A Proteomic Strategy for Quantifying Polyubiquitin Chain Topologies

Abstract We present a method to probe the structure of polyubiquitin (polyUb) chains by mass spectrometry. As polyUb chains can be linked by any of seven lysine residues in the sequence of ubiquitin, the structural diversity of the chains could be enormous (e.g., as many as 70 configurations for tri‐Ub). To quantify the frequency of every lysine linkage, polyUb chains were digested by trypsin and analyzed by tandem mass spectrometry using synthetic isotope‐labeled peptides as internal standards. Employing a purified ubiquitin dimer to optimize conditions for trypsinization, we found that all tryptic sites except K33 could be reliably cleaved under insolution and in‐gel conditions. Because K33 miscleavage led to large underestimation of K29 linkage, its digestion efficiency was calculated to correct the measurement. The reverse‐phase chromatography was also adjusted to capture the small hydrophilic peptide of K29 linkage. In addition, the analysis of the K6‐linkage peptide was complicated by methionine oxidation and therefore the sample was treated with hydrogen peroxide to fully oxidize methionine to sulfoxide. While the method was sensitive enough to detect K48 and K63 linkages directly from total yeast extract, all linkages were quantified from pre‐enriched Ub‐conjugates, revealing that the concentration ratio of K6:K11:K27:K29:K33:K48:K63 was ˜40:120:20:15:5:100:40 in yeast during the log growth phase, and the sum of non‐K48 linkages was more than twice as abundant as the canonical K48 linkage. Moreover, increasing the concentration of ubiquitin in yeast raised the level of almost all linkages. Inhibiting proteasome activity augmented only K48‐ and K29‐linked chains, recapitulating their roles in proteolysis and suggesting non‐proteolytic functions of other linkages in vivo.