Allele | panr2 | |||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Mutation Type | missense | |||||||||||||||||||||||||||||||||||||||||||||
Chromosome | X | |||||||||||||||||||||||||||||||||||||||||||||
Coordinate | 73,480,828 bp (GRCm39) | |||||||||||||||||||||||||||||||||||||||||||||
Base Change | T ⇒ C (forward strand) | |||||||||||||||||||||||||||||||||||||||||||||
Gene | Ikbkg | |||||||||||||||||||||||||||||||||||||||||||||
Gene Name | inhibitor of kappaB kinase gamma | |||||||||||||||||||||||||||||||||||||||||||||
Synonym(s) | IKK[g], 1110037D23Rik, NEMO | |||||||||||||||||||||||||||||||||||||||||||||
Chromosomal Location | 73,436,896-73,497,460 bp (+) (GRCm39) | |||||||||||||||||||||||||||||||||||||||||||||
MGI Phenotype |
FUNCTION: [Summary is not available for the mouse gene. This summary is for the human ortholog.] This gene encodes the regulatory subunit of the inhibitor of kappaB kinase (IKK) complex, which activates NF-kappaB resulting in activation of genes involved in inflammation, immunity, cell survival, and other pathways. Mutations in this gene result in incontinentia pigmenti, hypohidrotic ectodermal dysplasia, and several other types of immunodeficiencies. A pseudogene highly similar to this locus is located in an adjacent region of the X chromosome. [provided by RefSeq, Mar 2016] PHENOTYPE: Males hemizygous for targeted null mutations exhibit embryonic lethality by embryonic day 13.5 from apoptotic liver damage. Heterozygous females show patchy skin lesions with granulocyte infiltration, growth retardation, and shortened lifespan. [provided by MGI curators] |
|||||||||||||||||||||||||||||||||||||||||||||
Accession Number | ||||||||||||||||||||||||||||||||||||||||||||||
Mapped | Yes | |||||||||||||||||||||||||||||||||||||||||||||
Amino Acid Change | Leucine changed to Proline | |||||||||||||||||||||||||||||||||||||||||||||
Institutional Source | Beutler Lab | |||||||||||||||||||||||||||||||||||||||||||||
Gene Model | not available | |||||||||||||||||||||||||||||||||||||||||||||
AlphaFold | O88522 | |||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000004330 Gene: ENSMUSG00000004221 AA Change: L171P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000064017 Gene: ENSMUSG00000004221 AA Change: L152P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.996 (Sensitivity: 0.55; Specificity: 0.98) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000109762 Gene: ENSMUSG00000004221 AA Change: L153P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000109763 Gene: ENSMUSG00000004221 AA Change: L153P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000109764 Gene: ENSMUSG00000004221 AA Change: L152P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.996 (Sensitivity: 0.55; Specificity: 0.98) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000109765 Gene: ENSMUSG00000004221 AA Change: L165P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.998 (Sensitivity: 0.27; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000109768 Gene: ENSMUSG00000004221 AA Change: L153P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000116875 Gene: ENSMUSG00000004221 AA Change: L153P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000116615 Gene: ENSMUSG00000004221 AA Change: L153P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000120080 Gene: ENSMUSG00000004221 AA Change: L165P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000116230 Gene: ENSMUSG00000004221 AA Change: L19P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000126770 Gene: ENSMUSG00000004221 AA Change: L153P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.999 (Sensitivity: 0.14; Specificity: 0.99) |
|||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000136573 Gene: ENSMUSG00000004221 AA Change: L152P
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably damaging
PolyPhen 2 Score 0.996 (Sensitivity: 0.55; Specificity: 0.98) |
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably benign | |||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000118884 Gene: ENSMUSG00000004221
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably benign | |||||||||||||||||||||||||||||||||||||||||||||
SMART Domains |
Protein: ENSMUSP00000115829 Gene: ENSMUSG00000004221
|
|||||||||||||||||||||||||||||||||||||||||||||
Predicted Effect | probably benign | |||||||||||||||||||||||||||||||||||||||||||||
Meta Mutation Damage Score | Not available | |||||||||||||||||||||||||||||||||||||||||||||
Is this an essential gene? | Not available | |||||||||||||||||||||||||||||||||||||||||||||
Phenotypic Category | X-linked Recessive | |||||||||||||||||||||||||||||||||||||||||||||
Candidate Explorer Status | loading ... | |||||||||||||||||||||||||||||||||||||||||||||
Single pedigree Linkage Analysis Data |
|
|||||||||||||||||||||||||||||||||||||||||||||
Penetrance | 100% | |||||||||||||||||||||||||||||||||||||||||||||
Alleles Listed at MGI | All alleles(13) : Targeted, knock-out(3) Targeted, other(3) Gene trapped(6) Chemically induced(1) | |||||||||||||||||||||||||||||||||||||||||||||
Lab Alleles |
|
|||||||||||||||||||||||||||||||||||||||||||||
Mode of Inheritance | X-linked Recessive | |||||||||||||||||||||||||||||||||||||||||||||
