Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Nov 18.
Published in final edited form as: J Neurochem. 2016 Mar 15;137(2):154–163. doi: 10.1111/jnc.13526

NFκB-inducing kinase inhibits NFκB activity specifically in neurons of the CNS

Xianrong Mao 1, Bounleut Phanavanh 2, Hamdan Hamdan 3, Andréa Moerman-Herzog 2, Steven W Barger 2,4,5
PMCID: PMC5115916  NIHMSID: NIHMS827508  PMID: 26778773

Abstract

The control of NFκB in CNS neurons appears to differ from that in other cell types. Studies have reported induction of NFκB in neuronal cultures and immunostaining in vivo, but others have consistently detected little or no transcriptional activation by NFκB in brain neurons. To test if neurons lack some component of the signal transduction system for NFκB activation, we transfected cortical neurons with several members of this signaling system along with a luciferase-based NFκB-reporter plasmid; RelA was cotransfected in some conditions. No component of the NFκB pathway was permissive for endogenous NFκB activity, and none stimulated the activity of exogenous RelA. Surprisingly, however, the latter was inhibited by cotransfection of NFκB-inducing kinase (NIK). Fluorescence imaging of RelA indicated that coexpression of NIK sequestered RelA in the cytoplasm, similar to the effect of IκBα. NIK-knockout mice showed elevated expression of an NFκB-reporter construct in neurons in vivo. Cortical neurons cultured from NIK-knockout mice showed elevated expression of an NFκB-reporter transgene. Consistent with data from other cell types, a C-terminal fragment of NIK suppressed RelA activity in astrocytes as well as neurons. Therefore, the inhibitory ability of the NIK C-terminus was unbiased with regard to cell type. However, inhibition of NFκB by full-length NIK is a novel outcome that appears to be specific to CNS neurons. This has implications for unique aspects of transcription in the CNS, perhaps relevant to aspects of development, neuroplasticity, and neuroinflammation.

Keywords: β-galactosidase, luciferase, MAP3K14, NFκB, RelA, transcription

Graphical Abstract

Full-length NIK was found to inhibit (down arrow) transcriptional activation of NFκB in neurons, while it elevated (up arrow) activity in astrocytes. Deletion constructs corresponding to the N-terminus or C-terminus also inhibited NFκB in neurons, while only the C-terminus did so in astrocytes. One possible explanation is that the inhibition in neurons occurs via two different mechanisms, including the potential for a neuron-specific protein (e.g., one of the 14-3-3 class) to create a novel complex in neurons, whereas the C-terminus may interact directly with NFκB. [Structure of NIK is based on Liu et al. (Liu et al., 2012); N-terminal lobe is oriented at top.]

graphic file with name nihms827508u1.jpg

Introduction

The unique physiology and function of postmitotic neurons in the CNS create dramatic differences in their cellular physiology, including their regulation of gene expression. In many cell types, gene transcription can be activated by “NFκB,” a term used to refer to a group of Rel-family proteins that bind as homo- and heterodimers to a consensus sequence in gene enhancers. Many studies have attempted to explore the relationships between NFκB, development, survival, and synaptic plasticity in neurons. However, many of these studies have been limited by the inherent difficulties in performing a definitive assay of transcriptional activation that is specific to a given cell type in a mixed cell culture or a complex tissue. We have reported that NFκB is refractory in most types of CNS neurons, as assessed by quantitative activity assays in highly pure cultures of neurons or in neuron-glia cocultures where the activity reporter is specific to neurons (Mao et al., 2009; Massa et al., 2006). We have endeavored to uncover elements of the NFκB activation pathway that might explain this cell-type distinction.

Among other subdivisions of the NFκB family, pathways to activation have been divided into the canonical and the noncanonical. The canonical pathway involves the prevalent RelA/p50/IκBα complex. The binding of cell-surface proteins such as the tumor necrosis factor (TNF) receptor activates an IκB kinase complex comprising IKK1, IKK2, and IKK3 (the last is a scaffolding protein), resulting in the phosphorylation, ubiquitination, and proteasomal degradation of IκBα, IκBβ, and IκBε to liberate the active dimers RelA/p50, RelA/RelA, RelA/c-Rel, or c-Rel/c-Rel (Chen et al., 2005; Manavalan et al., 2010). The noncanonical NFκB pathway involves NFκB-inducing kinase (NIK, a.k.a. MAP3K14), IKK1, RelB, and p100 (the precursor of p52) (Sun, 2012). Activated NIK preferentially phosphorylates IKK1 rather than IKK2, and activated IKK1 subsequently phosphorylates p100, which is then partially degraded through a ubiquitin-dependent mechanism. The remaining portion is p52, and its dimerization with RelB constitutes active, noncanonical NFκB.

Apart from being actively involved in the noncanonical activation pathway, NIK function is not well understood. Although NIK mRNA is abundant, the protein has a short half-life, resulting in low steady-state levels in most cell types (Qing et al., 2005). Because IKK1 is actively involved in limiting IKK2-induced NFκB activity and IKK1 interacts functionally with NIK (Xiao et al., 2004), it is also possible that NIK plays a role in dampening excessive NFκB activity. Two major NIK domains have been identified: one is the kinase active site and the other is responsible for interaction with other molecules. Interestingly, two different domains of NIK have been found to inhibit NFκB activity (Lin et al., 1998; Xiao and Sun, 2000). One of these, located in the C-terminus, can strongly inhibit NFκB activity induced by TNF; the other, in the N-terminus, can suppress NFκB activity but not as strongly as the C-terminal sequence. NIK can shuttle between cytosol and nucleus, as it contains both a nuclear localization signal (NLS) and a nuclear export signal (NES); nuclear localization is typically found to be transient (Birbach et al., 2002); (Choudhary et al., 2005). Nuclear-localized NIK has been found to strongly inhibit NFκB activity induced by hepatitis B virus in hepatocytes (Park et al., 2005), consistent with the notion that the NIK→IKK1 pathway can antagonize the IKK2→NFκB pathway (Anest et al., 2003; Lawrence et al., 2005).

