Novel insights into the fucose metabolism – challenging the old dogma

GDP‐Fucose used for glycan fucosylation is synthesized de novo from either glucose (Glc) or mannose (Man) or utilized directly from salvaged/exogenous fucose (Fuc). Meticulous radioisotope dilution studies performed in the 1970’s only in HeLa cells using 0.3 μM Fuc showed that the de novo pathway provides >90% of GDP‐Fuc and salvage <10%. Employing different Fuc‐specific lectins, we showed that at low concentrations (<20 μM), the entry of Fuc into cells relies on transport via GLUT1 ( SLC2A1 ), but above this, most Fuc enters via amiloride‐sensitive macropinocytosis. Inhibitors of both clathrin‐ and caveolin‐dependent endocytosis had no effect on uptake. These findings were further confirmed using 3 H Fuc in four cell lines. We used differentially labelled 13 C Glc, Man and Fuc to study contributions of each pathway to Fuc in N‐glycans in six cell lines. Very low concentrations (0.5–30 μM) of exogenous Fuc progressively inhibit the de novo pathway, and completely by ~50 μM. Surprisingly, GDP‐Fuc derived from Glc is much more sensitive to this inhibition than that derived from Man, even though both contribute to GDP‐Man synthesis. When cultured in 30–50 μM Fuc, cells not only inhibit the de novo pathway, but also prefer exogenous over salvaged Fuc. Cells seem to distinguish multiple pools of GDP‐Fuc based on their origin. Using LC‐MS/MS and eight different cell lines, we studied the ongoing biosynthesis of >150 fucosylated N‐glycans containing 1–4 Fuc residues by comparing the proportion of 13 C Fuc incorporation at various concentrations for 24h. Over the labeling period, cells had to choose between GDP‐Fuc synthesized de novo or from exogenous fucose. N‐glycans with a single Fuc show a proportional increase in 13 C‐labeled species with increasing concentration of label. To our surprise, glycans with 2, 3, and 4 Fuc each required progressively higher exogenous 13 C Fuc to become fully labeled. Based on LC‐MS/MS and lectin staining, 1–6αFuc labels more efficiently with exogenous Fuc than 1–2α, 1–3α and 1–4αFuc, suggesting differential access to exogenous sugar that depends partially on Fuc linkage. K m differences between fucosyltransferases cannot explain preferential use of GDP‐Fuc synthesized from exogenous monosaccharide over de novo produced substrate. Rather, it suggests the existence of separate, ill‐defined pools of GDP‐Fuc. Altogether it raises the question whether salvaged fucose equivalent to exogenous Fuc? Their behavior clearly differs from de novo synthesized GDP‐Fuc. Support or Funding Information The Rocket Fund and R01DK099551.