Local Stock | Live Mice, Embryos, gDNA | |||||||||||||||||||||||||||||||||||||||||||||
MMRRC Submission | 030659-UCD | |||||||||||||||||||||||||||||||||||||||||||||
Last Updated | 2017-03-28 2:17 PM by Katherine Timer | |||||||||||||||||||||||||||||||||||||||||||||
Record Created | unknown | |||||||||||||||||||||||||||||||||||||||||||||
Record Posted | 2010-07-02 | |||||||||||||||||||||||||||||||||||||||||||||
Phenotypic Description | The panr2 ("pan-resistance") phenotype was detected in a screen of G3 progeny of N-ethyl-N-nitrosourea (ENU) mutagenized C57BL/6J males for altered responses to Toll-like receptor (TLR) ligands (TLR Signaling Screen) (1). Peritoneal macrophages from panr2 hemizygous male mice exhibit impaired responses to ligands for TLR3 (poly(I:C), TLR4 (LPS), TLR7 (R-848) and TLR9 (CpG oligodeoxynucleotides), and the heterodimers TLR1/2 (Pam3CSK4) and TLR2/6 (MALP-2, peptidoglycan). Multiple cytokines were affected, including TNF-α, IL-6, IL-12p40 and MCP-1, as well as the inflammatory mediator nitric oxide. Secretion of type 1 interferon (IFN) in response to TLR3 and TLR4 stimulation was not affected (Figure 1), nor was the production of type I IFN in response to double-stranded DNA (Double-stranded DNA Macrophage Screen). Panr2 mice display normal natural killer (NK) cell and cytotoxic T cell (CTL) function (In vivo NK cell and CD8+ T cell cytotoxicity screen).
Panr2 mice are highly susceptible to an otherwise sublethal dose of murine cytomegalovirus (MCMV; see the MCMV Susceptibility and Resistance Screen; Figure 2A). Death within six days of MCMV infection was indicative of an innate immune defect, since combined deficiency of B and T cells permits survival of a similar infection for more than two weeks (2). Similarly, panr2 mice succumbed to infection with the intracellular pathogen Listeria monocytogenes within four days (Figure 2B).
Examination of immune cell types in these animals revealed a selective reduction in memory, regulatory T cells (Tregs) and natural killer T (NKT) cells despite normal T cell development in the thymus (Figure 3). B cell development is also normal in these animals, but B cell activation is strongly impaired (Figure 4). The serum of unimmunized panr2 mice was deficient for all immunoglobulin (Ig) isotypes tested, including IgM. After immunization with a recombinant Semliki Forest virus (rSFV) expressing β-gal vector, a strong inducer of antibody responses in wild type mice, panr2 mice produced no specific IgG or IgM antibody.
Male panr2 mice are born at normal ratios. Histological analysis showed no defects in the skin, liver or intestinal epithelium, but male mice are azoospermic with cystic formations in the seminiferous tubules, as well as abnormal maturation of spermatids and a mean testis weight approximately 35% lower than wild type siblings (Figure 5). Inguinal lymph nodes were also absent or of diminished size in most mutant mice. |
|||||||||||||||||||||||||||||||||||||||||||||
Nature of Mutation | The Ikbkg gene on the X chromosome of panr2 mice was sequenced, and a T to C transition was identified in exon 4 (of 10 total exons) at position 588 of the Ikbkg transcript (using variant 1, Genbank: NM_178590). The mutation was identified in a hemizygous male.
457 TCATTGCTCGGAGAACTCCAGGAGAGCCAGAGC
148 -S--L--L--G--E--L--Q--E--S--Q--S-
The mutated nucleotide is indicated in red lettering, and results in a leucine to proline change at amino acid 153 of the IKK-γ or NEMO protein.
| |||||||||||||||||||||||||||||||||||||||||||||
Illustration of Mutations in Gene & Protein |
||||||||||||||||||||||||||||||||||||||||||||||
Protein Prediction | The 412-amino acid IKK-γ (inhibitor of kappa-B kinase-γ) or NEMO (NF-κB essential modulator) protein is a component of the IKK complex and is necessary for NF-κB activation (see Background) (3;4). The human NEMO protein is 419-amino acids long and has 86% identity to its mouse homologue (4). The IKBKG gene also encodes an N-terminally truncated form of the protein translated from methionine-38, but this isoform is not fully functional (5).
NEMO is composed of two coiled-coil (CC) domains located approximately at amino acids 63-194 and 250-284, a conserved ubiquitin binding sequence (aa 285-315) that is part of the NEMO ubiquitin binding (NUB) domain, a leucine zipper (LZ) motif (~aa 315-336), a proline-rich region (~aa 357-388) and a zinc finger (~aa 389-410) (3;4;6-9) (Figure 6A). Two smaller α-helical regions (αH1, αH2) are located before and after the CC1 domain. The CC2 and LZ domains are required for NEMO oligomerization, in which three LZ helices dock in the crevices defined by the external interfaces of the CC2 domains to form a six helix bundle (10). A coiled-coil domain spanning amino acids 47 to 80 and consisting of three α-helical subdomains also contributes to NEMO oligomerization, and is crucial for dimerization (11). The NUB domain encompasses parts of CC2 and the LZ, in addition to the conserved sequence mentioned above. This domain is also known as UBAN (ubiquitin binding in ABIN and NEMO), CoZi (CC2 and LZ) and NOA (NEMO, opineurin, ABIN) [reviewed by (12)]. The roles of these domains in NEMO function are further described below.