Because NFκB in CNS neurons is refractory to many conventional stimuli, we hypothesized that these cells are deficient in some component of the relevant signal-transduction pathways. Therefore, our initial aim was to effect repletion by expressing these components alone and in combinations. We serendipitously found that NIK expression repressed an empirical NFκB activation in neurons. The opposite effect was noted in astrocytes, suggesting neuron-specific manipulation of NIK and/or other components of the NFκB system. Such a phenomenon may begin to provide a mechanistic explanation for the distinctions CNS neurons exhibit with regard to NFκB activation.

Materials and Methods

Animals

Sprague-Dawley rats (Harlan Sprague Dawley) were used as the source of most primary neuron cultures from cerebral cortex. Two genetically modified mouse lines were also utilized: The κB/β-gal reporter line developed by Bhakar et al. (Bhakar et al., 2002) was generously provided by Dr. Philip A. Barker (McGill University, Montreal, Canada). This line is transgenic for a β-galactosidase gene under control of an NFκB-responsive promoter. The Map3k14−/− (NIK-knockout) line was generated by Yin et al. (Yin et al., 2001) and was generously provided by Amgen (Thousand Oaks, CA). Both lines were backcrossed into the C57BL/6 line for at least six generations. For some experiments these two lines were crossed to generate F2 animals that carried the κB/β-gal transgene in littermates that were either Map3k14−/− or Map3k14+/+. For tissue harvest, mice were anesthetized with isoflurane and decapitated for the removal of brains. Typically, one half of the brain was immersion-fixed in formalin and the opposite half was homogenized for β-galactosidase measurements. All analyses were performed on male mice; experiments were reviewed and approved by the Animal Care and Use Committee of the Central Arkansas Veterans Healthcare System.

Cell cultures

Primary neuronal cultures were established from the neocortices of 18-day rat embryos or 17-day mouse embryos as described previously (Mao et al., 1999); sexes were not distinguished. Neurons were maintained in Neurobasal medium containing B27 supplement (Invitrogen), 0.5 mM L-glutamine, and 10 μg/ml gentamycin sulfate. For cultures essentially devoid of glia, the mitotic inhibitor cytosine arabinoside (AraC, 3–10 μM) was added one day after plating and continued until a medium change on Day 5 (Moerman and Barger, 1999). All neuronal cultures used in this study were 8–10 days in vitro. Astrocyte cultures were established from neonatal animals and maintained in minimal essential medium with Earle’s salts (MEM, Invitrogen) supplemented to 10% with fetal bovine serum (FBS, Invitrogen) (Moerman and Barger, 1999). Microglia were removed by vigorous lavage, and the remaining cells had the appearance of type-I astrocytes.

We developed a system for establishing embryonic cell cultures from mice genetically ablated of the NIK gene (Yin et al., 2001). Although literature reports indicate that NIK knockout mice are fertile, we were unsuccessful at breeding the nullizygous animals. Therefore, we bred NIK+/− mice and developed a rapid-genotyping strategy by which NIK−/− and NIK+/+ embryos (or neonates) could identified during dissection so that cells from congenic progeny could be pooled into cultures. Real-time PCR was carried out on DNA prepared from a tail snip of each embryo to identify the presence of the wild-type NIK allele (primers: 5′-TCT GAG ATA GGC ATA TCC CTG GCT; 5′-AGT CCA ATT CCA TGT TGC TGC TGT) and the targeted disruption, including a portion of the neomycin-resistance gene (primer: ATC TTG TTC AAT GGC CGA TCC CAT; paired with the second NIK primer); we also included detection of Apoe as a quantitative standard by which to compare diploid copy numbers (primers: 5′-GGC CCA GGA GAA TCA ATG AGT; 5′-CCT GGC TGG ATA TGG ATG TTG). The entire DNA extraction and PCR protocol could be completed in 3 h; accounting for dissection time, this allowed the cells to be plated within 3.5 h postmortem.

Plasmids

pGL3-promΔ was created by removal of Sp1 sites from pGL3-prom to make it more specific for introduced cis elements (Mao et al., 2006). Four copies of a κB response element (ACGGGGTTTCACCGT) were then inserted to create pGL3-promΔ-4κB. The RelA and IκBα expression vectors were constructed in pEFGP-N1 and pDsRed2-N1, respectively, as described (Mao et al., 2006). Expression constructs for NIK, IKK1, and IKK2 (Geleziunas et al., 1998), as well as deletion constructs of NIK (1-220, 1-366, 1-570, 1-624, 350-947, 570-947, 735-947) (Lin et al., 1998) and a kinase-defective mutant of NIK (NIKK429A/K430A) (Foehr et al., 2000), were generously provided by Warner C. Greene (Univ. California-San Francisco). The pME-TRAF2 and pME-TRAF6 expression vectors (Akama and Van Eldik, 2000) were generously provided by Linda Van Eldik (Univ. Kentucky). Clones were verified by bidirectional sequencing. The pRL-CMV (Promega) plasmid expressing Renilla luciferase constitutively was used as a control for transfection efficiency and viability. All plasmids were prepared with Qiagen Midiprep kits, and DNA quality and quantity were determined by both spectrophotometry and visual inspection in agarose gels.

siRNA

Double-stranded Silencer® RNA of three different target sites in the rat NIK sequence were obtained from Ambion; sense strand sequences were as follows: siRNA 1: CGC GUG AUC ACC AAA GGC tt; siRNA 2: GCC CGA AAA AAA CGU AGG Att; siRNA 3: CGU AGG AAG AAG AGG UCG Att; a negative control RNA of undisclosed sequence was also provided by Ambion. The siRNA was mixed with the Lipofectamine 2000 along with transfected plasmids and applied to neuronal cultures in the existing medium at one of three different concentrations, indicated in figure legends.

