Inhibition of Helicobacter hepaticus-induced colitis by IL-10 requires the p50/p105 subunit of NF-kappa B

Defects within the innate immune system sensitize NF-kappaB-deficient (p50(-/-); p65(+/-)) mice to Helicobacter hepaticus (Hh)-induced colitis. Because IL-10 plays a central role in the inhibition of Hh-induced colitis, we hypothesized that the ability of IL-10 to inhibit the innate inflammatory response to Hh may be compromised in NF-kappaB-deficient mice. To test this hypothesis, we evaluated the ability of an IL-10-Ig fusion protein with IL-10-like properties to inhibit Hh-induced colitis in RAG-2(-/-) (RAG) and p50(-/-); p65(+/-); RAG-2(-/-) (3X/RAG) mice. As expected, IL-10-Ig efficiently inhibited the development of colitis in RAG mice. In contrast, the ability of IL-10-Ig to inhibit colitis was compromised in 3X/RAG mice. The defect in response to IL-10-Ig appeared to be primarily the result of the absence of the p50/p105 subunit, because the ability of IL-10-Ig to inhibit colitis was also compromised in p50(-/-); RAG-2(-/-) (p50/RAG) mice. Radiation chimeras demonstrated that the presence of p50/p105 within hemopoietic cells of the innate immune system was necessary for efficient inhibition of colitis by IL-10-Ig. Consistent with a defect in the suppressive effects of IL-10 in the absence of p50/p105, we found that the ability of IL-10 to control LPS-induced expression of IL-12 p40 was significantly compromised in macrophages lacking p50/p105. These results suggest that the absence of the p50/p105 subunit of NF-kappaB within hemopoietic cells of the innate immune system interferes with the ability of IL-10 to suppress inflammatory gene expression and Hh-induced colitis.     

N-terminal DNA-binding domains contribute to differential DNA-binding specificities of NF-kappa B p50 and p65.

We previously reported that either oxidation or alkylation of NF-kappa B in vitro abrogates DNA binding. We used this phenomenon to help elucidate structural determinants of NF-kappa B binding. We now demonstrate that Cys-62 of NF-kappa B p50 mediates the redox effect and lies within an N-terminal region required for DNA binding but not for dimerization. Several point mutations in this region confer a transdominant negative binding phenotype to p50. The region is highly conserved in all Rel family proteins, and we have determined that it is also critical for DNA binding of NF-kappa B p65. Replacement of the N-terminal region of p65 with the corresponding region from p50 changes its DNA-binding specificity towards that of p50. These data suggest that the N-terminal regions of p50 and p65 are critical for DNA binding and help determine the DNA-binding specificities of p50 and p65. We have defined within the N-terminal region a sequence motif, R(F/G)(R/K)YXCE, which is present in Rel family proteins and also in zinc finger proteins capable of binding to kappa B sites. The potential significance of this finding is discussed.     

Common structural constituents confer I kappa B activity to NF-kappa B p105 and I kappa B/MAD-3.

The vertebrate NF-kappa B/c-rel inhibitors MAD-3/I kappa B alpha, I kappa B gamma/pdI and bcl-3 all share a conserved ankyrin repeat domain (ARD) consisting of six complete repeats, a short acidic motif and/or an incomplete seventh repeat. We present here a detailed analysis of the domain in p105/pdI and MAD-3/I kappa B involved in inhibition of DNA binding and in protein interaction with rel factors. We demonstrate that in both cases an acidic region and six ankyrin-like repeats are sufficient and required for protein interaction with the rel factors. However, for p105/pdI to achieve the high affinity needed to suppress DNA binding, an incomplete seventh repeat is required in addition. Both pdI and MAD-3 associate with rel proteins by forming heterotrimeric complexes, as shown by native gel analysis and by cross-linking. Furthermore, we demonstrate that deletion of only three amino acids in the first repeat converts the subunit specificity of the p105 ARD into that of MAD-3/I kappa B. We conclude that functionally the ARD in these molecules has a modular structure, with different subregions determining the specificity for the NF-kappa B subunits p50 and p65.     