As a scaffolding protein, NEMO has been shown to interact with many proteins important in NF-κB activation or down-regulation [reviewed by (13;14)]. Through its αH1 and CC1 domains, NEMO binds to the IKK-1 and IKK-2 kinases of the IKK complex, the human T-cell leukemia virus type 1 Tax protein, or protein phosphatases (4;7;15;16), although the C-terminal region of NEMO has also been reported to be important for Tax interaction (6). This region of NEMO has also been reported to bind to other regulators of the IKK complex including the ubiquitin E3 ligase TNF receptor associated factor 6 (TRAF6) (17), the T and B cell receptor (TCR and BCR) scaffolding protein caspase recruitment domain 11 (CARD11; also known as CARMA1; see the record for king) (18), the adaptor protein CIKS/Act1 (19), the Ataxia telangiectasia mutated (ATM) kinase, protein inhibitor of activated STATy (PIASy) (20) and poly (ADP-ribose)-polymerase 1 (PARP1) that assembles PIASy, ATM and NEMO into a complex, the A20 binding inhibitor of NF-κB-1 (ABIN-1) (21), and NESCA, an adaptor involved in the neural growth factor (NGF) pathway (22). The stretch of amino acids located between the CC1 and CC2 domains interacts with the TNF receptor adaptor proteins receptor-interacting protein (RIP) and TANK, the viral FLICE inhibitory protein (v-FLIP), and the NF-κB inhibitors A20, ABIN-2 and the PP2A phosphatase (23-26), while the LZ domain binds to the TFG (Trk-fused gene) protein involved in oncogenic rearrangements (27). Residues between the LZ and the zinc finger bind to the bacterial E3 ubiquitin ligase IpaH9.8 (28), and the zinc finger interacts with the deubiquitinating enzyme CYLD (29). NEMO has also been shown to bind to K63-ubiquitinated proteins in the NUB domain region encompassing the CC2 and LZ domains (30;31). These proteins include RIP, the CARD11-associating protein Bcl10, and the interleukin-1-receptor-associated kinase 1 (IRAK1), which functions downstream of the IL-1 receptor and TLR signaling (30;32;33). Mutations of conserved ubiquitin-binding hydrophobic residues within this region abolish or reduce NF-κB activation (34-37). The NUB domain also binds to the the linear ubiquitin chain assembly complex (LUBAC) composed of the E3 ligases HOIL-1 and HOIP (38). Finally, a C-terminal domain of the protein encompassing the LZ and zinc finger domains binds to the positive regulator v-CLAP (viral CARD-like apoptotic protein), the viral counterpart of Bcl10 (39), and to the NF-κB inhibitor CSN3, a subunit of the COP9 signalosome (CSN) involved in the ubiquitin-proteasomal pathway (40).
Many of the NEMO-interacting proteins regulate NF-κB activity by modifying NEMO in response to various stimuli including genotoxic stress, TLR signaling, and T and B cell receptor signaling (41-43). These modifications include phosphorylation (44-46), ubiquitination, and sumoylation (28;38;43;47-49). K63- and linear ubiquitination of NEMO by various E3 ubiquitin ligases has been demonstrated to be critical for NF-κB activation. TRAF6 ubiquitinates NEMO on multiple sites including lysines 278, 314, 318, 319 and 392 (42;47), while LUBAC attaches linear ubiquitin to NEMO at residues 278 and 302 (38). Other E3 ubiquitin ligases have been implicated in NEMO K63-linked ubiquitination including Parkin and cellular inhibitor of apoptosis protein 1 (c-IAP1) (50). In humans, mutations in Parkin cause juvenile Parkinson disease (OMIM #600116) (51). The bacterial IpaH9.8 protein also ubiquitinates NEMO, but does so in a K27-linked manner on residues Lys 302 and Lys 314 resulting in protein degradation (28). It is likely that the CSN signalosome also contributes to NEMO degradation by regulating its ubiquitination, thus inhibiting NF-κB activation (40). Deubiquitination of K63-linked NEMO is also a major mechanism of down-regulating NF-κB activity. A20, complexed with ABIN-1, deubiquitinates NEMO, as does CYLD (18;21;29). The zinc finger factor ZNF216, an A20-like protein, can bind to polyubiquitin chains and may also regulate the ubiquitination of NEMO or interfere with the binding of NEMO to ubiquitinated proteins (26). Other NF-κB inhibitors like ABIN-2, CARD8 prevent the association of NEMO with positive regulators (25;52). NEMO is phosphorylated by IKK on amino acids 31, 43 and 369 (45) and by ATM on residue 85. Ser 85 phosphorylation is preceded by NEMO sumoylation at residues Lys 277 and Lys 309 by PIASy (20;41;49).
The structures of various NEMO domains have recently been investigated. The minimal IKK associating domain of NEMO is localized to amino acids 44-111, and a crystal structure of this fragment in complex with the NEMO binding domain of IKK-2 revealed a heterotetrameric complex with an elongated and atypical parallel four-helix bundle (Figure 7A; PDB 3BRV). Each NEMO molecule is a crescent-shaped α helix, which packs head-to-head to form a dimerization interface at the N- and C- termini (residues 51-65 and 103-107) that forms an aperture running down the length of the bundle. The two IKK peptides bridge across the NEMO helices, on both sides of this aperture, and bind to cavities lined by NEMO residues 85-101 mainly through hydrophobic interactions with Leu 93, Phe 92, Met 94, Phe 97, Ala 100 and Arg 101 or via hydrogen bonding with Ser 85, Glu 89 and Arg 101. The NEMO dimer is stabilized by interactions at the N- and C- termini. At the N-termini these interactions include residues Leu 51, Cys 54, Glu 57, Leu 61, and Arg 62 from each monomer. At the C-termini, residues Leu 103, Val 104 and Leu 107 are involved (53). The NMR structure of the zinc finger domain reveals a global ββα fold that forms two surfaces, a large hydrophobic face defined mostly by one side of the α-helix and the β1 strand, and a polar face consisting of the very acidic side of the helix and the β2 strand (54). It is likely that both surfaces are involved in protein-protein interactions (Figure 7B; PDB 2JVX).
In addition to the IKK associating and the zinc finger domains of NEMO, the structure of the ubiquitin binding NUB region has also been determined. Although NEMO is strongly implicated in binding to and recognizing K63-ubiquitinated proteins, recent evidence suggests that the NUB domain preferentially binds to linear ubiquitin (35;38;55). These observations can be reconciled through recent evidence that suggests that NEMO recognizes K63-linked ubiquitin with high affinity using a bipartite domain consisting of the NUB domain and the zinc finger, which can also bind ubiquitin molecules through its hydrophobic face (31;56). Several crystal or NMR structures of the NEMO NUB domain have recently been reported, either alone or in complex with linear and K63-linked diubiquitin (35-37;55). A structure of the NUB domain complexed with linear diubiquitin revealed a parallel coiled-coil dimer that forms a heterotetrameric complex (Figure 7C; PDB 2ZVO). Each NEMO dimer binds two ubiquitin molecules referred to as Ubdistal (more N-terminal) and Ubproximal (more C-terminal) with both protomers contributing to each of the diubiqutin binding surfaces comprised of residues Glu 289 to Glu 320. The Ubproximal binds to one of the two NEMO protomers using mainly polar interactions with residues Arg 309, Arg 312, Glu 313, Glu 317 and Glu 320, while the interactions with Ubdistal are mainly hydrophobic and involve Val 293, Tyr 301, Lys 302 and Phe 305. A salt bridge using NEMO residue Asp 304 also contributes to this binding. Ubiquitin molecules contain a hydrophobic patch centered on Ile 44 that is usually recognized by ubiquitin-binding domains. NEMO recognizes this region in Ubdistal, but not in Ubproximal. Ubiquitin binding caused a straightening of the NEMO helices, a structural change that may extend throughout the protein and alter additional interactions with other proteins (35). The binding of the NUB domain to K63-linked diubiquitin revealed a similar parallel coiled-coil dimeric structure of the NEMO domains. However, unlike linear diubiquitin, both K63-linked ubiquitin molecules bound to the same region of each NEMO monomer (amino acids 293–308) and link two NEMO dimers together (Figure 7D; PDB 3JSV). NEMO recognizes the canonical Ile 44-centered hydrophobic patch for both ubiquitin molecules (36).
The panr2 mutation results in a leucine to proline substitution at amino acid 153 of the NEMO protein. This residue is highly conserved, and occurs in the CC1 domain. The affected protein is expressed normally in macrophages (Figure 6B).
| |||||||||||||||||||||||||||||||||||||||||||||
Expression/Localization | ||||||||||||||||||||||||||||||||||||||||||||||
Background | The NF-κB signaling pathway functions in essentially all mammalian cell types and is activated in response to injury, infection, inflammation and other stressful conditions requiring rapid reprogramming of gene expression (Figure 8). The NF-κB family of transcription factors consists of the evolutionary conserved proteins p65/RelA, c-Rel, RelB, p50 (derived from the p105 precursor) and p52 (derived from the p100 precursor; see the record for xander). In the resting cell, NF-κB dimers are kept inactive in the cytoplasm through their association with IκB inhibitory molecules, including p105 and p100. In response to stimulation, IκBs are phosphorylated by the IκB kinase (IKK) complex at conserved serine residues. This modification induces the K48-linked polyubiquitination of IκB molecules and subsequent recognition by the 26S proteasome as substrates for proteolysis. Degradation of IκBs allows the NF-κB dimers to translocate into the nucleus, where they are able to activate the transcription of target genes, including various cytokines [for review see (58)].
NF-κB regulates both innate and adaptive immune responses, including those mediated by the TLR receptors, TNF receptors, the IL-1 receptor, as well as BCR and TCR. Upon stimulation, these receptors recruit adaptor proteins and activate immune complexes leading to IKK complex activation. TLR signaling depends on two pathways utilizing different adaptor proteins, the myeloid differentiation (MyD) 88-dependent pathway (see pococurante and lackadaisical) acting in conjunction with MAL (MyD88-adaptor like; see torpid), and the TRIF (Toll-interleukin 1 receptor (TIR) domain-containing adaptor inducing IFN-β)/TRAM (TRIF related adaptor molecule)-dependent pathway (see records for Lps2 and Tram KO). The MyD88-dependent pathway leads to early NF-κB activation, while the TRIF-dependent pathway results in delayed NF-κB activation as well as the production of type I IFNs via TBK1 and IKK-ε. Most TLRs signal through the MyD88-dependent pathway, except for TLR3 and TLR4. TLR4 uses both MyD88 and TRIF adaptor proteins, while TLR3 is dependent upon TRIF alone. MyD88-dependent signaling requires the recruitment of the IRAK1 and IRAK4 (see otiose) to the receptor, while both MyD88 and TRIF recruit TRAF6. TRAF6 then activates the TGF-β–activated kinase (TAK) 1 resulting in subsequent phosphorylation and activation of the IKK complex kinases [reviewed in (65;66)]. BCR and TCR-induced NF-κB activation requires costimulatory signals and requires the formation of a lipid raft-associated multiprotein complex that includes the scaffolding proteins CARD 11, Bcl10 and MALT1 (Mucosa-Associated Lymphoid tissue lymphoma Translocation-associated gene 1) (67-69). Similar to the TLR pathway, TRAF6 is required downstream of the Bcl10 complex in order to ubiquitinate NEMO at various lysine residues, modifications that are necessary for IKK complex activation (43;47;48). In contrast, both the tumor suppressor CYLD (downstream of TNF) and the NF-κB inhibitor A20 (downstream of TNF and TCR signaling) have been shown to deubiquitinate NEMO and down-regulate NF-κB activation (18;21;29).
Similar to IKK-2 and p65/RelA NF-κB knockouts (70-72), NEMO-deficient male mice die during gestation from uncontrolled liver apoptosis (60;73;74). Interestingly, embryonic death occurs earlier in these animals than in IKK-2 or RelA-deficient mice. Accordingly, NEMO-deficient cells are completely defective in NF-κB responses in response to a wide variety of stimuli. These results are explained by the partial functional redundancy with RelA provided by other NF-κB subunits and that IKK-1 partially compensates for the lack of IKK-2 (62;75). Heterozygous NEMO knockout females contain both wild-type and NEMO-deficient cells in a mosaic pattern due to X-inactivation. These females develop inflammatory skin lesions shortly after birth that resolve over time (73;74). Keratinocyte-specific NEMO and IKK-2 knockout animals develop similar skin lesions after birth (76;77). The lack of NF-κB activation results in susceptibility of cells to TNF-α-induced apoptosis, causing the hepatocyte cell death seen in male embryos and the skin lesions seen in heterozygous females and keratinocyte-specific IKK knockouts (60;73;74;76;77).
Tissue-specific deletion of IKK complex components in other tissues has provided further understanding of the role NF-κB activation plays in several cell and tissue types including B cells, T cells, gut epithelium, and liver. B-cell-specific knockouts of both IKK-2 and NEMO resulted in a lack of B cells in the spleen suggesting that canonical NF-κB activation is required for the maintenance of mature B cells (78), while the development of B cells is dependent on non-canonical NF-κB activation by IKK-1 and NIK (79;80). Mice with T-cell-restricted NEMO ablation or with replacement of IKK-2 with a kinase-dead mutant prevented survival of peripheral T cells (81). Ablation of NEMO or both IKK-1 and IKK-2 in gut epithelium caused severe chronic intestinal inflammation in mice, suggesting that NF-κB activation is required to control epithelial integrity and the interaction between the mucosal immune system and gut microflora. In this model, MyD88 or TNFR1 deficiency prevented the development of intestinal inflammation, demonstrating that TLR activation by intestinal bacteria and TNF-α signaling is essential for the development of intestinal inflammation caused by NF-κB inactivation (82). Deletion of NEMO in liver parenchymal cells causes the spontaneous development of hepatocellular carcinoma triggered by NEMO-deficient hepatocyte death (83). Finally, in a mouse model of multiple sclerosis, CNS-specific ablation of IKK-2 or NEMO ameliorated the disease by preventing the expression of NF-κB regulated inflammatory cytokines and other molecules (84). The phenotypes seen in NEMO-deficient, IKK-2, and IKK-1 knockout mice are partially recapitulated by the transgenic overexpression of repressor or dominant-negative proteins of NF-κB activation [reviewed in (61)].
Several human diseases are now known to be caused by mutations in the IKBKG gene with symptoms partially dependent on the level of NF-κB inactivation (85). Mutations that completely disrupt NEMO function and NF-κB activation result in incontinentia pigmenti (IP; OMIM #308300), a rare X-linked multisystem disorder presenting almost exclusively in females as most males die in utero (85;86). The disorder is highly variable but characterized by a distinctive swirling pattern of the skin; defects of teeth, hair, and nails; and ophthalmic, central nervous system, and musculoskeletal abnormalities. The skin phenotype is the most common defect and is caused by resolution of inflammatory skin lesions identical to those seen in NEMO-deficient mice (73;74;76), eventually resulting in hyperpigmented streaks caused by the loss of melanin from dying basal keratinocytes and its deposition within dermal macrophages. Females with IP rarely have immunological defects due to skewed X-inactivation in the peripheral blood cells. Skewed X-inactivation also occurs in hepatocytes. It is likely that males carrying these mutation die in utero due to hepatocyte apoptosis, similar to NEMO-deficient male mouse embryos (60;73;74).
Mutations in NEMO resulting in partially functional protein result in male survival with varying phenotypes including hypohidrotic (anhydrotic) ectodermal dysplasia and immunodeficiency (HED-ID; OMIM #300291) or immunodeficiency without HED (87;88). HED or EDA is a rare developmental syndrome in which patients present with missing or sparse hair, missing or misshapen teeth, and absent or reduced ability to sweat. The HED phenotype is caused by lack of NF-κB activation by the EDA (ectodysplasin) signaling pathway specific to the development of the so-called skin appendages, ectoderm-derived tissues including hair follicles, nails, teeth and exocrine glands in mammals. EDA signals through a TNF-related pathway involving the EDA receptor (EDAR; see achtung2) and the adaptor protein EDARADD (EDAR-associated death domain; see achtung and gizmo). Mutations in all of these factors causes HED without immunodeficiency in humans (OMIM #129490, #224900, #305100) (89). In patients with HED-ID, mutations in NEMO affect the EDA pathway, as well as NF-κB activation in the immune system. The immunodeficiency phenotypes in patients with hypomorphic NEMO alleles are variable, but include persistent, life-threatening bacterial and viral infections associated with lack of NK cell function and variable impairment of CD40, TLR/IL-1, TNF-α, and TCR signaling. Some patients also exhibit a hyper-IgM phenotype, and have defects in class switching (88;90)
In response to genotoxic stress such as DNA double-strand breaks, NEMO becomes sumoylated by PIASy resulting in nuclear relocation, and subsequent phosphorylation by ATM followed by mono-ubiquitination. This results in the shuttling of NEMO to the cytoplasm where it facilitates the formation of an ATM-IKK complex resulting in the activation of the IKK complex and the induction of NF-κB (41;49). An additional protein, p53-inducible death domain-containing protein (PIDD), appears to be involved in DNA damage-induced activation of NF-kB independent of the ATM pathway and also induces the symoylation and ubiquitination of NEMO in the nucleus (91). As DNA damage is one of the causative factors in the pathogenesis of premature aging (progeroid) syndromes such as Werner’s and Hutchinson-Gilford’s (OMIM #277700, #176670), it is possible that NF-κB activation through NEMO plays a role in these diseases. Patients with Werner syndrome often present with inflammatory disorders that are potentially caused by NF-κB activation (92).
| |||||||||||||||||||||||||||||||||||||||||||||
Putative Mechanism | Studies of panr2 males reveal that NEMO is important for the development of lymph nodes. This supports the observations that IKK-mediated phosphorylation of the p50 precursor, p105, is required for complete lymph node development (93), as are the NF-κB subunits p50 and p52 (94). Activation of both subunits via signaling from LTβR is required for secondary lymphoid organogenesis, with combined p50/p52 deficiency preventing lymph node development altogether (94). Collectively, these data suggest that canonical NF-κB activation through the IKK complex is important for the development of lymphoid architecture. On the other hand, azoospermia has not been associated with mutations of the IKK complex. However, NF-κB activity is developmentally regulated during spermatogenesis in mice, andIκBα and β are both expressed in the testes (95).
Conditional deletion of Ikbkg is known to affect a variety of tissues, including the skin (76), liver (83) and intestinal epithelium (82). In addition, the survival of T (81) and B (78) lymphocytes is precluded by conditional deletion of Ikbkg. However, these tissues and immune cells are normal or present in panr2 mice, although the survival of certain T cell subsets and B cell function are compromised. An absence of ectodermal dysplasia, which develops in mice with mutations of Eda (96), Edar (97) and Edaradd (98), suggests that the panr2 mutation is permissive to signals transduced from EDAR, but not from the TLRs, CD40, LTβR, BCR and/or TCR. The specific defects seen in memory, Tregs and NKT cells in panr2 mice reflects the phenotype observed in mice with the T cell-specific ablation of IKK-2 (81) or in mice with an engineered mutation of p105 that cannot be phosphorylated by IKK (Nfkb1SSAA/SSAA) (93).
To characterize the biochemical consequences of the panr2 mutation, NF-κB and MAPK signaling events were examined in TLR-stimulated macrophages (Figure 9). In response to both LPS (Figure 8A) and MALP-2 (not shown), p38 MAPK phosphorylation occurred normally in panr2 cells, yet phosphorylation of p105, MEK (MAPK or ERK kinase) and ERK was severely impaired and p105 protein levels were stabilized. In spite of a mild impairment of IκBα phosphorylation, IκBα degradation remained intact. However, nuclear translocation of p65 was suppressed in panr2 cells (Figure 9C), with accumulation occurring instead in the cytoplasmic fraction.