Transient transfection and luciferase activity assay

Cells were cultured in 24-well plates and transfected using Lipofectamine 2000 (Invitrogen) essentially as described previously (Mao et al., 2006). Astrocytes were transfected when the cultures reached 60–70% confluency, whereas neurons were plated and transfected at a density of 1 × 105/cm2. Total quantity of plasmid per well was standardized at 1 μg; pBluescript II was used to adjust to this total. After preincubation, DNA/Lipofectamine mixtures were added directly to the existing culture medium. Cells were harvested ~24 h after transfection for luciferase assay. Dual-luciferase reporter assay system (Promega) was used to detect both firefly and Renilla luciferase activities through a Veritas luminometer (Promega).

β-galactosidase assays

The measurement of β-galactosidase activity in lysates of mouse brain was performed with the Galacto-Light Plus kit (Life Technologies) according to manufacturer’s instructions. Half-brains were Dounce-homogenized in 100 mM sodium phosphate buffer (pH 7.8) containing 0.2% Triton X100, 0.2 mM phenylmethylsulfonyl fluoride, 5 μg/ml leupeptin, 0.5 mM dithiothreitol, and 0.03% H2O2. Homogenates were heated to 55 °C for 10 min, cleared by centrifugation, and an aliquot was removed for protein assay (BCA, Pierce). β-gal assays were performed on 20 μl of resulting extract. Homogenization buffer itself was used as a blank, values from which were subtracted from the tissue-sample values.

For histochemistry of β-galactosidase in situ, half-brains were immersion-fixed in formalin for 48 h and then cryoprotected by transfer to 25% sucrose. Tissue was embedded in Tissue-Tek CRYO-OCT Compound (Fisher Scientific), frozen, and cut in 30-μm sections on a cryostat. Tissue sections were incubated 30 min in development solution [phosphate-buffered saline containing 5 mM potassium ferricyanide, 2.5 mM, potassium ferrocyanide, 2.4 mM 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal), 2 mM MgCl2, 0.02% NP40, and 0.01% sodium deoxycholate], rinsed twice in PBS containing 3% DMSO, and twice in PBS alone; they were then post-fixed for 15 min in 4% paraformaldehyde, rinsed in PBS, and mounted on glass slides for microscopy.

Statistics

Pairwise comparisons were made by Student’s t-test. Other datasets were analyzed by one-way ANOVA followed by Bonferroni post hoc test. A p-value of 0.05 or less was considered statistically significant.

Results

Neuron-specific suppression of NFκB by NIK

Despite the ability of some conditions to produce a nuclear translocation of Rel-family proteins in CNS neurons, numerous approaches have demonstrated little or no induction of NFκB-mediated transcriptional activation in these cells (Barger et al., 2005; Mao et al., 1999; Mao et al., 2009; Massa et al., 2006; Moerman et al., 1999). To address the possibility that Rel transcription factors and/or components of their activation pathway are expressed at low levels in CNS neurons, we transfected several members of the NFκB signal-transduction pathway into primary cortical neurons along with an NFκB-responsive luciferase reporter-gene plasmid. While an intermediate amount of RelA produced a substantial increase in reporter-gene expression (Fig. 1), neither this induction nor the basal level of expression was affected by cotransfection with p300, TNF receptor-associated factor (TRAF) 2, or TrCP1 (Mao et al., 2009). In one set of experiments, we also tested transfection of the proximal IκB kinases IKK1, -2, or -3. Neither of these kinases nor a combination of all three was able to enhance the activity of RelA (Fig. 1). However, when NIK was included among the IKKs, a significant reduction in the activity of RelA was observed. By contrast, transfection of NIK into astrocytes activated endogenous NFκB and superactivated cotransfected RelA (Fig. 2). This activation property of NIK was dependent on its kinase activity, as an inactive mutant (NIKK429A/K430A) was unable to induce NFκB in astrocytes.

Figure 1. NIK suppresses NFκB activity in neurons.

Figure 1

Open in a new tab

Primary neocortical neurons were transfected with a κB-driven firefly luciferase reporter with or without RelA; a combination of RelA with IKK1, -2, and -3; or all of these combined with NIK. All plasmids were 100 ng/well except RelA, which was 10 ng/well. [The dose of RelA plasmid was selected to be intermediate with respect to the maximal level of the assay’s dynamic range. Relative luciferase activity from RelA (100 ng) transfection was 111.0.] After 24 h, firefly luciferase activities were determined relative to Renilla luciferase reference. Values reflect the mean ± SEM of triplicate cultures. (*p<0.05 vs. RelA/IKK1,2,3)

Figure 2. NIK activates NFκB in astrocytes.

Figure 2

Open in a new tab

Primary cultures of rat astrocytes were transfected with a κB-driven firefly luciferase reporter (and constitutive Renilla luciferase for reference). The reporter was induced by cotransfection of RelA in all conditions designated by a black bar; white bar represents basal reporter activity. Schematic representations of the NIK protein and several deletion constructs are depicted alongside their effects on the κB enhancer alone (burgandy bars) or on the activation by RelA (black bars). Triangular ramps indicate increasing doses of plasmid: 0.03, 0.1, and 0.3 μg. IκBα was used as a positive control for RelA inhibition. Values reflect the mean ± SEM of triplicate cultures 24 h after transfection. Note that only the mutants lacking an active kinase domain are inhibitory in astrocytes. (*p<0.01; ANOVA and Scheffe post-hoc)

Individual domains of NIK have been proposed to act as dominant-negative mutants, binding up associated members of the signaling cascade without propagating a signal (Lin et al., 1998; Xiao and Sun, 2000). We conducted a structural analysis with a variety of N- and C-terminal-truncated NIK mutants to explore in greater detail the sequence requirements for the function of NIK in neurons and astrocytes. A C-terminal fragment comprising the last 212 amino acids of NIK suppressed RelA activity in astrocytes (Fig. 2), consistent with reports testing this C-terminal domain in other systems (Lin et al., 1998; Xiao and Sun, 2000). N-terminal NIK fragments were essentially neutral in astrocytes, failing to elevate RelA-generated activity in the way full-length NIK did. Even constructs containing the active kinase site were moot, presumably because the deletion of the C-terminus abolishes interactions with IKK1 (Lin et al., 1998).

