YIPF1, YIPF2, and YIPF6 are medial-/trans-Golgi and trans-Golgi network-localized Yip domain family proteins, which play a role in the Golgi reassembly and glycan synthesis.

In this study, we attempted to explore the function of three uncharacterized mammalian homologs of yeast Yip domain family proteins-YIPF6, a homolog of Yip1p, and YIPF1 and YIPF2, which are homologs of Yif1p. Immunofluorescence staining revealed that YIPF1, YIPF2, and YIPF6 mainly localize in the medial-/trans-Golgi and also partially in the trans-Golgi network (TGN). On treatment with brefeldin A (BFA), the homologs co-migrated partly with medial-/trans-Golgi markers and also with a TGN marker in earlier time point, but finally redistributed within cytoplasmic punctate structures that were distinct from medial-/trans-Golgi and the TGN markers. YIPF6 formed a stable complex separately with YIPF1 and YIPF2, and knockdown of YIPF6 reduced YIPF1 and YIPF2 levels. These results suggest that YIPF6 forms complexes with YIPF1 and YIPF2 for their stable expression and localization within the Golgi apparatus. Knockdown experiments showed that YIPF1 and YIPF2, by contrast, are not necessary for the expression and localization of YIPF6. The structure of the Golgi apparatus and its disassembly after BFA treatment were not significantly affected by the knockdown of YIPF1, YIPF2, or YIPF6. However, reassembly of the Golgi apparatus after the removal of BFA was markedly delayed by the knockdown of YIPF1 and YIPF2, but not by that of YIPF6. These results strongly suggest that free YIPF6 after disassociating with YIPF1 and YIPF2 interferes with the reassembly of the Golgi apparatus. Knockdown of YIPF1 and YIPF2, but not that of YIPF6, also reduced intracellular glycans in HT-29 cells. Thus, we confirmed that YIPF1, YIPF2, and YIPF6 play a significant role in supporting normal glycan synthesis.

Low cytoplasmic pH reduces ER-Golgi trafficking and induces disassembly of the Golgi apparatus.

The Golgi apparatus was dramatically disassembled when cells were incubated in a low pH medium. The cis-Golgi disassembled quickly, extended tubules and spread to the periphery of cells within 30 min. In contrast, medial- and trans-Golgi were fragmented in significantly larger structures of smaller numbers at a slower rate and remained largely in structures distinct from the cis-Golgi. Electron microscopy revealed the complete disassembly of the Golgi stack in low pH treated cells. The effect of low pH was reversible; the Golgi apparatus reassembled to form a normal ribbon-like structure within 1-2h after the addition of a control medium. The anterograde ER to Golgi transport and retrograde Golgi to ER transport were both reduced under low pH. Phospholipase A2 inhibitors (ONO, BEL) effectively suppressed the Golgi disassembly, suggesting that the phospholipase A2 was involved in the Golgi disassembly. Over-expression of Rab1, 2, 30, 33 and 41 also suppressed the Golgi disassembly under low pH, suggesting that they have protective role against Golgi disassembly. Low pH treatment reduced cytoplasmic pH, but not the luminal pH of the Golgi apparatus, strongly suggesting that reduction of the cytoplasmic pH triggered the Golgi disassembly. Because a lower cytoplasmic pH is induced in physiological or pathological conditions, disassembly of the Golgi apparatus and reduction of vesicular transport through the Golgi apparatus may play important roles in cell physiology and pathology. Furthermore, our findings indicated that low pH treatment can serve as an important tool to analyze the molecular mechanisms that support the structure and function of the Golgi apparatus.