The impaired phosphorylation and stabilization of p105 leads to the subsequent inhibition of TPL2 kinase activity, which likely explains the observed defects in MEK/ERK activation. The TPL2/MEK/ERK pathway promotes the secretion of TNF-α, rather than the accumulation of the 26 kDa membrane-associated TNF-α precursor protein inside the cell (99). To examine the accumulation of TNF-α inside panr2 cells, macrophages were stimulated with LPS in the presence or absence of the secretory pathway inhibitor brefeldin A. Brefeldin A-treated panr2 cells accumulated roughly 2-fold less intracellular TNF-α, in contrast to a 3-4 fold reduction of surface TNF-α (Figure 9B), and a 5-fold reduction of TNF-α in the supernatant (Figure 1). This disparity between intracellular and secreted TNF is consistent with observations in TPL2-deficient (99) and mutant (100) cells, but also imply that synthesis of preTNF is affected independently of TPL2.
The biochemical results of the panr2 mutation suggest that IκBα degradation is not by itself sufficient for immune competence, at least in macrophages, since TLR-induced IκBα degradation proceeds normally in panr2 cells. Nuclear translocation of p65 is severely affected, however, implying a requirement for NEMO beyond IκB degradation. In agreement with this finding, cytoplasmic retention of the majority of p65 can occur in the combined absence of IκBα, IκBβ and IκBε (101). The molecule(s) that retain p65 in the cytoplasm in this context are not known, although greater quantities of p100 and p105 can be recovered by p65 immunoprecipitation. As p105 phosphorylation and degradation are impaired in TLR-stimulated panr2 cells, this may account for p65 cytoplasmic retention in the presence of IκBα degradation.
TRAF6-mediated ubiquitination of NEMO is also necessary for complete activation of TLR-induced cytokine secretion. However, mice lacking the target lysine residue have intact TLR-induced IκBα degradation, NF-κB activation and ERK phosphorylation, but an impaired cytokine response (42), suggesting that NEMO ubiquitination may regulate processes beyond IκBα degradation and ERK activation. The panr2 mutation, in contrast, uncouples ERK phosphorylation and nuclear translocation of p65 without impairing IκBα degradation. This is likely caused by alteration of the scaffolding function of NEMO, disrupting some but not all functions of NEMO and the IKK complex. This separation of function may offer a mechanistic explanation for rare immune deficiencies in humans. Indeed, the variable phenotypes seen in humans with mutations in the IKBKG gene are due to the location of the individual mutations and the involvement of NEMO in multiple protein interactions and pathways. Examination of 72 patients carrying 32 different NEMO mutations allowed the identification of NEMO regions involved in the development of HED, the hyper-IgM phenotype, and other immunological defects (88). HED was attributed to the C-terminal portion of the CC domains and adjacent regions, and the C-terminal 30 amino acids including the ZF. The very N-terminus of NEMO is not required for NF-κB activation in skin appendage development as patients with this mutation do not show signs of HED (88;102). This is also the case for a small deletion in the CC2 domain (Δ271-276 in human NEMO) and a larger deletion in the proline-rich region (Δ351-373). Presumably, these regions of NEMO are not necessary for interactions with proteins needed for NF-κB activation in skin appendages, but are needed for interactions with molecules necessary for NF-κB activation in the immune system. The hyper-IgM phenotype occurred with mutations affecting the ZF domain. The region immediately preceeding the leucine zipper is required for signaling by CD40 and TNF-α, but not IL-1/TLR or TCR. Certain mutations, such as ZF truncations and Δexon4-6 splice mutations globally affect function (88). Many of the human NEMO mutations disrupt specific NEMO functions including impaired binding to IKK-2 or ubiquitinated proteins, impaired ubiquitination, and loss of oligomerization (85).
The same residue of NEMO altered due to the panr2 mutation is mutated to an arginine in a human patient with HED-ID (90) further confirming the importance of this leucine residue in NEMO function. Although L153 does not occur in the IKK-1/2 binding region (4), mutating this residue may affect the ability of the CC1 domain to interact with proteins. Disruption of the coiled coil structure is likely with the panr2 mutation as it results in the substitution of a proline, a residue known to disrupt secondary structures (103). The human L153R mutation impaired TLR induced NF-κB activation and TNF-α production (88;104), TNF-α induced NF-κB activation, and also increased TNF-α-induced programmed cell death with decreased A20 expression (88). The susceptibility to apoptosis caused by the L153R mutation likely contributed to the recurrent inflammatory colitis seen in this patient (104). About 20% of patients with NEMO immunodeficiency exhibit colitis (88). NEMO is ubiquitinated at K278 in response to the nucleotide-binding oligomerization domain protein 2 (NOD2; also known as CARD15)/RICK signaling pathway affected in Crohn’s disease (OMIM #266600) (48). These data complement the results from deleting IKK complex components in the gut epithelium of mice (82). Unlike panr2, L153R mutant cells do not degrade IκBα in response to TLR stimulation (90), potentially accounting for a disruption of EDAR signaling and the development of ectodermal dysplasia in this patient. Mice expressing a constitutively active from of IκBα also develop ectodermal dysplasia (105). The lack of skin and hair defects in panr2 mutant mice may also be consistent with the largely intact nuclear translocation of p65 in an IKBKG mutant patient without ectodermal dysplasia (106).
| |||||||||||||||||||||||||||||||||||||||||||||
Primers | Primers cannot be located by automatic search. | |||||||||||||||||||||||||||||||||||||||||||||
Genotyping | Panr2 genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide change.