When these constructs were tested in neurons, suppression of RelA activity was observed with all constructs (Fig. 3). Each N- and C-terminal fragment was as effective as full-length NIK in this regard. Indeed, the C-terminal domain evoked a more severe inhibition than full-length NIK itself. Thus, the inhibitory ability of the NIK C-terminus was unbiased with regard to cell type; the functional distinction between the actions of NIK in neurons versus glia depended on the intact NIK molecule.

Figure 3. Structural determinants of NIK suppression of NFκB activity in neurons.

Figure 3

Open in a new tab

Highly purified cultures of rat cortical neurons were transfected with a κB-driven firefly luciferase reporter (and constitutive Renilla luciferase for reference). The reporter was induced by cotransfection of RelA in all conditions designated by a black bar; white bar represents basal reporter activity. Schematic representations of the NIK protein and several deletion constructs are depicted alongside their effects on reporter activity (in the presence of RelA). IκBα was used as a positive control for RelA inhibition. Triangular ramps indicate increasing doses of plasmid: 0.1, 0.3, and 1.0 μg. Values reflect the mean ± SEM of triplicate cultures 24 h after transfection. *RelA alone was significantly greater than all other conditions. Note that all portions of NIK inhibit RelA activity in neurons (*p<0.01; ANOVA and Scheffe post-hoc).

Physical interactions of NIK with RelA

The influence of NIK on RelA could be mediated by several functional mechanisms. We previously analyzed the cellular localization of RelA in neurons by using a RelA-EGFP fusion construct, permitting visualization of transiently transfected cells. RelA-EGFP expressed alone localized to the nucleus, consistent with its NFκB transactivation capacity in reporter assays (Barger et al., 2005). When IκBα was cotransfected with RelA-EGFP, the surfeit of RelA was apparently bound by IκBα and retained in the cytosol. Similar results were obtained when NIK transfection replaced that of IκBα (Fig. 4A). To test the possibility that this resulted from a physical sequestration of RelA by NIK, we conducted coimmunoprecipitation assays. NTera2 cells were cotransfected with RelA and NIK. NIK was apparent in crude lysates of the transfectants and in the RelA-precipitated pellets, but it was absent from control pellets (Fig. 4B).

Figure 4. Physical interaction between RelA and NIK.

Figure 4

Open in a new tab

A. A RelA-EGFP fusion protein was transfected into primary cortical neurons alone or with an expression construct for NIK or IκBα. Note the strong nuclear localization of RelA-EGFP and the reversal of this localization by NIK and IκBα. B. NTera2 cells were cotransfected with RelA and either empty expression vector (mock) or the NIK vector. Western blotting with a NIK antibody was performed on raw lysates or after immunoprecipitation with normal goat serum (NGS) or a goat RelA antibody (Ab). The NIK protein migrates at ~105 kDa.

Disinhibition of NFκB by suppression of NIK

To test whether endogenous NIK contributes to the refractory nature of NFκB in CNS neurons, small interfering RNA (siRNA) was used to knock down NIK expression in cultures of cortical neurons. Three siRNA oligonucleotides were designed by the manufacturer (Ambion) in accordance with an operationally defined algorithm. Each NIK siRNA and control oligonucleotides (scrambled-sequence double-stranded RNA, Srr siRNA, and Gapdh siRNA; also from Ambion) were cotransfected with a κB-driven luciferase reporter plasmid. The doses of the oligonucleotides were optimized empirically. As shown in Fig. 5, NIK siRNAs #1 and #2 were able to significantly increase reporter activity at various concentrations. Another sequences, designated NIK siRNA #3, was ineffective at modulating NFκB reporter activity. The differential effects on activity corresponded to differential effects on NIK protein levels, which were significantly depleted by NIK siRNAs #1 and #2 but not #3 (not shown). NIK siRNA was also cotransfected with RelA in neurons to determine if RelA activity could be further enhanced. NIK siRNA #1 and NIK siRNA #2 significantly boosted this additional RelA activity (not shown). In all these experiments, control oligonucleotides were ineffective. These data suggest that endogenous NIK actively suppresses both basal and induced NFκB activity in neurons.

Figure 5. NIK knockdown disinhibits NFκB activity in neurons.

Figure 5

Open in a new tab

Primary neocortical neurons were cotransfected with an NFκB-responsive reporter and one of five different dsRNA oligonucleotides (25 nM): scrambled-sequence dsRNA, siRNA directed against the sequences of Srr or Gapdh, or one of two siRNAs directed against NIK. One set of cultures was transfected with RelA as a positive control. After 72 h, neurons were harvested and firefly luciferase activities were determined relative to Renilla luciferase reference. Values reflect the mean ± SEM of triplicate cultures. (*p<0.05; ANOVA and Scheffe post-hoc)

Disinhibition of NFκB by genetic ablation of NIK

We developed a system for establishing embryonic cell cultures from mice genetically ablated of the NIK gene (Yin et al., 2001). We began by establishing mixed glial cultures from neonatal NIK−/− mice and NIK+/+ littermates; astrocytes obtained from these cultures were then transfected with an NFκB-driven reporter plasmid. NIK−/− astrocytes showed lower basal NFκB activity and lower responses to CD40L (a noncanonical NFκB agonist) relative to NIK+/+ cells; however, responses to TNF (a canonical NFκB agonist) were essentially intact (Fig. 6A), consistent with NIK’s predominant role in the noncanonical pathway (Sun, 2012). Cotransfection of a NIK expression construct restored responses to CD40L in NIK−/− astrocytes. Neurons from NIK−/− embryos were similarly transfected with an NFκB reporter plasmid. Basal activity did not appear altered by NIK ablation in neurons; however, the absence of NIK permitted induction by overexpression of TRAF2 and TRAF6 (Fig. 6B), which were ineffective in wild-type neurons.

