Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Apr 1;53(12):2064-73.
doi: 10.1021/bi401551r. Epub 2014 Mar 18.

Probing Kinase Activation and Substrate Specificity With an Engineered Monomeric IKK2

Affiliations
Free PMC article

Probing Kinase Activation and Substrate Specificity With an Engineered Monomeric IKK2

Arthur V Hauenstein et al. Biochemistry. .
Free PMC article

Abstract

Catalytic subunits of the IκB kinase (IKK), IKK1/IKKα, and IKK2/IKKβ function in vivo as dimers in association with the necessary scaffolding subunit NEMO/IKKγ. Recent X-ray crystal structures of IKK2 suggested that dimerization might be mediated by a smaller protein-protein interaction than previously thought. Here, we report that removal of a portion of the scaffold dimerization domain (SDD) of human IKK2 yields a kinase subunit that remains monomeric in solution. Expression in baculovirus-infected Sf9 insect cells and purification of this engineered monomeric human IKK2 enzyme allows for in vitro analysis of its substrate specificity and mechanism of activation. We find that the monomeric enzyme, which contains all of the amino-terminal kinase and ubiquitin-like domains as well as the more proximal portions of the SDD, functions in vitro to direct phosphorylation exclusively to residues S32 and S36 of its IκBα substrate. Thus, the NF-κB-inducing potential of IKK2 is preserved in the engineered monomer. Furthermore, we observe that our engineered IKK2 monomer readily autophosphorylates activation loop serines 177 and 181 in trans. However, when residues that were previously observed to interfere with IKK2 trans autophosphorylation in transfected cells are mutated within the context of the monomer, the resulting Sf9 cell expressed and purified proteins were significantly impaired in their trans autophosphorylation activity in vitro. This study further defines the determinants of substrate specificity and provides additional evidence in support of a model in which activation via trans autophosphorylation of activation loop serines in IKK2 requires transient assembly of higher-order oligomers.

