Kovalenko et al.

Supplementary information

Figures Legends:

Supplementary Fig. 1: «IKK is not the CYLD kinase»

HEK 293T cells were transfected and analyzed as described in Fig. 2a. A dominant negative version of IKK2 which is catalytically defective (IKK2 SS/AA) but still able to interact with NEMO (data not shown) was used to determine whether endogenous IKK was directly or indirectly responsible for CYLD phosphorylation.

Supplementary Fig. 2:«NEMO ubiquitination does not induce proteasome-dependent degradation»

a. HEK 293T cells were transfected with 500 ng of HA-NEMO. Twenty-four hours later, cells were treated for 6 hours with 20 M of the proteasome inhibitor MG132. Cytoplasmic extracts were prepared and NEMO expression was assessed with anti-NEMO. b. Same protocol as in a but NEMO was co-transfected with control vector (Pc) or HA-Ubiquitin (Ub) in the absence (lane 2) or presence (lane 3) of MG132.

Supplementary Fig. 3: «(N393)CYLD exhibits restricted substrate specificity and behaves as a specific negative regulator of TRAF2/NF-B signaling»

a. Deubiquitinating activity of (N393)CYLD. HEK 293T were transfected with 500 ng of TRAF2, 500 ng of HA-Ubiquitin and either 2ug of pcDNA3 (lanes 1,2), wt CYLD-Flag (lanes 3,4) or (N393)CYLD (lanes 5,6). TRAF2 ubiquitination was assessed after Western blotting with anti-TRAF2 and CYLD expression with anti-Flag. b. Same as in a., except that 500 ng of HA-NEMO were transfected instead of TRAF2 and NEMO ubiquitination was analyzed with anti-NEMO. c. Same as in a. except that 500 ng of Flag-TRAF6 were transfected instead of TRAF2 and TRAF6 ubiquitination was analyzed with anti-Flag. d. Same as in a. except that ubiquitination of total cytoplasmic proteins was analyzed with anti-HA after transfecting 500 ng of HA(K48R)ubiquitin.e. Luciferase assay of NF-B activation. HEK 293T cells were transfected with TRAF2 or TRAF6 and the effect of co-transfecting wt CYLD or (N393)CYLD analyzed.

Supplementary Fig. 4: «CYLD acts upstreamof IKK»

a. CYLD overexpression inhibits IKK activation by TNF. HEK 293T cells were transfected with 10 ng of VSV-IKK2 and 2 g of PcDNA3 (Pc) or wt CYLD (CYLD). After 24hours, cells were treated or not with TNF (10 ng/ml for 10 min), the cytoplasmic extracts prepared and immunoprecipitated with anti-VSV. The IKK kinase assay was carried out as described in Yamaoka et al.4. b. CYLD overexpression inhibits TNF-induced IB degradation. HEK 293T cells were transfected with PcDNA3, Myc-IB0.5 gor Myc-IB plus CYLD (2g). Twenty-four hours later cells were stimulated or not with 10 ng/ml TNF for the times indicated and IB degradation was analyzed by Western blotting with anti-IB. IB on the right of the figure indicates the endogenous molecule. c. CYLD does not inhibit the IB degradation step. HEK 293T cells were transfected with Myc IB (0.5 g) and the indicated combinations of the empty vector (Pc), IKK2 SS/EE (a constitutively active form of IKK2; IKK2*, 0.5 g) and wild-type CYLD (2g). Twenty four hours later, cells were treated or not with 20 M of MG132 for 6 hours, as indicated. Cytoplasmic extracts were analyzed with anti-IB. In this setting, CYLD overexpression did not affect IB induced phosphorylation (as shown here using MG132), IB ubiquitination (as shown in Fig. 3e) or IB degradation (as shown here in the absence of MG132).