Stairway to the Golgi: Two paths VPS13B can go by

Bridge-like lipid transfer proteins (LTPs), such as VPS13 or ATG2, have generated increasing interest due to their ability to move lipids rapidly between membranes—likely >100 times faster than conventional, shuttle-like LTPs (1, 2). These LTPs are characterized by a hydrophobic groove that spans their length—which can reach ∼16 nm for VPS13— physically tethering and connecting membranes for efficient bulk lipid transfer (3) (Fig. 1 A). Among the four members of the human VPS13 family, VPS13B is mostly associated with Golgi membranes, and mutations in the VPS13B gene causes Cohen syndrome, a rare genetic disorder characterized by developmental delay, intellectual disability, joint hypermobility, myopia, and low neutrophil count (4). Now, two new studies uncover that VPS13B operates at two critical interfaces of the early secretory pathway: (a) between cis- and trans-Golgi cisternae, and (b) between endoplasmic reticulum (ER) exit sites (ERES) and the Golgi (Fig. 1 B).

In the report by Ugur et al. (5), the intracellular localization of VPS13B was explored with 3D super-resolution microscopy—providing an isotropic resolution below 20 nm—, revealing that VPS13B presents a punctate signal that best colocalizes with the cis-Golgi protein GM130. By subjecting cells to a hypotonic shock that causes membrane contact site expansion, they mapped VPS13B at the interface between cis- (proximal) and trans- (distal) Golgi cisternae. Unlike other VPS13 family members, VPS13B does not contain a functional FFAT motif necessary for ER tethering and does not localize all over the ER network, supporting the notion that VPS13B does not act at canonical, VAP-dependent ER–Golgi membrane contact sites. Next, the authors identified the protein of unknown function FAM177A1 as a Golgi-localized interactor of VPS13B. Remarkably, mutations in FAM177A1 are linked to a rare neurodevelopmental disorder that shares some similarities to Cohen syndrome (6), thereby suggesting a possible functional overlap between these two proteins. Despite their close proximity at the Golgi, neither VPS13B nor FAM177A1 are required for the other’s localization (5). Functional studies showed that both VPS13B and FAM177A1 knockout cells exhibit delayed Golgi reassembly after Brefeldin A (BFA)-induced Golgi breakdown. As the Golgi is also disassembled during mitosis, it will be interesting to investigate whether these two proteins are involved in regulating Golgi reformation after cell division, and if so, how are they regulated throughout the cell cycle. To conclude, the authors propose a working model in which VPS13B-facilitated lipid transfer between Golgi cisternae is critical for efficient Golgi reformation and homeostasis.

In a complementary study by Du et al. (7), a different interactor of VPS13B was identified: the ER exit site (ERES)-localized Sec23-interacting protein (Sec23IP, also known as p125). Structurally, Sec23IP belongs to the phospholipase A1 family of proteins (8). To examine the biochemistry of VPS13B-Sec23IP binding, the authors mapped the interaction domains between VPS13B and Sec23IP (7). They found that this interaction is disrupted in Cohen syndrome-associated mutations of VPS13B, further linking this pathway to disease mechanisms. Next, they investigated the cellular function of this interaction, showing data that supports the idea that Sec23IP recruits a pool of VPS13B to the ERES–Golgi interface, a site of active ER-to-Golgi trafficking. The loss of either VPS13B or Sec23IP disrupted the formation of a tubular ER–Golgi intermediate compartment (tERGIC), a recently proposed specialized ER–Golgi transport intermediate for the fast export of soluble clients of the SURF4 cargo receptor (9). Expression of a putative lipid transfer-inactive mutant of VPS13B was unable to restore tERGIC formation (7), suggesting that lipid transfer may be needed for tERGIC biogenesis. Finally, to investigate whether and how these proteins control membrane trafficking, Du et al. (7) performed a total secretome analysis and revealed that the secretion of a number of procollagens is partially inhibited in VPS13B-depleted cells, while the secretion of other cargo proteins is accelerated. Taken together, these data suggest a possible role for VPS13B–Sec23IP interaction in cargo sorting and/or selective cargo export at ERES. It will be interesting to compare the VPS13B-dependent secretome to the SURF4-dependent secretome (10), as both pathways appear to be related, at least in their reliance on tERGIC carriers for efficient ER export. Interestingly, overexpression of Sec23IP and VPS13B caused peripheral ERES to relocate to the perinuclear region, further strengthening the idea that VPS13B–Sec23IP interaction tethers ERES to the Golgi (7). While observing this phenomenon in live cells may help elucidate the dynamics of the process, the current data raise exciting questions about how Sec23IP-mediated recruitment influences Golgi-associated ERES and ER export.

Together, these two studies highlight the functional versatility of VPS13B, in part facilitated by its newly identified partners FAM177A1 and Sec23IP (Fig. 1 B). The proposed capacity of VPS13B to transfer lipids between membranes could play a crucial role in Golgi reassembly (5) and the formation of tERGIC structures and efficient ER-to-Golgi trafficking (7), as both processes require en masse transfer of membrane from the ER to the Golgi. But how can VPS13B mediate, from a mechanistic perspective, this membrane transfer? To address this question, the first step will be to determine if VPS13B is indeed a bona fide bridge-like LTP and measure the rate of lipid transfer. If so, the next key questions will be (a) whether VPS13B facilitates lipid transfer bidirectionally or unidirectionally, (b) which specific membranes are involved in this transfer, and (c) what mechanisms (if any) drive it.

In the absence of an active driving force dictating directionality, the establishment of a bridge for lipid transfer between two membranes with different lateral tensions may open a new mode for tension equilibration: the emergence of a tension-driven lipid flow that moves lipids from the membrane under low tension to the one under high tension (akin to a Marangoni flow in fluid mechanics) (11) (Fig. 1 C). Interestingly, in vitro measurements of the mechanical tensions of ER- and Golgi-derived membranes suggest that ER membranes are generally under higher tension than Golgi membranes (12). Following this line of thought, VPS13B-mediated lipid transfer from early Golgi cisternae to perinuclear ERES may locally reduce ERES membrane tension. Although such Golgi-to-ER lipid flow might seem counterintuitive given VPS13B’s observed role in ER-Golgi trafficking and Golgi reassembly, we propose that this reverse lipid flow is acting to transiently reduce ERES tension, thereby facilitating membrane deformation and the formation of long transport intermediates for the export of bulky cargoes, such as procollagens, as has been previously proposed (13). The overall consequence of this Golgi-to-ER lipid transfer would be to boost ER-to-Golgi trafficking, particularly of challenging cargoes such as procollagens, explaining the phenotypes observed by (a) Du et al. (7) (ER accumulation of procollagen in VPS13B-depleted cells, and defective biogenesis of tERGIC), and (b) Ugur et al. (5) (delayed Golgi reassembly after BFA washout, possibly caused by inefficient ER export). Whether the two paths VPS13B can go by (linked to FAM177A1 or Sec23IP) are also functionally related will be important to study in the future.

In conclusion, assessing if bridge-like LTPs support tension-driven lipid flow between membranes (e.g., ERES and Golgi or between proximal and distal Golgi cisternae) will crucially help us fully understand the functional organization of the early secretory pathway. New tools to report on local variations of membrane tension at these sites, combined with real-time observations of VPS13B lipid transfer activity, would aid answer these fundamental questions. Testing these ideas will not only be important for understanding the role of VPS13B in lipid transfer and organelle homeostasis but also its involvement in diseases like Cohen syndrome, offering insights for potential therapies.