Figure 6. Absence of NIK is permissive for NFκB activity in neurons.

Figure 6

Open in a new tab

Primary neocortical cultures of astrocytes (A) or neurons (B) were established from wild-type or NIK−/− mice. A κB-driven firefly luciferase reporter was transfected alone or together with expression vectors for NIK (astrocytes) or TRAF2 and -6. Some astrocyte cultures were treated with rat TNFα (10 ng/ml) or murine CD40L (1 μg/ml). After 24 h, cultures were harvested and firefly luciferase activities were determined relative to Renilla luciferase reference. Values reflect the mean ± SEM of triplicate cultures. *p<002, **p<0.01.

The tonic inhibition of neuronal NFκB by NIK was also examined in mouse brains. To assess NFκB activity, we made use of a mouse line that carries a β-gal gene driven by a promoter sensitive to NFκB (Bhakar et al., 2002). We crossed this line to the NIK-knockout line to produce littermates that carried the βgal reporter in the context of a wild-type (NIK+/+;κB/βgalTg/0) or nullizygous (NIK−/−;κB/βgalTg/0) NIK genotype. At 3 months of age, mice were euthanized, and the brains were bisected midsagittally. One half was processed for β-gal histology and the other for a quantitative assay of β-gal activity. A marked increase in β-gal development was evident, primarily in the hippocampus, in the NIK−/− mice (Fig. 7B) compared to NIK+/+ (Fig. 7A). Enzymatic activity assays also showed an increase, though the effect was diluted to some extent by inclusion of the entire brain (Fig. 7C).

Figure 7. NIK ablation disinhibits NFκB in vivo.

Figure 7

Open in a new tab

Mice carrying a β-gal gene under control of a κB enhancer were crossed with NIK−/− mice. At 3 mo of age, brains were sectioned and processed for X-gal histochemistry (A, B) or brains were homogenized for enzymatic assay of β-gal activity (C) (*p=0.026).

Discussion

Post-mitotic CNS neurons have been reported to show NFκB responses that are exceedingly attenuated. NFκB activity can be observed readily in brain homogenates or mixed cultures, but this may be attributable almost exclusively to glia and other nonneuronal cell types. Although evidence to the contrary has been reported, especially in the early history of such studies (Grilli et al., 1996; Kaltschmidt et al., 1995; Meffert et al., 2003), more recent findings indicate that neurons exhibit little or no activation of NFκB in response to tumor necrosis factor, interleukin-1β, lipopolysaccharide, HIV gp120, nerve growth factor, brain-derived neurotrophic factor, glutamate, cholinergic or adrenergic agonists, cytoskeletal drugs, or oxidants such as hydrogen peroxide (Barger and Mao, 2012; Jarosinski et al., 2001; Lian et al., 2012; Lian et al., 2015; Listwak et al., 2013; Mao et al., 2009; Massa et al., 2006; Saha and Pahan, 2007; Srinivasan et al., 2004).

A recalcitrant NFκB system may be one aspect of the special protection from immune attack accorded to neurons. Though the CNS is not devoid of immune system activity, the inability to replace neurons in considerable numbers means that immunological cytotoxicity (e.g., removal of virus-infected cells) could have dire consequences for the organism. Such immune attack is often targeted toward host cells expressing class I major histocompatibility genes. As these genes are robustly induced by NFκB, it may be adaptive for neurons to suppress this transcription factor generally. Nevertheless, the mechanisms responsible for the refractory nature of NFκB in neurons have been elusive. Because overexpression of a DNA-binding protein comprising NFκB such as RelA can produce transcriptional activation in neurons, we assumed that meager activity levels resulted from a passive deficiency, perhaps in some member of the upstream signal-transduction such as IKKs. In attempting to reverse such potential deficiencies, we were surprised to find that expression of NIK uniquely suppressed the activity afforded through overexpression of RelA. In fact, NIK behaved analogous to IκBα in assays of transcriptional activation and manipulation of RelA subcellular localization. This stands in contrast to the situation in astrocytes and most other cell types, where NIK typically enhances the activity of RelA and NFκB in general.

NIK has previously been shown to suppress NFκB activation only in mutant forms. Similar to our results in both neurons and astrocytes, a kinase-inactive mutant failed to enhance RelA activity, but dominant-negative inhibition of RelA required a C-terminal fragment (Lin et al., 1998). This is probably because the full-length NIK, even in the absence of a kinase active site, has more functional properties than individual N- or C-terminal constructs. It is possible that the C-terminal region is not accessible to repress NFκB activity in full-length NIK. It will be of interest to determine whether NIK expressed in neurons differs intrinsically from the NIK found in other cell types, e.g., by conformation or post-translational modification. Though we detected physical interactions between RelA and what appeared to be full-length NIK, it is possible that a fraction of the NIK in neurons is fragmented by proteases or otherwise altered at levels sufficient to bind and inactivate NFκB.

An alternative to an intrinsic difference in neuronally expressed NIK is the possibility that other proteins uniquely expressed in CNS neurons interact with NIK to convert its influence from activation to repression. Of note is the NIK- and IKK-binding protein (NIBP) which was first identified as a NIK-binding protein via a yeast two-hybrid approach in brain homogenates and subsequently localized prominently to neurons (Hu et al., 2005). However, NIBP has also been localized to other cell types, and—in those cells, at least—it appears to enhance rather than repress NFκB activity. Stronger candidates may be represented by A20 and/or its binding partners: A20-binding inhibitor of NFκB (ABIN) (Heyninck et al., 1999), ABIN-2 (Van Huffel et al., 2001), and 14-3-3 proteins (Matitau and Scheid, 2008). A20 is an important inhibitor of NFκB that has been shown to physically interact with IKK3, and while it does not inhibit NFκB activation via the noncanonical pathway (Heyninck et al., 1999), that fact does not rule out a physical interaction with NIK. CNS neurons constitutively express substantial amounts of A20 and lose this expression under stressful and degenerative conditions (Peluffo et al., 2013).

