Role of trans-Golgi network vesicles in mitochondrial fusion

Abstract

The continuous cycles of membrane fusion and fission are hallmarks of mitochondrial behaviour. In recent decades, the study of mitochondrial dynamics has come to the forefront of cell biology due to a growing recognition that the ever-changing shape of mitochondria has a basic impact on their function, affecting a range of essential processes, from energy production to endocrine signalling. Under homeostatic conditions, the processes of mitochondrial fusion and fission are balanced, creating a network of mitochondria in the cell of intermediate length and moderate interconnectivity. However, when cells are stressed, for example by nutrient surplus or deficit, the population of mitochondria fragments or elongates, respectively. While mitochondrial fusion and fission are both essential for maintaining cellular homeostasis, the processes are unequal in terms of the difficulties in elucidating their underlying mechanisms. The reason for this disparity is largely due to the Abbe diffraction limit, a natural constraint on resolving objects below approximately 200 nm. This is not a significant problem for the study of mitochondrial division, because the process involves a divergence of structures, so quantification is not limited by conventional microscopy techniques. Mitochondrial fusion, on the other hand, involves the convergence of membranes, according to which it is unfeasible to distinguish genuine membrane fusion events from random collisions or stable contacts. Studies of mitochondrial division have revealed roles for other organelles in facilitating membrane scission. Whether inter-organelle contacts contribute to mitochondrial fusion remains poorly understood. The results presented here indicate that trans-Golgi network vesicles (TGNv) are recruited to mitochondrial fusion sites and accelerate the speed of fusion. Super resolution live cell imaging revealed that TGNv are recruited downstream of the endoplasmic reticulum (ER), suggesting that the ER might serve as a platform for TGNv recruitment to the mitochondrial fusion site. A screen examining the set of phosphatidylinositol phosphate (PIP) species as potential regulators of mitochondrial morphology uncovered the localization of phosphatidylinositol (3,4)-bisphosphate [PI(3,4)P2] in close proximity to the mitochondrial network. PI(3,4)P2 is produced by Class II phosphatidylinositol 3-kinase (PI3KIIs) α and β isoforms from phosphatidylinositol 4 phosphate (PI4P). Interestingly, we showed that loss of PI3KIIs resulted in mitochondrial morphology changes characterized by mitochondrial fragmentation. Mechanistic studies revealed that cells lacking PI3KIIs exhibited impaired mitochondrial fusion and were unable to undergo stress-induced mitochondrial hyperfusion (SIMH) in response to cycloheximide (CHX) treatment. Furthermore, super-resolution imaging showed that PI3K α is localized to TGNv and live-cell super-resolution as well as confocal imaging with photoactivatable GFP (PAGFP) revealed the presence of both PI(3,4)P2 and PI3K α at mitochondrial fusion sites, highlighting a role for TGNv-containing PI3KIIs and PI(3,4)P2 in regulating the process of mitochondrial fusion. Altogether, TGNv are recruited to mitochondrial fusion sites downstream of ER, accelerating membrane merging through the localized activity of PI3KC2α, which produces PI(3,4)P2. Thus, inter-organelle contacts are not only important for mitochondrial division but also contribute to the complex process of mitochondrial fusion, thereby promoting cellular and organismal homeostasis.