Substrate-binding sites of UBR1, the ubiquitin ligase of the N-end rule pathway - PubMed (original) (raw)

CUP9 mutants, UBR1-CUP9 interactions, and effects of macromolecular crowding on the specificity of interaction between purified proteins. A, map of the 306-residue CUP9 and its truncation mutants.B, GST pulldown assays with UBR11-1140f and GST fusions to indicated CUP9 derivatives (see also the legend to Fig. 4, “Experimental Procedures” and the main text). Lane 1, 5% refers to a directly loaded sample of UBR11-1140f-containing yeast extract that corresponded to 5% of the amount of extract in GST pulldowns of this panel.Lane 2, GST pulldown with UBR11-1140f and GST fusion to wild-type CUP9. Lanes 3-7, same as lane 2 but with GST fusions to CUP9 mutants. Lane 8, same as lane 2, but GST lacking the CUP9 moiety. The band of UBR11-1140f (retained on glutathione-Sepharose beads) is indicated. C, lane 1, same as_lane 1_ in B, except that the test _S. cerevisiae_extract contained full-length fUBR1. Lanes 2-9, GST pulldowns with fUBR1 and the indicated GST-CUP9 fusions. As indicated above the lanes, all samples except those in lanes 3 and 5 contained a mixture of Arg-Ala (RA) and Leu-Ala (LA) dipeptides, each of them at 1 m

m

(see“Experimental Procedures”). Lane 10, same as lane 2, but with GST alone. D, lane 1, molecular mass markers;single and_double asterisks_ denote, respectively, 220 and 60 kDa. Lane 2, Coomassie-stained gel of purified full-length fUBR1. E, lane 1, same as lane 1 in C, but 2% of totalfUBR1 in the assay. Lanes 2 and 3, same as_lanes 3_ and 2 (respectively) in C, but an independent experiment. Lane 4, same as lane 1 but with purified fUBR1. Lanes 5 and 6, same as lanes 2 and 3 but with purified fUBR1 to which was added an empty (lacking fUBR1) S. cerevisiae extract. Note that purified fUBR1 assayed in the presence of “added-back” yeast extract exhibited specific (dipeptides-dependent) binding to GST-CUP9 (lanes 5 and 6), indistinguishably from fUBR1 that had not been initially purified (lanes 2 and 3). F, lanes 1-4, same as lanes 1-4 in E but an independent experiment. Lane 4, same as lane 1, but 10% of the input sample of the purified fUBR1 (see lane 2 in_D_). Lanes 5 and 6, same as lanes 2 and_3_ but with purified fUBR1. Note the lack of dependence of UBR1-CUP9 interaction on the presence of cognate dipeptides. Lanes 7_and 8, same as lanes 5 and 6 but GST pulldowns were carried out in the presence of added ovalbumin, at 10 mg/ml. G, lanes 1-3, same as lanes 1-3 in E or F, but an independent experiment. Lane 4, same as lane 4 in_F, but an independent experiment. Lanes 5 and 6, same as lanes 7 and 8 in F, but with oval buminat 50 mg/ml. Lanes 7 and 8, same as lanes 5 and_6_, but with purified EF1A added to the assay (to 1.2 mg/ml), instead of ovalbumin. Lanes 9 and 10, same as lanes 7 and_8_, but in the absence of added EF1A. H, Coomassie-stained gel of purified EF1A (see “Experimental Procedures”). I, lane 1, same as lane in F, but an independent experiment.Lanes 2 and 3, same as lanes 2 and 3 in_E_ or F, but with GST-CUP9L294P, a previously characterized missense mutant of CUP9 (in the region of its degron)that did not bind to UBR1 and was metabolically stabilized in vivo(8). Note the absence of binding of fUBR1 (in the extract) to GST-CUP9L294P, irrespective the absence or presence of dipeptides. Lane 4, same as_lane 4_ in F, but an independent experiment. Lanes 5_and 6, same as lanes 2 and 3 but with purifiedfUBR1. Note that purified fUBR1 binds to degron-impaired GST-CUP9L294P and does so irrespective of the presence or absence of dipeptides. J, lane 1, same as lanes 4 in F or_I, but an independent experiment. Lanes 2 and 3, same as lanes 5 and 6 in F, but in the presence of ovalbumin (at 50 mg/ml). Lanes 4 and 5, same as lanes 2 and 3 but with GST-CUP9L294P, instead of “wild-type” GST-CUP9. K, lane 1, same as lane 1_in F, but an independent experiment. Lanes 2 and 3, same as lanes 2 and 3 in F and G, but with GST-CUP9222-306, an N-terminally truncated CUP9. Note the retention of specific (dipeptides-modulated) binding of this CUP9 derivative tofUBR1. Lane 4, same as lane 4 in F, but an independent experiment. Lanes 5 and 6, same as lanes 2 and 3 but with purified fUBR1. Lanes 7 and_8, same as lanes 5 and 6, but with added ovalbumin (ov), at 50 mg/ml. Note the restoration of specific binding.L, an alternative UBR1-CUP9 binding assay in which the interactions were probed using coimmunoprecipitation of fUBR1 andpyCUP9his6, a GST-lacking, full-length CUP9 bearing the N-terminal pyepitope tag (see “Experimental Procedures”). FLAG-tagged fUBR1 in the yeast extract was incubated with beads-immobilized anti-FLAG antibody, followed by isolation offUBR1-bearing beads by centrifugation, washes with binding buffer, and incubation with either purifiedpyCUP9his6alone(lanes2_and_3, upperpanel) or in the presence of (added) UBR1-lacking yeast extract (lanes 4-7). The incubations were carried out in either the absence (lanes 2, 4, and 6) or the presence of Arg-Ala (RA) and Leu-Ala (LA) dipeptides, each of them at 1 m

m

. The beads containing immobilized fUBR1 were pelleted by centrifugation and washed, and the relative amounts of retainedpyCUP9his6 were then determined by SDS-PAGE and immunoblotting with anti-py antibody (see “Experimental Procedures”). The upper panel shows the amounts ofpyCUP9his6 that were bound to fUBR1 in the absence of added yeast extract (lanes 2 and 3) or in the presence of extract. Lane 1 shows the amount ofpyCUP9his6 that corresponds to 10% of its input in each of the binding assays. The lower panel is the result of SDS-PAGE and immunoblotting (with anti-FLAG antibody) of fUBR1 that was eluted from anti-FLAG beads by the FLAG peptide, to verify the approximate equality of “input” levels of fUBR1 in each assay. Note that in the absence of added (UBR1-lacking) yeast extract, the interaction betweenfUBR1 and pyCUP9his6 was nonspecific, in that it was independent of the absence or presence of cognate dipeptides (lanes 2 and 3; compare with lanes 4-7). Lanes 6 and_7_, same as lanes 4 and 5, but the inputpyCUP9his6 was 2-fold higher.