A role for NF-kappa B subunits p50 and p65 in the inhibition of lipopolysaccharide-induced shock

To evaluate the possibility that NF-kappaB subunits p50 and p65 have a role in limiting the systemic inflammatory response induced by endotoxin, we compared the susceptibility of wild-type (WT), p65+/-, p50-/-, and p50-/-p65+/- (3X) mice to LPS-induced shock. Interestingly, whereas p65+/- mice were no more sensitive than WT mice to LPS-induced shock, 3X mice were exquisitely sensitive to the toxic effects of LPS. Mice lacking p50 alone displayed an intermediate phenotype. Sensitivity to LPS was a property of the innate immune system and was characterized by elevated circulating levels of TNF in both p50-/- and 3X mice. The ability of LPS to induce shock depended upon TNF, and 3X mice were significantly more sensitive to the toxic effects of TNF than were p50-deficient mice. The expression of several LPS-inducible proinflammatory genes, including IFN-gamma, was significantly higher within the spleens of p50-/- mice than in the spleens of WT mice, and interestingly, the expression of IFN-gamma was augmented still further within the spleens of 3X mice. These results demonstrate that NF-kappaB subunits p50 and p65 have critical inhibitory functions during the systemic response to LPS and raise the possibility that these functions could be essential in preventing mortality associated with systemic inflammatory response syndromes.     

Evidence for differential functions of the p50 and p65 subunits of NF-kappa B with a cell adhesion model.

The p50 and p65 subunits of NF-kappa B represent two members of a gene family that shares considerable homology to the rel oncogene. Proteins encoded by these genes form homo- and heterodimers which recognize a common DNA sequence motif. Recent data have suggested that homodimers of individual subunits of NF-kappa B can selectively activate gene expression in vitro. To explore this possibility in a more physiological manner, murine embryonic stem (ES) cells were treated with phosphorothio antisense oligonucleotides to either p50 or p65. Within 5 h after exposure to phosphorothio antisense p65 oligonucleotides, cells exhibited dramatic alterations in adhesion properties. Similar findings were obtained in a stable cell line that expressed a dexamethasone-inducible antisense mRNA to p65. Although antisense oligonucleotides raised against both p50 and p65 elicited a significant reduction in their respective mRNAs, only the cells treated with antisense p50 maintained a normal morphology. However, 6 days following removal of leukemia-inhibiting factor, a growth factor which suppresses embryonic stem cell differentiation, adhesion properties of cells treated with the antisense p50 oligonucleotides were markedly affected. The ability of the individual antisense oligonucleotides to elicit differential effects on cell adhesion, a property dependent upon the stage of differentiation, suggests that the p50 and p65 subunits of NF-kappa B regulate gene expression either as homodimers or as heterodimers with other rel family members. Furthermore, the finding that reduction in p65 expression alone had profound effects on cell adhesion properties indicates that p65 plays an important role in nonstimulated cells and cannot exist solely complexed with the cytosolic inhibitory protein I kappa B.     

Interaction of the C-terminal region of p105 with the nuclear localisation signal of p50 is required for inhibition of NF-kappa B DNA binding activity.

DNA binding of the homodimeric p50 subunit of NF-kappa B was inhibited by a bacterially expressed protein containing the ankyrin repeats present in the C-terminus of the p105 precursor but not by the I kappa B protein MAD-3. However p50 was retained on protein affinity matrices containing either the C-terminal ankyrin repeats of p105 or MAD-3. To investigate the interaction between p50 and proteins containing ankyrin repeats we have used a number of approaches to probe the accessibility of the p50 nuclear localisation signal in the protein complex. A monoclonal antibody recognising a linear epitope either very close to, or including, the nuclear localisation signal of the p50 protein could immunoprecipitate p50 homodimers but was unable to precipitate the protein when it was bound to the C-terminal region of p105. A close association between the nuclear localisation signal of p50 and the C-terminal region of p105 was also suggested by protease accessibility experiments. While the nuclear localisation signal of free p50 is extremely susceptible to cleavage with trypsin the same site is masked in the presence of the C-terminal ankyrin repeats of p105 and, to a lesser extent MAD-3. Removal of the nuclear localisation signal by trypsin digestion generates a protein that is fully competent for DNA binding but is refractile to inhibition by the C-terminal ankyrin repeats of p105. Addition of DNA destabilises complexes between p50 and ankyrin repeat containing proteins, increasing the susceptibility of the nuclear localisation signal to trypsin cleavage. The data suggest that there is a rapid exchange of p50 between complexes containing DNA or I kappa B proteins via a metastable complex containing DNA, p50 and I kappa B.     