It is intriguing that NIK expression is lost in a very large fraction of neuroblastomas (Vandesompele et al., 2001). NFκB is associated with proliferative gene targets in many cases (Ledoux and Perkins, 2014); it is possible that NIK plays some role in suppressing these genes in neuroblasts as a contribution to their differentiation into post-mitotic neurons. Chromosomal instabilities that result in deletions in the NIK genomic region (q21.31 region of Chromosome 17) may act to inhibit such differentiation during the development of neuroblastoma. Indeed, using four different types of prediction, the computational oncogene prioritization software CGPrio (Furney et al., 2008) scores NIK at a ≥95% chance of acting as a tumor suppressor in the context of neuroblastoma.

Level of expression is an important parameter in transgene studies, including many of those performed here. Obviously, overexpression of RelA is sufficient to activate transcription through cis elements containing the κB enhancer sequence. This appears to result from overwhelming the levels of endogenous IκBs, as we have shown inactivation of transfected RelA in neurons by cotransfection with a robust expression vector for IκBα. Some studies in which RelA or other members for the NFκB transcription-factor family were overexpressed in neurons have shown intriguing effects on synaptic plasticity or other aspects of neurophysiology. However, it is not clear that these interpretations are physiologically relevant without equivalent coexpression of IκB. Caveats regarding expression levels were obviated here with experiments in which NIK was ablated genetically without transfection of other components. Similarly, advances have been made with a floxed RelA gene, where excision after pharmacologically evoked Cre recombinase expression created changes in dendritic spines in cultured neurons. This approach was recently employed in a study implicating a Polo-like kinase (Plk) in control of neuronal NFκB (Mihalas et al., 2013). It will be of interest to determine if NIK and Plks interact.

In conclusion, we have found that the expression of NIK in CNS neurons inhibits rather than activates NFκB activity. Indeed, this repressive activity of NIK explains, at least partially, a generalized recalcitrance of NFκB in CNS neurons reported previously by this laboratory and others. It will be important to determine whether this is an intrinsic property of NIK, as expressed and perhaps modified in neurons, or alternatively that other proteins specific to neurons interact with NIK in a way that establishes this somewhat paradoxical repression.

Acknowledgments

The κB/β-gal mouse was generously provided by Dr. Philip A. Barker (McGill University, Montreal, Canada). The NIK-knockout mouse was obtained by agreement with Amgen (Thousand Oaks, CA). We are also grateful to several individuals who provided expression plasmids, as described the Materials and Methods section. We appreciate technical assistance, including animal husbandry, from Mandy Porter and Deiadra Brown. This work was supported by funds from the NIH (R01NS046439 and P01AG12411). The authors have no conflicts of interest to declare.

Abbreviations

AraC

cytosine arabinoside

FBS

fetal bovine serum

β-gal

β-galactosidase

IKK

inhibitor of κ light polypeptide gene enhancer in B-cells

MAP3K14

mitogen-activated protein kinase kinase kinase 14

MEM

minimal essential medium (Earle’s salts)

