Schröfelbauer et al.NEMO directs IKK to NFB

Supplemental Figure S1

Figure S1.(A) Western blot analysis of phosphorylation of IB and p65 in the presence or absence of MG132 upon stimulation with IL-1. (B) Kinase assay in NEMO deficient cells stably expressing IKKEE with or without reconstitution of WT-NEMO. Kinase reaction was performed with immunoprecipitated IKK in the presence of GST-IB1-54 as substrate.

Supplemental Figure S2

Figure S2.(A) Expression levels of wild-type MEFs and NEMO deficient cells stably reconstituted with indicated NEMO expression vectors. (B) Western blot analysis of IB, IB and IB steady-state levels in NEMO deficient cells stably expressing IKKEE and either empty vector (EV), WT or ZF mutant NEMO. (C) NEMO associated IKK kinase activity was measured in indicated cells in an IP-kinase assay using either a N-terminal fragment (GST-IB1-54) or full length IκB (His- IBFL). (D, E) NEMO associated kinase activity was measured in indicated cell lines upon stimulation with (D) LPS (E) TNF. (F, G) NFB DNA binding activity was analyzed in NEMO deficient cells reconstituted with indicated NEMO mutants upon stimulation with (F) LPS and (G) TNF. (H) Kinase activity (top) and EMSA (bottom) of WT MEFs and NEMO deficient cells stably expressing C25 stimulated with indicated amounts of IL-1.

Supplemental Figure S3

Figure S3. Interaction between Myc-NEMO, HA-IKK and Flag-IB was analyzed in transfected 293T cells by immunoprecipitating NEMO using a -Myc antibody. Co-immunoprecipitation of IKK and IB was analyzed by western blotting.

Supplemental Figure S4

Figure S4. 293T cells were transfected with increasing amounts of NEMO together with HA-IKKEE. IKK was immunoprecipitated with anti-HA antibody and in vitro IKK kinase activity was measured using GST-IB1-54 as substrate.

Supplemental Figure S5

Figure S5.(A) Western blot analysis of steady state phosphorylation of p105 in indicated cells in the presence or absence of MG132. (B) Western blot analysis of steady state p65 phosphorylation in indicated cell lines with or without MG132 treatment. (C)In vitro IKKEE kinase assay with saturating amounts of FL-IB, p105 and p65AD substrates in the presence or absence of NEMO. Time indicates the kinase reaction time.

Supplemental Figure S6

Figure S6. (A) NEMO deficient cells reconstituted with indicated NEMO mutants were stimulated with LPS. Phosphorylation of p65 and IB was analyzed by western blotting. (B) NEMO deficient cells expressing C389/93S mutated NEMO and IB, IB, IB deficient cells were treated with the IKK inhibitor sc-514 (50 g/ml) and stimulated with IL-1 for indicated times. Phosphorylation of p105 and p65 was analyzed by western blotting. (C) NEMO associated kinase activity

was measured in an in vitro kinase assay of NEMO deficient cells stably reconstituted with WT or C25 NEMO upon starvation for indicated times using FL-IB as substrate.

Supplemental Experimental Procedures

Mathematical Modeling

A simple ordinary differential equation (ODE's)-based model representing the association and dissociation of NEMO, IKK, and a generic IκB into binary and ternary complexes was used to study the effect of NEMO concentration on the level of ternary complex. Reactions corresponding to eq 1 were included in the model and assumed to follow mass action kinetics.

NEMO + IKK <-> IKK:NEMO(k1ON,OFF)(1.a)

NEMO + IκB <-> NEMO:IκB (k2ON,OFF)(1.b)

IKK:NEMO + IκB <-> IKK:NEMO:IκB (k3 ON,OFF)(1.c)

NEMO:IκB + IKK <-> IKK:NEMO:IκB (k4 ON,OFF)(1.d)

To generate figure 3H the steady state for the system of differential equations representing the model was solved for different initial concentrations of NEMO. The function NDSolve from the package Mathematica 7 (2008, Wolfram Research Inc, Champaign, IL) was used to calculate the numerical solutions. The parameters for the model were chosen to illustrate the existence of an optimum level of NEMO that leads to maximum formation of ternary complex taking into account the relative stability of the different complexes (k1ON/OFF=1/0.01, k2ON/OFF=1/0.1, k3ON/OFF=1/0.001, k4ON/OFF=1/0 arbitrary units). As long as the ternary is more stable than either binary complex the choice of parameters only affects the shape of the peak.

The effect of competitive substrates was analyzed in terms of the quasi-steady state approximation as discussed by Schnell et al (Mendoza, 2000). Under this approximation, the fraction of an enzyme (IKK in this case) bound to IκB and the alternative substrate S as the reaction proceeds can be estimated as per eq 2:

(2.a)

(2.b)

Where the KM’s are given by

(3)

Where kON and kOFF are the complex formation and dissociation constants and kENZ is the kinetic constant for the catalytic phosphorylation step (i=IκB,S). Under the QSSA, the rate of phosphorylation of each substrate is proportional to the concentration of its respective complex with IKK.

The effect of NEMO has been included as a mutiplicative modifier to kONIkB. Figure 4 was generated assuming an initial concentration of IKK and IκB of 1 and 100 a.u. respectively and varying the value of the NEMO modifier and the ratio S/IκB as indicated. The base value for the parameters kON, and kOFF, and kENZ are taken to be equal for S and IκB; 1x104, 1x103, and 15 in arbitrary units of mass and time. These values were chosen to reflect the observed time scale on which IκB is phosphorylated. but it is worth pointing out that under the QSSA approximation the effects discussed are a consequence of the functional form of equations 2 and 3.