Glycogen synthase kinase-3 beta regulates NF-kappa B1/p105 stability

A number of different kinases have been implicated in NF-kappa B regulation and survival function. Here we investigated the molecular cross-talk between glycogen synthase kinase-3 beta (GSK-3 beta) and the p105 precursor of the NF-kappa B p50 subunit. GSK-3 beta forms an in vivo complex with and specifically phosphorylates NF-kappa B1/p105 at Ser-903 and Ser-907 in vitro. In addition, the p105 phosphorylation level is reduced in fibroblasts lacking GSK-3 beta as compared with wild-type cells. GSK-3 beta has a dual effect on p105: it stabilizes p105 under resting conditions and primes p105 for degradation upon tumor necrosis factor (TNF)-alpha treatment. Indeed, constitutive processing of p105 to p50 occurs at a higher rate in cells lacking GSK-3 beta with respect to wild-type cells and can be reduced upon reintroduction of GSK-3 beta by transfection. Moreover, p105 degradation in response to TNF-alpha is prevented in GSK-3 beta-/- fibroblasts and by a Ser to Ala point mutation on p105 at positions 903 or 907. Interestingly, the increased sensitiveness to TNF-alpha-induced death occurring in GSK-3 beta-/- fibroblasts, which is coupled to a perturbation of p50/105 ratio, can be reproduced by p105 silencing in wild-type fibroblasts.     

Interaction of TAFII105 with selected p65/RelA dimers is associated with activation of subset of NF-kappa B genes

TAF(II)105, a substoichiometric coactivator subunit of TFIID, is important for activation of anti-apoptotic genes by NF-kappaB in response to the cytokine tumor necrosis factor (TNF)-alpha. In the present study we have analyzed the mechanism of TAF(II)105 function with respect to its regulation of p65/RelA, a component of NF-kappaB. We found two independent p65/RelA-binding domains within the N terminus of TAF(II)105. One of these domains appears to be crucial for TAF(II)105-mediated anti-apoptotic gene activation in response to TNF-alpha. Analysis of the interaction between TAF(II)105 and different NF-kappaB complexes has revealed substantial differences in the affinity of TAF(II)105 toward different p65/RelA-containing dimers. We have identified the TNF-alpha induced anti-apoptotic A20 gene as a target gene of TAF(II)105. A20 has a differential protective effect on cell death induced by TNF-alpha in the presence of either the dominant negative mutant of TAF(II)105 (TAF(II)105DeltaC) or the superdominant IkappaBalpha. The results suggest that the inhibitory effect of TAF(II)105DeltaC on NF-kappaB-dependent genes is restricted to a subset of anti-apoptotic genes while the effect of IkappaBalpha is more general. Thus, an interaction between NF-kappaB and a specific coactivator is important for specifying target gene activation.     

The NF-kappa B precursor p105 and the proto-oncogene product Bcl-3 are I kappa B molecules and control nuclear translocation of NF-kappa B.

We have examined the interaction of the NF-kappa B precursor p105 with NF-kappa B subunits. Similar to an I kappa B molecule, p105 associates in the cytoplasm with p50 or p65. Through this assembly, p105 efficiently blocks nuclear transfer of either subunit. Moreover, the p105 protein inhibits DNA binding of dimeric NF-kappa B subunits in a similar, but not identical, manner to its isolated C-terminal domain, which contains an ankyrin-like repeat domain (ARD). The proto-oncogene product Bcl-3 also controls nuclear translocation of p50, but not of p65. Hence, p50 can be retained in the cytoplasm via at least three distinct interactions: through direct interactions either with its own precursor, with Bcl-3 or indirectly through I kappa B alpha or -beta when attached to p65. We discuss a function of p105 as a cytoplasmic assembly unit for homo- and heteromeric NF-kappa B complexes and of Bcl-3 as an I kappa B with novel subunit specificity.