NFκB

nuclear factor of κ light chain enhancer of activated B-cells

NIK

NFκB-inducing kinase

PBS

phosphate-buffered saline

TNF

tumor necrosis factor

TRAF

TNF receptor-associated factor

X-gal

5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside

References

  1. Akama KT, Van Eldik LJ. B-amyloid stimulation of inducible nitric-oxide synthase in astrocytes is interleukin-1β- and tumor necrosis factor-α (TNFα)-dependent, and involves a TNFα receptor-associated factor- and NFκB-inducing kinase-dependent signaling mechanism. J Biol Chem. 2000;275:7918–7924. doi: 10.1074/jbc.275.11.7918. [DOI] [PubMed] [Google Scholar]
  2. Anest V, Hanson JL, Cogswell PC, Steinbrecher KA, Strahl BD, Baldwin AS. A nucleosomal function for IκB kinase-α in NF-κB-dependent gene expression. Nature. 2003;423:659–663. doi: 10.1038/nature01648. [DOI] [PubMed] [Google Scholar]
  3. Barger SW, Mao X. NF-κB in Neurons—Mechanisms and Myths. In: Albensi BC, editor. Transcription Factors CREB and NF-κB: Involvement in Synaptic Plasticity and Memory Formation. Bentham Science Publishers; Oak Park IL: 2012. pp. 113–129. [Google Scholar]
  4. Barger SW, Moerman AM, Mao X. Molecular mechanisms of cytokine-induced neuroprotection: NFκB and neuroplasticity. Curr Pharm Des. 2005;11:985–998. doi: 10.2174/1381612053381594. [DOI] [PubMed] [Google Scholar]
  5. Bhakar AL, Tannis LL, Zeindler C, Russo MP, Jobin C, Park DS, MacPherson S, Barker PA. Constitutive nuclear factor-κ B activity is required for central neuron survival. J Neurosci. 2002;22:8466–8475. doi: 10.1523/JNEUROSCI.22-19-08466.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Birbach A, Gold P, Binder BR, Hofer E, de Martin R, Schmid JA. Signaling molecules of the NF-κ B pathway shuttle constitutively between cytoplasm and nucleus. J Biol Chem. 2002;277:10842–10851. doi: 10.1074/jbc.M112475200. [DOI] [PubMed] [Google Scholar]
  7. Chen LF, Williams SA, Mu Y, Nakano H, Duerr JM, Buckbinder L, Greene WC. NF-κB RelA phosphorylation regulates RelA acetylation. Mol Cell Biol. 2005;25:7966–7975. doi: 10.1128/MCB.25.18.7966-7975.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Choudhary S, Boldogh S, Garofalo R, Jamaluddin M, Brasier AR. Respiratory syncytial virus influences NF-κB-dependent gene expression through a novel pathway involving MAP3K14/NIK expression and nuclear complex formation with NF-κB2. J Virol. 2005;79:8948–8959. doi: 10.1128/JVI.79.14.8948-8959.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Foehr ED, Bohuslav J, Chen LF, DeNoronha C, Geleziunas R, Lin X, O’Mahony A, Greene WC. The NF-κ B-inducing kinase induces PC12 cell differentiation and prevents apoptosis. J Biol Chem. 2000;275:34021–34024. doi: 10.1074/jbc.C000507200. [DOI] [PubMed] [Google Scholar]
  10. Furney SJ, Calvo B, Larranaga P, Lozano JA, Lopez-Bigas N. Prioritization of candidate cancer genes--an aid to oncogenomic studies. Nucleic Acids Res. 2008;36:e115. doi: 10.1093/nar/gkn482. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Geleziunas R, Ferrell S, Lin X, Mu Y, Cunningham ET, Jr, Grant M, Connelly MA, Hambor JE, Marcu KB, Greene WC. Human T-cell leukemia virus type 1 Tax induction of NF-κB involves activation of the IκB kinase α (IKKα) and IKKβ cellular kinases. Mol Cell Biol. 1998;18:5157–5165. doi: 10.1128/mcb.18.9.5157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Grilli M, Goffi F, Memo M, Spano P. Interleukin-1b and glutamate activate the NF-kB/Rel binding site from the regulatory region of the amyloid precursor protein gene in primary neuronal cultures. J Biol Chem. 1996;271:15002–15007. doi: 10.1074/jbc.271.25.15002. [DOI] [PubMed] [Google Scholar]
  13. Heyninck K, De Valck D, Vanden Berghe W, Van Criekinge W, Contreras R, Fiers W, Haegeman G, Beyaert R. The zinc finger protein A20 inhibits TNF-induced NF-κB-dependent gene expression by interfering with an RIP- or TRAF2-mediated transactivation signal and directly binds to a novel NF-κB-inhibiting protein ABIN. J Cell Biol. 1999;145:1471–1482. doi: 10.1083/jcb.145.7.1471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hu WH, Pendergast JS, Mo XM, Brambilla R, Bracchi-Ricard V, Li F, Walters WM, Blits B, He L, Schaal SM, Bethea JR. NIBP, a novel NIK and IKK(β)-binding protein that enhances NF-(κ)B activation. J Biol Chem. 2005;280:29233–29241. doi: 10.1074/jbc.M501670200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Jarosinski KW, Whitney LW, Massa PT. Specific deficiency in nuclear factor-kB activation in neurons of the central nervous system. Lab Invest. 2001;81:1275–1288. doi: 10.1038/labinvest.3780341. [DOI] [PubMed] [Google Scholar]
  16. Kaltschmidt C, Kaltschmidt B, Baeuerle PA. Stimulation of ionotropic glutamate receptors activates transcription factor NF-kB in primary neurons. Proc Natl Acad Sci USA. 1995;92:9618–9622. doi: 10.1073/pnas.92.21.9618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lawrence T, Bebien M, Liu GY, Nizet V, Karin M. IKKα limits macrophage NF-κB activation and contributes to the resolution of inflammation. Nature. 2005;434:1138–1143. doi: 10.1038/nature03491. [DOI] [PubMed] [Google Scholar]
  18. Ledoux AC, Perkins ND. NF-κB and the cell cycle. Biochem Soc Trans. 2014;42:76–81. doi: 10.1042/BST20130156. [DOI] [PubMed] [Google Scholar]