Figures

Figure 1
Figure 1
Interactions between individual IKK2 subunits. (A) As revealed by X-ray crystallography, a single IKK2 subunit contains three domains: the amino-terminal kinase domain (KD), a central ubiquitin-like domain (ULD), and a carboxy-terminal scaffold dimerization domain (SDD). The proximal portion of the SDD lies adjacent to the KD and ULD opposite the distal end. (B) Human IKK2 dimer in its open conformation, as it appears in X-ray crystal structures. Dimerization is mediated by distal portions of the SDD. (C) In their open dimeric conformation, individual IKK2 subunits can associate through an extensive V-shaped interface involving residues from the KD, ULD, and SDD. (D) Association through the V-shaped interface supports KD–KD interactions through which IKK2 can become active via activation loop trans autophosphorylation.
Figure 2
Figure 2
Engineered monomeric IKK2 (IKK2mono). (A) Schematic diagram of the domain organization of the IKK2 subunit (above) and IKK2mono (below). Coloring and abbreviations are shown as in Figure 1. Numbers correspond to the domain borders in the human IKK2 subunit, and GG refers to the diglycine linker joining SDD helices α2s and α4s. (B) Ribbon diagram model of the expected IKK2mono structure on the basis of the X-ray crystal structure of human IKK2EE(11–669). (C) Coomassie-stained 10% SDS-PAGE analysis monitoring IKK2mono purification by Ni affinity (lanes 2–6) and size-exclusion (lanes 7–9) chromatography and the final concentrated protein (lane 10). (D) Representative chromatogram of size-exclusion chromatography on IKK2mono.
Figure 3
Figure 3
Circular dichroism spectroscopy of purified IKK2 proteins. The mean residue molar ellipticity for IKK2EEmono (black line) and IKK2EE(11–669) (gray line) is plotted as a function of wavelength. Both samples show strong α-helical signals, with the monomer showing a slight decrease, as predicted because of the removal of three largely α-helical passes at the dimer interface.
Figure 4
Figure 4
IKK2mono is specific for IκBα residues S32 and S36. In vitro kinase assays were performed with full-length human IKK2EE (lanes 1–6), IKK2EE(11–669) (lanes 7–11), IKK2EE(1–420) (lanes 12–16), and IKK2mono (lanes 17–21) against both native sequence or S32A/S36A (AA) GST–IκBα(1–54) substrates as well as native, S32A/S36A, or C-terminal S283E/S288E/T291E/S293E/T296E/T299E (E6) mutant full-length IκBα substrates. Two-hundred nanograms of kinase was used in each assay except for the IKK2EE(1–420) reactions, which contained 400 ng of enzyme each. Substrate specificity is revealed by autoradiography (top panels) and Coomassie-stained SDS-PAGE substrate loading controls (bottom panels).
Figure 5
Figure 5
IKK2 oligomerization state in solution. (A) Full-length IKK2 (black line) and IKK2EE(11–669) (dark gray) are both dimers in solution, as revealed by multiangle laser light scattering following size-exclusion chromatography (SEC–MALLS). IKK2mono (light gray) is a monomer. (B) Increasing the IKK2mono concentration to as high as 187 μM does not significantly affect its profile as a monomer in solution.
Figure 6
Figure 6
Activation loop phosphorylation activity of IKK2 dimers and monomers. (A) Western blot analysis with an anti-phospho-S181 antibody on recombinant human IKK2 proteins purified from baculovirus-infected Sf9 insect cells before and after treatment with Mg/ATP (top). Only IKK2mono in which V-shaped interface residues I413 and L414 are both mutated to alanine fails to be recognized by the antibody (lanes 13–16). Western with anti-His antibody confirms the level of protein in each reaction (bottom). (B) Catalytically inactive version of full-length IKK2 in which catalytic base D145 is mutated to Asn permits direct detection of activation loop trans autophosphorylation. Western blot reveals that constitutively active versions of both dimeric and monomeric IKK2 bearing the S177E/S181E mutations (EE) phosphorylate the IKK2D145N substrate in vitro (lanes 21 and 22).
Figure 7
Figure 7
Role of trans autophosphorylation in activation of IKK2mono. (A) Autoradiography of in vitro kinase assays with GST–IκBα(1–54) as substrate and various versions of IKK2mono as enzymes (top). Both mutation of the V-shaped (I413A/L414A) and KD–KD (V229A/H232A) interfaces yields IKK2mono proteins that almost completely lack kinase activity toward GST–IκBα(1–54) substrate (lanes 1–3). Conversion of both activation loop residues S177 and S181 to phosphomimetic glutamic acids (EE) results in IKK2mono enzymes that display significantly higher kinase activity. Coomassie-stained SDS PAGE substrate and anti-His western blot enzyme loading controls are shown below. (B) Western blot with anti-phospho-S181 antibody reveals that only native sequence IKK2mono is capable of activation loop autophosphorylation (compare lanes 6, 8, and 9; above). Anti-His western blot loading control (bottom).
Figure 8
Figure 8
Activation loop trans autophosphorylation by mutant IKK2mono enzymes can be partially recovered at high concentration. The activation loop phosphorylation state of the inactive IKK2D145N mutant is monitored via western blot with anti-phospho-S181 antibody. Enzyme concentrations are given; IKK2D145N was 300 nM in each reaction. Dimeric and constitutively active IKK2EE(11–669) readily phosphorylates activation loop S181 (lanes 1 and 2). IK2EEmono also phosphorylates the inactive IKK2D145N substrate, although a higher concentration of enzyme is required to observe comparable levels of phosphorylation (compare lanes 3 and 4, 11 and 12, and 19 and 20). Mutation of V-shaped interface residues I413 and L414 to alanine yields an IKK2EEmono enzyme that is even more defective in trans autophosphorylation of IKK2D145N in vitro. Mutation of KD–KD interaction residues V229 and H232 to alanine yields an IKK2EEmono enzyme that fails to trans autophosphorylate even when provided in molar excess to the IKK2D145N substrate protein.

Similar articles

See all similar articles

References

    1. Rothwarf D. M.; Zandi E.; Natoli G.; Karin M. (1998) IKKγ is an essential regulatory subunit of the IκB kinase complex. Nature 395, 297–300. - PubMed
    1. Karin M.; Ben-Neriah Y. (2000) Phosphorylation meets ubiquitination: The control of NF-κB activity. Annu. Rev. Immunol. 18, 621–663. - PubMed
    1. Israël A. (2010) The IKK complex, a central regulator of NF-κB activation. Cold Spring Harbor Perspect. Biol. 2, a000158-1–a000158-14. - PMC - PubMed
    1. Rudolph D.; Yeh W. C.; Wakeham A.; Rudolph B.; Nallainathan D.; Potter J.; Elia A. J.; Mak T. W. (2000) Severe liver degeneration and lack of NF-κB activation in NEMO/IKKγ-deficient mice. Genes Dev. 14, 854–862. - PMC - PubMed
    1. Delhase M.; Hayakawa M.; Chen Y.; Karin M. (1999) Positive and negative regulation of IκB kinase activity through IKKβ subunit phosphorylation. Science 284, 309–313. - PubMed

Publication types

MeSH terms

Feedback