Abstract
O(2) sensing in diverse protozoa depends on the prolyl 4 hydroxylation of Skp1 and modification of the resulting hydroxyproline with a series of five sugars. In yeast, plants, and animals, Skp1 is associated with F-box proteins. The Skp1-F-box protein heterodimer can, for many F-box proteins, dock onto cullin-1 en route to assembly of the Skp1-cullin-1-F-box protein-Rbx1 subcomplex of E3(SCF)Ub ligases. E3(SCF)Ub ligases conjugate Lys48-polyubiquitin chains onto targets bound to the substrate receptor domains of F-box proteins, preparing them for recognition by the 26S proteasome. In the social amoeba Dictyostelium, we found that O(2) availability was rate-limiting for the hydroxylation of newly synthesized Skp1. To investigate the effect of reduced hydroxylation, we analyzed knockout mutants of the Skp1 prolyl hydroxylase and each of the Skp1 glycosyltransferases. Proteomic analysis of co-immunoprecipitates showed that wild-type cells able to fully glycosylate Skp1 had a greater abundance of an SCF complex containing the cullin-1 homolog CulE and FbxD, a newly described WD40-type F-box protein, than the complexes that predominate in cells defective in Skp1 hydroxylation or glycosylation. Similarly, the previously described FbxA-Skp1CulA complex was also more abundant in glycosylation-competent cells. The CulE interactome also included higher levels of proteasomal regulatory particles when Skp1 was glycosylated, suggesting increased activity consistent with greater association with F-box proteins. Finally, the interactome of FLAG-FbxD was modified when it harbored an F-box mutation that compromised Skp1 binding, consistent with an effect on the abundance of potential substrate proteins. We propose that O(2)-dependent posttranslational glycosylation of Skp1 promotes association with F-box proteins and their engagement in functional E3(SCF)Ub ligases that regulate O(2)-dependent developmental progression.
Highlights
From the ‡Department of Biochemistry & Molecular Biology, Oklahoma Center for Medical Glycobiology, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73104; §Department of Biochemistry & Molecular Biology, Oklahoma State University, Stillwater, Oklahoma 74078
The cullin-RING ubiquitin ligases (CRLs)1 are a prominent subgroup of these enzymes [1] and consist of an E3 architecture that includes a substrate receptor, an adaptor, the cullin scaffold, the RING protein, and an exchangeable E2 ubiquitin donor that has been charged with ubiquitin (Ub) by an E1 enzyme
Deneddylation mediated by the eight-subunit COP9 signalosome is required for in vivo activity, suggesting that Cand1 serves as a substrate exchange factor to allow for re-equilibration of SCF complexes from preexisting subunits
Summary
Metabolic Labeling and Isolation of Skp1—Cells (gnt1–) were grown axenically in HL-5 medium supplemented with 50 g/ml streptomycin and 50 g/ml ampicillin until they achieved a density of 2 ϫ 106 cells per milliliter. HO-Skp was isolated via incubation with 5 g of affinity-purified UOK85 rabbit IgG followed by incubation with Protein A beads and centrifugation as described above. Cell lysates from clones were initially analyzed for modification of the culE locus via Western blotting using the M2 anti-FLAG mAb (Sigma) and verified via PCR using primers BsRF and CE3FRev. To disrupt phyAO, the floxed blasticidin S resistance cassette was removed through transient expression of Cre recombinase as previously described [21]. Co-immunoprecipitation—Aliquots of 3 ϫ 107 slug cells, or 1.5 ϫ 107 exponentially growing vegetative cells in order to match total protein content, were lysed in 250 l of IP Buffer (0.2% or 1.0% Nonidet P-40 (as indicated), 50 mM HEPES-NaOH, pH 7.4, or 50 mM Tris-HCl, pH 8.0 (as indicated), 50 mM NaCl, 1 mM MgCl2, 0.1 mM EDTA, and the above-mentioned protease inhibitors) and allowed to incubate on ice for 10 min. To ensure a high level of rigor with regard to the protein–protein interactions described, proteins reported in supplemental Tables S2–S4 were excluded from further analysis if their spectral counts from negative-control co-IPs were more than zero and Ͼ10% of those observed in the “bait” pulldowns, or if they were ribosomal subunits, chaperones (except for a subgroup in the FLAG-FbxD co-IPs), or intraorganellar or secretory proteins (most members of the latter category were already excluded by the previous criteria)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