Primers for PCR amplification
Panr2(F): 5’- TGATGAGCCTTCTTGATGGTTCAGC -3’
Panr2(R): 5’- CCCCTTGGCAAAAGAGAGTGAGTG -3’
PCR program (use SIGMA JumpStart REDTaq)
1) 94°C 2:00
2) 94°C 0:30
3) 56°C 0:30
4) 72°C 1:00
5) repeat steps (2-4) 29X
6) 72°C 7:00
7) 4°C ∞
Primers for sequencing
Panr2_seq(F): 5’- CTACCACTTGAGAGTCAAGGTTG -3’
Panr2_seq(R): 5’- TTCAGATGCTCAAGCTAGGC -3’
The following sequence of 1424 nucleotides (from Genbank genomic region NC_000086 for linear DNA sequence of Ikbkg) is amplified:
7079 tg
7081 atgagccttc ttgatggttc agctctcagt atttaaaaaa ttccagtctc ctctcttatt 7141 ctgaaacttc tcacctgttg tataatgtgg actaggtgtt aatttttgtt gctgtctttg 7201 agtctagcct ccttttacag agtctcaaaa gagactgaga aaggttcact accacttgag 7261 agtcaaggtt ggtgactgga ggaaggcaaa tagccattgg gctaatgagg cttcaaagta 7321 ccaaggttta taggtcttga gaggcaggag aagccctcca tatatcagct gcttgaaggc 7381 tgccttaggt cttagaagtg ccaaaagtgg gaagcttggg agaaagcagt actgataggg 7441 aatgaccttt tttcttgtag cagatggctg aggacaaggc ctctgtgaaa gctcaggtga 7501 catcattgct cggagaactc caggagagcc agagccgttt ggaggctgcc accaaggatc 7561 ggcaagcttt agagggaagg tgagtgggga ggggaagcta acccaggagg ccccttccag 7621 ggtttttaat gagacaaatg taggagttct aggaattcct ctctctgaaa ggaaggaccg 7681 agtcttttta gattccccag agccctcagg gcttccttgt gagaagcctg gtctggtatt 7741 cagggattta gttctagctg ttgctggaaa gactctgagt ctgtactggg tgcttagtta 7801 gggcttccgt tgctgtggaa agatagcacg acaatgacaa cttttataag gaaaacattc 7861 tatctggctt acattcagag gtttagtcca ttattgtcat ggtagcaagc atgatggtaa 7921 gcagacagac atgatgctgg acaggtgtct gaaagttctt catcgggatt ggcaggcagc 7981 aggaagagag agtaaaccag tgagcctagc ttgagcatct gaaacctcaa agtccatccc 8041 cagtgatccg cttcttttaa caaggccaca cctactctta acaaggccac ttatcctaat 8101 ccttttaaat aattccactc tctagaatcc tacgggggcc attttcactc aaaacatcat 8161 actgggatct agcttattct ttaggaagca ctgcagggcc cctgagacca cactcaagag 8221 ctcactatcc agactgatat gcagtgggat tctaccaagg cagaaagtcc taggtgaaca 8281 agactggata cattccacag caaccctgtg aatagcacag agccctcatg aactacttct 8341 cattaaagac tatggattcc atgtttctgt tgggagttgt tcacgttggt atcctctgcc 8401 aggcatgtac cagtatttct gaattctggc agaaaagtga gtttgactga ctttgtcaac 8461 taatggtctt gggactgtca ctcactctct tttgccaagg gg PCR primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the mutated T is shown in red text.
| |||||||||||||||||||||||||||||||||||||||||||||
References | 58. Hayden, M. S., and Ghosh, S. (2008) Shared Principles in NF-kappaB Signaling. Cell. 132, 344-362. 66. Kawai, T. and Akira, S. (2007) TLR signaling, Semin. Immunol. 19, 24-32. 106. Orange, J. S., Jain, A., Ballas, Z. K., Schneider, L. C., Geha, R. S., and Bonilla, F. A. (2004) The presentation and natural history of immunodeficiency caused by nuclear factor kappaB essential modulator mutation, J Allergy Clin. Immunol. 113, 725-733. | |||||||||||||||||||||||||||||||||||||||||||||
Science Writers | Nora G. Smart | |||||||||||||||||||||||||||||||||||||||||||||
Illustrators | Diantha La Vine | |||||||||||||||||||||||||||||||||||||||||||||
Authors | Owen M. Siggs, Bruce Beutler |