O-Fucosylation by Pofut2: Identifying Targets and Refining the Consensus Sequence

Protein O-f fucosyltransferase 2 (Pofut2) is an ER resident glycosyltransferase that adds fucose to serines or threonines in properly folded Thrombospondin type 1 repeats (TSRs) containing the putative consensus sequence CXa(Xb)Xc( S/T)CXdXeG. O-Fucose can be elongated to a disaccharide by a β3-glucosyltransferase. Deficiencies in glycosylation of TSRs lead to severe developmental defects, presumably due to loss of glycosylation of one or more targets. Elucidating the molecular mechanism contributing to these defects requires an accurate list of target proteins and an understanding the function of O-fucosylation of targets. Mouse genome searches for this consensus sequence within a TSR yielded 52 potential targets. In this study, I refine the consensus sequence and list of Pofut2 targets by mapping sites of O-fucosylation using mass spectral methods and by mutagenesis of a representative TSR. In addition, I investigate the function of O-fucosylation on ADAMTSL2 using cell based assays and mutagenesis.;Using mass spectral methods I confirmed O-fucosylation of 16 TSRs within seven target proteins. My analysis led to the identification of amino acids within the consensus sequence that may block fucosylation. To refine the consensus sequence, I performed site directed mutagenesis on TSR3 from human thrombospondin 1, testing the importance of the invariant residues (cysteines and glycine) and “X” positions by inserting amino acids of predicted targets with distinct characteristics from those found in confirmed O-fucose sites. I demonstrated that mutating Xa to aspartic acid or leucine severely limits Pofut2 activity and that the conserved glycine is not essential for fucosylation.;Mutations to predicted Pofut2 target ADAMTSL2 lead to the human disorder geleophysic dysplasia (GD). Two GD mutations fall within consensus sequences for O-fucosylation, including a serine to leucine mutation at Xa. I documented O-fucose on TSRs from ADAMTSL2 and demonstrated secretion of ADAMTSL2 requires fucosylation. In collaboration with Andrew Taibi, I showed GD mutations cause decreased protein secretion. In a unique case, where N-glycan and O-fucose consensus sequences overlap, I demonstrate removal of the N-glycan site or both N-glycan site and O-fucose site causes decreased secretion. My results suggest that some GD mutations may cause reduced secretion of ADAMTSL2 due to alterations in TSR glycosylation.