Recycling of Golgi glycosyltransferases requires direct binding to coatomer

Abstract

The glycosyltransferases of the mammalian Golgi complex must recycle between the stacked cisternae of that organelle to maintain their proper steady-state localization. This trafficking is mediated by COPI-coated vesicles, but how the glycosyltransferases are incorporated into these transport vesicles is poorly understood. Here we show that the N-terminal cytoplasmic tails (N-tails) of a number of Golgi glycosyltransferases which share a ϕ-(K/R)-X-L-X-(K/R) sequence bind directly to the δ- and ζ-subunits of COPI. Mutations of this N-tail motif impair binding to the COPI subunits, leading to mislocalization of the transferases to lysosomes. The physiological importance of these interactions is illustrated by mucolipidosis III patients with missense mutations in the N-tail of GlcNAc-1-phosphotransferase that cause the transferase to be rapidly degraded in lysosomes. These studies establish that direct binding of the N-tails of mammalian Golgi glycosyltransferases with COPI subunits is essential for recycling within the Golgi.

A tale of short tails, through thick and thin: investigating the sorting mechanisms of Golgi enzymes

“…Curiously, C2GnT, which was previously reported to interact with coatomer indirectly via GOLPH3, was also found to interact directly with coatomer [56,58,63]. The cytoplasmic tail of C2GnT, appears to have a putative GOLPH3 binding site that overlaps the d-COP/f1-COP binding motifs, suggesting redundancy in the sorting signals in the cytoplasmic tail of at least one Golgi enzyme (Table 1), and disruption of both is required to ablate an interaction with coatomer [63]. However, it should be noted that in this study deletion of GOLPH3 did not affect the Golgi localisation of C2GnT, in contrast to what had been reported previously [54,61].…”

“…The poly‐arginine stretch 3 RGERRRR 9 in the cytoplasmic tail of human glucosidase‐1 ( MOGS ) is positioned 29 residues distal to the membrane and is sufficient to target the protein to the ER . In contrast, a similar 1 MRRRSR 6 motif positioned immediately proximal to the membrane in the tail of GalNAc‐T2 (( GALNT2 ) is sufficient for Golgi‐targeting . The molecular identity of the receptors and/or coat components involved in either arginine‐based interaction is unclear but it would be surprising if it was not COPI‐dependent.…”

“…The d-COP/f1-COP consensus binding motif was identified as φ(K/R)XLX (K/R) and it was shown that mutagenesis of this motif was, in most instances, sufficient to both ablate coatomer binding in vitro and disrupt Golgi-localisation in vivo [63]. Curiously, C2GnT, which was previously reported to interact with coatomer indirectly via GOLPH3, was also found to interact directly with coatomer [56,58,63]. The cytoplasmic tail of C2GnT, appears to have a putative GOLPH3 binding site that overlaps the d-COP/f1-COP binding motifs, suggesting redundancy in the sorting signals in the cytoplasmic tail of at least one Golgi enzyme (Table 1), and disruption of both is required to ablate an interaction with coatomer [63].…”

Multiple features within the syntaxin Sed5p mediate its Golgi localization

“…To date, relatively few N‐terminal COPI‐binding motifs have been characterized . In mammalian cells, some Golgi resident glycosyltransferases have been shown to directly interact with coatomer however the motif that mediates this association—“ϕ‐(K/R)‐X‐L‐X‐(K/R)”—is not similar to the motif we have identified in Sed5p (Figure ). The COPI‐coatomer binding motif from Sed5p is also able to mediate binding to mammalian coatomer, and thus a binding site for N‐terminal tribasic motifs on COPI‐coatomer would seem to be an evolutionarily conserved feature.…”

“…In many respects, the Golgi is unique among endomembrane organelles as its individual cisternae from cis , medial to trans harbor distinct enzyme compositions critical to the glycosylation of proteins and lipids—the glycosyltransferases and glycosidases. The mechanisms by which these enzymes are retained in the Golgi have been extensively studied . Moreover, following the distribution of Golgi membrane proteins by live cell imaging in yeast cells has provided critical insights into the biogenesis of this organelle .…”