NF-κB is a blanket term referring to homodimers and heterodimers of a subset of the Rel family of transcription factors (1, 2). Five members of the NF-κB family have been identified in mammals: RelA (p65), RelB, and c-Rel, which contain transactivation domains, and p50 and p52, which are expressed as precursor proteins p105 (NF-κB1) and p100 (NF-κB2), respectively. The p50/p65 dimer is the most common form found in the cytoplasm of unstimulated cells, where it is usually bound to the α- or β-isoform of the inhibitor of κB (IκBα or IκBβ) via ankyrin repeats (3). Stimulation of the cell by a number of different sources (e.g., tumor necrosis factor α [TNFα], interleukin-1β [IL-1β], oxidative stress, and lipopolysaccharide [LPS]) leads to activation of the IκB kinase (IKK) complex. Activated IKK phosphorylates IκBα at Ser32 and Ser36 (or the respective serine residues on IκBα), which causes its polyubiquitination and 26S proteasomic degradation (4). Disruption of the NF-κB/IκBα complex results in the inactivation of a nuclear export signal and the accumulation of NF-κB in the nucleus (5), where it initiates gene transcription. A large number of different genes with NF-κB–binding domains in their promoters have been identified, including several proinflammatory genes (6, 7).
NF-κB activation is associated with many chronic inflammatory diseases, including rheumatoid arthritis (RA), osteoarthritis, inflammatory bowel disease, asthma, ulcerative colitis, multiple sclerosis, and atherosclerosis (1, 2). Immunohistochemical analysis of synovium from RA patients detected nuclear localization of NF-κB, which is indicative of its activation (8, 9). Experiments in cultured rheumatoid synovial fibroblasts demonstrated constitutive activation of NF-κB that was further augmented by addition of the proinflammatory cytokines IL-1β or TNFα (10, 11). Importantly, expression of a dominant-negative mutant of the IKKβ subunit of the IKK complex blocked the activation of NF-κB and the release of cytokines from these cells (12). The involvement of NF-κB in inflammatory joint disease has been also corroborated by studies in animal models, in which inhibition of NF-κB activity strongly reduced the severity of arthritis (13).
The involvement of NF-κB in inflammatory diseases and the large number of proinflammatory mediators regulated by this transcription factor makes this molecule an attractive target for antiinflammatory treatments (7, 14). Agents currently used to inhibit NF-κB signaling are small compounds that inhibit proteasome function and therefore IκBα degradation, decoy oligonucleotides, peptides that interfere with the nuclear translocation of NF-κB and, recently, small compounds that inhibit the enzymatic activity of the IKKβ subunit (14–16). However, the specificity of some of these compounds has not been conclusively determined, and the in vivo use of other biologic inhibitors may be hampered owing to the drawbacks of the currently available delivery vectors.
We recently described a novel way of regulating NF-κB activity in vitro by combining the protein-transducing domain (PTD) of the human immunodeficiency virus Tat protein with the super-repressor IκBα (Tat-srIκBα) (17). The Tat PTD, in common with similar domains found in VP22 from herpes simplex virus (18) and Antennapedia from Drosophila (19), is a region rich in positively charged amino acids that are thought to interact with negatively charged phospholipids in mammalian plasma membranes. This interaction facilitates entry of the protein into the cell. The srIκBα is a mutant form of IκBα in which Ser32 and Ser36 have been substituted for alanines. Because it cannot be phosphorylated by IKK, srIκBα binds irreversibly to NF-κB (20, 21). Previous studies have shown that proteins fused to Tat PTD are carried inside cells when added exogenously, and furthermore, these “cargo” proteins retain their biologic activity within the cell (22–25). Tat-srIκBα fusion protein was capable of entering Jurkat T cells and HeLa cells in vitro, and could be coimmunoprecipitated with p65. Furthermore, it prevented TNFα- and IL-1β–induced NF-κB–mediated transcription, demonstrating that the srIκBα portion retained its biologic function within the cell (17).
Here we describe the in vivo effects of Tat-srIκBα in the rat carrageenan-induced pleurisy model, a well-characterized model of inflammation that has previously been used for the development of antiinflammatory and antirheumatic drugs (26). Following intrapleural injection with carrageenan, proinflammatory mediators are released, and the pleural cavity fills with edematous fluid (27). In the early phase, neutrophils are the predominant infiltrating cell type, peaking in numbers at ∼6 hours; the neutrophils are later replaced by macrophages and lymphocytes (28). NF-κB activation has been demonstrated in cells recovered from pleurisy exudates at both early and late time points (28, 29). We show here that Tat-srIκBα, when administered intravenously in the early phase, can inhibit neutrophil recruitment to the pleural cavity and can modulate local production of cytokines and chemokines. This inhibition was associated with enhanced apoptosis of cells entering the pleural cavity.
- Top of page
- MATERIALS AND METHODS
The NF-κB signaling pathway has emerged as a key mediator of inflammation and therefore represents a target for the treatment of inflammatory disorders (14, 16). In RA, synovial fibroblasts cultured in vitro in the presence of TNFα underwent apoptosis when srIκB mutant was expressed by means of an adenovirus vector (44), suggesting that NF-κB plays a critical role in supporting the survival of this cell type in RA joints. That the hyperplasia in RA joint is NF-κB driven has been suggested by other studies (45). Also, the production of TNFα by macrophages, a critical cell type that controls aspects of inflammation in RA joints, was shown to be dependent on NF-κB activity (46).
