, W. J. Britton. 2005. Transmembrane TNF is sufficient to initiate cell migration and granuloma formation and provide acute, but not long-term, control of Mycobacterium tuberculosis infection. J. Immunol. 174: 4852-4859. [Abstract/Free Full Text]
  22. Flynn, J. L., C. A. Scanga, K. E. Tanaka, J. Chan. 1998. Effects of aminoguanidine on latent murine tuberculosis. J. Immunol. 160: 1796-1803. [Abstract/Free Full Text]
  23. Botha, T., B. Ryffel. 2003. Reactivation of latent tuberculosis infection in TNF-deficient mice. J. Immunol. 171: 3110-3118. [Abstract/Free Full Text]
  24. Keane, J., S. Gershon, R. P. Wise, E. Mirabile-Levens, J. Kasznica, W. D. Schwieterman, J. N. Siegel, M. M. Braun. 2001. Tuberculosis associated with infliximab, a tumor necrosis factor {alpha}-neutralizing agent. N. Engl. J. Med. 345: 1098-1104. [Abstract/Free Full Text]
  25. Lee, J. H., N. R. Slifman, S. K. Gershon, E. T. Edwards, W. D. Schwieterman, J. N. Siegel, R. P. Wise, S. L. Brown, J. N. Udall, Jr, M. M. Braun. 2002. Life-threatening histoplasmosis complicating immunotherapy with tumor necrosis factor {alpha} antagonists infliximab and etanercept. Arthritis Rheum. 46: 2565-2570. [Medline]
  26. Wallis, R. S., M. Broder, J. Wong, A. Lee, L. Hoq. 2005. Reactivation of latent granulomatous infections by infliximab. Clin. Infect. Dis. 41: (Suppl 3):S194-S198. [Medline]
  27. Scallon, B., A. Cai, N. Solowski, A. Rosenberg, X. Y. Song, D. Shealy, C. Wagner. 2002. Binding and functional comparisons of two types of tumor necrosis factor antagonists. J. Pharmacol. Exp. Ther. 301: 418-426. [Abstract/Free Full Text]
  28. Musicki, K., H. Briscoe, S. Tran, W. J. Britton, B. M. Saunders. 2006. Differential requirements for soluble and transmembrane tumor necrosis factor in the immunological control of primary and secondary Listeria monocytogenes infection. Infect. Immun. 74: 3180-3189. [Abstract/Free Full Text]
  29. Torres, D., L. Janot, V. F. Quesniaux, S. I. Grivennikov, I. Maillet, J. D. Sedgwick, B. Ryffel, F. Erard. 2005. Membrane tumor necrosis factor confers partial protection to Listeria infection. Am. J. Pathol. 167: 1677-1687. [Abstract/Free Full Text]
  30. Pasparakis, M., L. Alexopoulou, V. Episkopou, G. Kollias. 1996. Immune and inflammatory responses in TNF {alpha}-deficient mice: a critical requirement for TNF {alpha} in the formation of primary B cell follicles, follicular dendritic cell networks and germinal centers, and in the maturation of the humoral immune response. J. Exp. Med. 184: 1397-1411. [Abstract/Free Full Text]
  31. Rothe, J., W. Lesslauer, H. Lotscher, Y. Lang, P. Koebel, F. Kontgen, A. Althage, R. Zinkernagel, M. Steinmetz, H. Bluethmann. 1993. Mice lacking the tumour necrosis factor receptor 1 are resistant to TNF-mediated toxicity but highly susceptible to infection by Listeria monocytogenes. Nature 364: 798-802. [Medline]
  32. Bachmann, M. F., M. R. Dyer. 2004. Therapeutic vaccination for chronic diseases: a new class of drugs in sight. Nat. Rev. Drug Discov. 3: 81-88. [Medline]
  33. Alzani, R., A. Corti, L. Grazioli, E. Cozzi, P. Ghezzi, F. Marcucci. 1993. Suramin induces deoligomerization of human tumor necrosis factor {alpha}. J. Biol. Chem. 268: 12526-12529. [Abstract/Free Full Text]
  34. He, M. M., A. S. Smith, J. D. Oslob, W. M. Flanagan, A. C. Braisted, A. Whitty, M. T. Cancilla, J. Wang, A. A. Lugovskoy, J. C. Yoburn, et al 2005. Small-molecule inhibition of TNF-{alpha}. Science 310: 1022-1025. [Abstract/Free Full Text]
  35. Steed, P. M., M. G. Tansey, J. Zalevsky, E. A. Zhukovsky, J. R. Desjarlais, D. E. Szymkowski, C. Abbott, D. Carmichael, C. Chan, L. Cherry, et al 2003. Inactivation of TNF signaling by rationally designed dominant-negative TNF variants. Science 301: 1895-1898. [Abstract/Free Full Text]
  36. Salzmann, M., M. F. Bachmann. 1998. Estimation of maximal affinities between T-cell receptors and MHC/peptide complexes. Mol. Immunol. 35: 65-71. [Medline]
  37. Edwards, C. K., III, A. M. Bendele, L. I. Reznikov, G. Fantuzzi, E. S. Chlipala, L. Li, L. L. Moldawer, J. D. Mountz, Y. Y. Li, C. A. Dinarello. 2006. Soluble human p55 and p75 tumor necrosis factor receptors reverse spontaneous arthritis in transgenic mice expressing transmembrane tumor necrosis factor {alpha}. Arthritis Rheum. 54: 2872-2885. [Medline]
  38. Georgopoulos, S., D. Plows, G. Kollias. 1996. Transmembrane TNF is sufficient to induce localized tissue toxicity and chronic inflammatory arthritis in transgenic mice. J. Inflamm. 46: 86-97. [Medline]



This article has been cited by other articles: