Figure 2.

Phosphorylation on Ser670 is required for Mdm2-mediated ubiquitination and down-modulation of GRK2 in the G2. (A) Lack of Ser670 phosphorylation stabilizes GRK2 protein levels in the G2 phase of the cell cycle. HeLa cells stably overexpressing wild-type (HeLa WT5) or mutant GRK2-S670A (HeLa A1) and parental cells were incubated in the absence (asynchronous, As) or presence of etoposide (Et) for the G2 synchronization as in Figure 1. Cells were lysed in RIPA buffer, and total GRK2 and tubulin were analyzed by immunoblotting. GRK2 levels were normalized with tubulin as a control of protein loading. Representative blots are included. Dotted lines indicate that intervening lanes have been spliced out. (B) Enhanced interaction of GRK2 with the cell cycle regulator Pin1 relays on Ser670 phosphorylation during the G2. Cellular lysates of asynchronously growing or G2-arrested HeLa WT5, HeLa A1 and parental cells were immunoprecipitated with a specific Pin1 antibody and presence of GRK2 analyzed by immunoblot. Immunoprecipitated Pin1 was immunodetected after membrane stripping and used to normalize Pin1-bound GRK2 levels. Input levels of GRK2 and Pin1 proteins and tubulin as control of loading were determined in whole lysates. Values are means ± SEM from 3-5 independent experiments performed in duplicate (* p < 0.05, ** p > 0.01). Illustrative blots are shown. (C) TUBEs affinity pulldowns were performed on HeLa WT5 and A1 cells cultured for 18h in the absence or presence of 5 µM etoposide for the G2 synchronization. Protein ubiquitination was stabilized with the addition of the proteasome inhibitor MG-132 (40 µM) 1h prior to cellular lysis. Both TUBE-bound polyubiquitinated proteins and unbound proteins were immunoblotted with the specific mouse monoclonal antibody c5/1.1 that detects GRK2 independently of its ubiquitination state, as reported previously [33]. After stripping, blots were re-probed with the anti-ubiquitin antibody FK2, which recognizes only polyubiquitinated proteins, confirming similar trapping by TUBEs in the G2-arrested cells. Representative blots of two experiments are shown. (D,E) In vitro ubiquitination patterns driven by Mdm2 vary with the extent of GRK2 phosphorylation on Ser670. (D) Recombinant GRK2-wt, S670A or K220R proteins were incubated in the presence of E1 and E2 protein (UbcH5b), GST-Mdm2 as E3 ligase and different purified ubiquitin proteins (WT, no-K (unable to form polyubiquitin chains into protein targets) and AA (unable to be conjugated to E1 thus blocking any ubiquitination)). Ubiquitination was detected by immunoblotting with an antibody able to recognize both free and pan-conjugated ubiquitin as indicated in Methods. Equal loading of GRK2 protein levels was confirmed by blot re-probing with a rabbit polyclonal antibody (C-15, Santa Cruz) that preferentially recognizes non-ubiquitinated forms of the kinase protein. (E) Effect of ubiquitin-chain elongation mutants on GRK2 polyubiquitination induced by Mdm2. Ubiquitination of GRK2-wt was performed as above in the presence of additional ubiquitin mutants that interfere with the assembly of specific polyubiquitin chains (K48R and triple mutant K29R/K48R/K63R (3KTR)). Samples were analyzed by Western blot using mouse monoclonal antibody c5/1.1. †, denotes protein bands in the blot corresponding to E1 or GST-Mdm2-related mono-ubiquitination signal (E1 enzyme, MW~110 KDa; GST-Mdm2, apparent MW~108). §, denotes GRK2-related multi-mono-ubiquitination band. #, denotes GRK2-related mono-ubiquitination band. * ubiquitin-unrelated GRK2 signal. Dashed boxes indicate polyubiquitinated proteins. Detailed information about the Western blots can be found in Figure S2.