An increasing number of inhibitors are currently being developed that block the NF-κB signaling pathway at various steps. Compounds that inhibit proteasome function, and therefore IκBα degradation, are used in experimental settings to assess involvement of NF-κB (15). However, because such inhibitors would be expected to block the physiologic turnover of unrelated proteins as well, nonspecific effects will most likely be a serious drawback for their use in the clinic. Use of decoy oligonucleotides to block DNA binding of NF-κB and RNA interference technology to modulate the expression of genes that participate in the NF-κB pathway (i.e., IKKβ) could be more specific than proteasome inhibitors (14, 47). Their use, however, could be limited because of the need for safe and efficient delivery systems, and although virus vectors are able to achieve such expression, they also have the potential to induce NF-κB activation, hence curtailing the effects of the inhibitor and complicating the interpretation of the results. Another class of recently developed small-molecule inhibitors blocks NF-κB activation by inhibiting the IKKβ subunit of the IKK complex (16). As is the case for other kinase inhibitors, their specificity for IKKβ has to be demonstrated before they can be used as antiinflammatory drugs.
To directly and specifically inhibit NF-κB action, we previously constructed a membrane-penetrating form of the super-repressor IκBα, Tat-srIκBα, and showed that, when added exogenously to cells in culture, the repressor was able to inhibit the biologic activity of NF-κB (17). Here, we extended our studies by investigating the effects of the Tat-srIκBα chimera in an in vivo model of inflammation. When injected intravenously shortly before administration of the inflammatory stimulus, Tat-srIκBα reduced the number of leukocytes migrating from the bloodstream to the site of inflammation. In addition, inflammatory cells recovered from the pleural cavity of Tat-srIκBα–treated animals displayed elevated caspase activity compared with those from controls, suggesting that, in addition to their reduced migratory response, these cells were more prone to apoptosis. These results are consistent with studies that demonstrate activation and a prosurvival role for NF-κB in neutrophils following stimulation of the cells with proinflammatory cytokines or type I interferon (30, 48).
Accelerated apoptosis is one possible mechanism through which Tat-srIκBα reduces neutrophil numbers in the pleural cavity when given intravenously. Neutrophils that take up the inhibitor in the bloodstream and are then exposed to proinflammatory stimuli are prone to apoptosis because of their inability to activate NF-κB and, hence, are less capable of transmigration. Alternatively, because adhesion molecule promoters contain NF-κB–binding domains (E-selectin, intercellular adhesion molecule 1, and vascular cell adhesion molecule 1 ), it is possible that failure to up-regulate the expression of such molecules on vascular endothelial cells may contribute to the decrease in the number of cells that accumulate at sites of inflammation. Consistent with this model, it was recently demonstrated that NF-κB action in cells of nonhematopoietic origin is important for leukocyte recruitment during LPS-induced pneumonia (50).
In contrast to the effects of systemic administration, when the Tat-srIκBα inhibitor was delivered locally, only a marginal reduction in neutrophil migration was observed following induction of inflammation. Therefore, the route of administration can determine the potency of inhibition in this in vivo model. A plausible explanation for this difference is that when administered intravenously, Tat-srIκBα is taken up by circulating cells, blocking the activity of NF-κB possibly before the cells are exposed to stimuli generated by the intrapleural injection of carrageenan. In contrast, when Tat-srIκBα is administered intrapleurally, neutrophils are most likely exposed to significant doses of the inhibitor after they are activated and induced to migrate into the pleural cavity.
As expected, production of the key proinflammatory cytokines TNFα and IL-1β was reduced in pleural exudates from Tat-srIκBβ–treated animals, although it was not completely abolished. Lack of complete inhibition could be due to the action of regulatory transcription factors other than NF-κB, as has been reported for the IL-1β genes, for which it was shown that NF-κB is only a component that can amplify a core inducible activity regulated by CCAAT/enhancer binding protein β (NF-IL6) (38, 39). Therefore, complete inhibition of transcription of certain proinflammatory cytokines in vivo may depend upon targeting other transcription factors as well.
Surprisingly, the concentrations of the neutrophil chemoattractants CINC-1 and CINC-3 were strongly increased. Over the course of these experiments, we noticed that exudates with the lowest number of infiltrating cells contained the highest levels of CINC-3. This may indicate that inflammatory neutrophils that infiltrate the site of inflammation are able to down-regulate the production of CINC-1 and CINC-3, and possibly other mediators, by resident cells. This type of down-regulation could represent a feedback mechanism that prevents continuous infiltration of cells. Similar observations in a model of murine LPS-induced neutrophil recruitment to bronchoalveolar spaces have been reported (51). Therefore, membrane-permeable inhibitors of signaling cascades delivered in vivo may be useful tools for uncovering novel biologic mechanisms.
In summary, our results demonstrate the importance of NF-κB in neutrophil migration and survival during inflammation and establish a methodology that can be applied to regulate signaling pathways in vivo. This methodology opens the way for the development of a new type of antiinflammatory agent that can be used in RA and other inflammatory disorders.