  19. Lian H, Shim DJ, Gaddam SS, Rodriguez-Rivera J, Bitner BR, Pautler RG, Robertson CS, Zheng H. IκBα deficiency in brain leads to elevated basal neuroinflammation and attenuated response following traumatic brain injury: implications for functional recovery. Mol Neurodegener. 2012;7:47. doi: 10.1186/1750-1326-7-47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lian H, Yang L, Cole A, Sun L, Chiang AC, Fowler SW, Shim DJ, Rodriguez-Rivera J, Taglialatela G, Jankowsky JL, Lu HC, Zheng H. NFκB-activated astroglial release of complement C3 compromises neuronal morphology and function associated with Alzheimer’s disease. Neuron. 2015;85:101–115. doi: 10.1016/j.neuron.2014.11.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lin X, Mu Y, Cunningham ET, Jr, Marcu KB, Geleziunas R, Greene WC. Molecular determinants of NF-κB-inducing kinase action. Mol Cell Biol. 1998;18:5899–5907. doi: 10.1128/mcb.18.10.5899. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Listwak SJ, Rathore P, Herkenham M. Minimal NF-κB activity in neurons. Neuroscience. 2013;250:282–299. doi: 10.1016/j.neuroscience.2013.07.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Liu J, Sudom A, Min X, Cao Z, Gao X, Ayres M, Lee F, Cao P, Johnstone S, Plotnikova O, Walker N, Chen G, Wang Z. Structure of the nuclear factor κB-inducing kinase (NIK) kinase domain reveals a constitutively active conformation. J Biol Chem. 2012;287:27326–27334. doi: 10.1074/jbc.M112.366658. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Manavalan B, Basith S, Choi YM, Lee G, Choi S. Structure-function relationship of cytoplasmic and nuclear IκB proteins: an in silico analysis. PLoS One. 2010;5:e15782. doi: 10.1371/journal.pone.0015782. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Mao X, Moerman-Herzog AM, Wang W, Barger SW. Differential transcriptional control of the superoxide dismutase-2 κB element in neurons and astrocytes. J Biol Chem. 2006;281:35863–35872. doi: 10.1074/jbc.M604166200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mao X, Moerman AM, Lucas MM, Barger SW. Inhibition of the activity of a neuronal kB-binding factor (NKBF) by glutamate. J Neurochem. 1999;73:1851–1858. [PubMed] [Google Scholar]
  27. Mao XR, Moerman-Herzog AM, Chen Y, Barger SW. Unique aspects of transcriptional regulation in neurons--nuances in NFκB and Sp1-related factors. J Neuroinflammation. 2009;6:16. doi: 10.1186/1742-2094-6-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Massa PE, Aleyasin H, Park DS, Mao X, Barger SW. NFkB in neurons? The uncertainty principle in neurobiology. J Neurochem. 2006;97:607–618. doi: 10.1111/j.1471-4159.2006.03810.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Matitau AE, Scheid MP. Phosphorylation of MEKK3 at threonine 294 promotes 14-3-3 association to inhibit nuclear factor κB activation. J Biol Chem. 2008;283:13261–13268. doi: 10.1074/jbc.M801474200. [DOI] [PubMed] [Google Scholar]
  30. Meffert MK, Chang JM, Wiltgen BJ, Fanselow MS, Baltimore D. NF-κ B functions in synaptic signaling and behavior. Nat Neurosci. 2003;6:1072–1078. doi: 10.1038/nn1110. [DOI] [PubMed] [Google Scholar]
  31. Mihalas AB, Araki Y, Huganir RL, Meffert MK. Opposing action of nuclear factor κB and Polo-like kinases determines a homeostatic end point for excitatory synaptic adaptation. J Neurosci. 2013;33:16490–16501. doi: 10.1523/JNEUROSCI.2131-13.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Moerman AM, Barger SW. Inhibition of AMPA responses by mutated presenilin 1. J Neurosci Res. 1999;57:962–967. [PubMed] [Google Scholar]
  33. Moerman AM, Mao X, Lucas MM, Barger SW. Characterization of a neuronal kB-binding factor distinct from NF-kB. Mol Brain Res. 1999;67:303–315. doi: 10.1016/s0169-328x(99)00091-1. [DOI] [PubMed] [Google Scholar]
  34. Park SG, Ryu HM, Lim SO, Kim YI, Hwang SB, Jung G. Interferon-γ inhibits hepatitis B virus-induced NF-κB activation through nuclear localization of NF-κB-inducing kinase. Gastroenterology. 2005;128:2042–2053. doi: 10.1053/j.gastro.2005.03.002. [DOI] [PubMed] [Google Scholar]
  35. Peluffo H, Gonzalez P, Acarin L, Aris A, Beyaert R, Villaverde A, Gonzalez B. Overexpression of the nuclear factor κB inhibitor A20 is neurotoxic after an excitotoxic injury to the immature rat brain. Neurol Res. 2013;35:308–319. doi: 10.1179/1743132812Y.0000000139. [DOI] [PubMed] [Google Scholar]
  36. Qing G, Qu Z, Xiao G. Stabilization of basally translated NF-κB-inducing kinase (NIK) protein functions as a molecular switch of processing of NF-κB2 p100. J Biol Chem. 2005;280:40578–40582. doi: 10.1074/jbc.M508776200. [DOI] [PubMed] [Google Scholar]
  37. Saha RN, Pahan K. Differential regulation of Mn-superoxide dismutase in neurons and astroglia by HIV-1 gp120: Implications for HIV-associated dementia. Free Radic Biol Med. 2007;42:1866–1878. doi: 10.1016/j.freeradbiomed.2007.03.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Srinivasan D, Yen JH, Joseph DJ, Friedman W. Cell type-specific interleukin-1β signaling in the CNS. J Neurosci. 2004;24:6482–6488. doi: 10.1523/JNEUROSCI.5712-03.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sun SC. The noncanonical NF-κB pathway. Immunol Rev. 2012;246:125–140. doi: 10.1111/j.1600-065X.2011.01088.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Van Huffel S, Delaei F, Heyninck K, De Valck D, Beyaert R. Identification of a novel A20-binding inhibitor of nuclear factor-κ B activation termed ABIN-2. J Biol Chem. 2001;276:30216–30223. doi: 10.1074/jbc.M100048200. [DOI] [PubMed] [Google Scholar]
  41. Vandesompele J, Speleman F, Van Roy N, Laureys G, Brinskchmidt C, Christiansen H, Lampert F, Lastowska M, Bown N, Pearson A, Nicholson JC, Ross F, Combaret V, Delattre O, Feuerstein BG, Plantaz D. Multicentre analysis of patterns of DNA gains and losses in 204 neuroblastoma tumors: how many genetic subgroups are there? Med Pediatr Oncol. 2001;36:5–10. doi: 10.1002/1096-911X(20010101)36:1<5::AID-MPO1003>3.0.CO;2-E. [DOI] [PubMed] [Google Scholar]
  42. Xiao G, Fong A, Sun SC. Induction of p100 processing by NF-κB-inducing kinase involves docking IκB kinase α (IKKα) to p100 and IKKα-mediated phosphorylation. J Biol Chem. 2004;279:30099–30105. doi: 10.1074/jbc.M401428200. [DOI] [PubMed] [Google Scholar]
  43. Xiao G, Sun SC. Negative regulation of the nuclear factor κ B-inducing kinase by a cis-acting domain. J Biol Chem. 2000;275:21081–21085. doi: 10.1074/jbc.M002552200. [DOI] [PubMed] [Google Scholar]
  44. Yin L, Wu L, Wesche H, Arthur CD, White JM, Goeddel DV, Schreiber RD. Defective lymphotoxin-β receptor-induced NF-κB transcriptional activity in NIK-deficient mice. Science. 2001;291:2162–2165. doi: 10.1126/science.1058453. [DOI] [PubMed] [Google Scholar]

RESOURCES