Mitochondrial Dynamics, Mitophagy, and Mitochondria-Endoplasmic Reticulum Contact Sites Crosstalk Under Hypoxia

Abstract

Mitochondria are double membrane organelles within eukaryotic cells, which act as cellular power houses, depending on the continuous availability of oxygen. Nevertheless, under hypoxia, metabolic disorders disturb the steady-state of mitochondrial network, which leads to dysfunction of mitochondria, producing a large amount of reactive oxygen species that cause further damage to cells. Compelling evidence suggests that the dysfunction of mitochondria under hypoxia is linked to a wide spectrum of human diseases, including obstructive sleep apnea, diabetes, cancer and cardiovascular disorders. The functional dichotomy of mitochondria instructs the necessity of a quality-control mechanism to ensure a requisite number of functional mitochondria that are present to fit cell needs. Mitochondrial dynamics plays a central role in monitoring the condition of mitochondrial quality. The fission-fusion cycle is regulated to attain a dynamic equilibrium under normal conditions, however, it is disrupted under hypoxia, resulting in mitochondrial fission and selective removal of impaired mitochondria by mitophagy. Current researches suggest that the molecular machinery underlying these well-orchestrated processes are coordinated at mitochondria-endoplasmic reticulum contact sites. Here, we establish a holistic understanding of how mitochondrial dynamics and mitophagy are regulated at mitochondria-endoplasmic reticulum contact sites under hypoxia.

Keywords: hypoxia; mitochondria; mitochondria-endoplasmic reticulum contact sites; mitochondrial dynamics; mitophagy.

FIGURE 1

Schematic diagrams of mitochondrial fission-fusion cycles. Mitochondria constantly undergo fission and fusion cycles to maintain a characteristic morphology in dynamic balance, allowing the mitochondrial network to adapt to cellular demands and immediately respond to imposed stresses. (A) Mitochondrial fission is mainly executed by DRP1, which is phosphorylated on Ser637, keeping it in the cytosol. When fission occurs, the cytosolic DRP1 is dephosphorylated by calcineurin and recruited to the OMM through interactions with the corresponding receptors, including MFF, FIS1 and MiD49/51, and then DRP1 self-assembles to form a ringlike structure and induces GTP hydrolysis to mediate membrane constriction. (B) Mitochondrial fusion begins with the OMM fusion, which is controlled by the MFNs. The MFNs is inserted into the OMM via two TM domains, separated by a short loop, exposing the HR2 domain and the GTPase with HR1 domain, both of which face the cytoplasm. The specific trans-structure underlies MFNs-mediated OMM fusion. (C) Following OMM fusion, the interaction between L-OPA1 and CL drives the fusion of IMM. The interaction is enhanced in the presence of S-OPA1, which is generated by proteolytic cleavage of L-OPA1.

FIGURE 2

Schematic diagrams of the structural elements of fusion proteins. (A) The MFNs contains two TM domains, separated by a short loop, exposing the N-terminal region harbouring the GTPase and HR1 domain, and the C-terminal containing the HR2 domain. (B) The structure of unprocessed OPA1 includes the N-terminal MTS, TM domain, GTPase domain, middle stalk region, and the C-terminus containing the GED. OPA1 can be proteolytically processed by YME1L or OMA1 at S1 or S2, respectively, leading to the accumulation of L‐OPA1 or S‐OPA1.

FIGURE 3

Schematic diagrams of OPA1-dependent IMM fusion. The process of mitochondrial IMM fusion includes several sequential steps: tethering, membrane docking and fusing. (A) It exhibits efficient and fast fusion at equimolar levels of S-OPA1 and L-OPA1. (B) When S-OPA1 is overexpressed, the activity of fusion is inhibited and the fusion efficiency is reduced. (C) When the S-OPA1 level is lower than L-OPA1, slow and inefficient fusion occurs. (D) The activity of the IMM fusion peaks at a ratio of 1:1S-OPA1: L-OPA1.

FIGURE 4

Schematic diagrams of the regulation of mitochondrial dynamics equilibrium and its connection to mitochondrial morphology. Under steady-state conditions, the fission-fusion cycle is constantly maintained in an appropriate equilibrium, bringing about characteristic, filamentous or tubular mitochondrial morphologies. This dynamic balance can be affected by many factors. The dynamic actin cycle plays an important role in the regulation of mitochondrial dynamics equilibrium. Actin assembly induces mitochondrial fission, whereas actin disassembly promotes fusion. Moreover, S-OPA1 and the corresponding receptors of DRP1 are involved in the regulation of mitochondrial dynamics balance. The overexpression of MFF/FIS1/S-OPA1 can tip the balance toward fission, similarly, low-to-moderate levels of MIDs prefer to shift the balance toward fission, leading to a fragmented mitochondrial network. In contrast, the overexpression of MIDs or the low-to-moderate level of S-OPA1, as well as the inhibition of MFF, can swing the dynamic equilibrium in the direction of fusion, resulting in a hyperfused mitochondrial morphology.

FIGURE 5

Schematic diagrams of mitochondrial fission at MERCSs. Multiple constriction events of the inner and outer mitochondrial membranes occur at MERCSs. (A,B) The pre-constriction events include INF2-mediated actin polymerization and CoMIC, both of which occur at MERCSs, contributing to the IMM constriction. INF2 and spire1C induce actin polymerization that increases ER-mitochondria interactions, thereby stimulating the ER-to-mitochondrial calcium transfer through the MCU. The subsequent increase in intra-mitochondrial Ca²⁺ initiates CoMIC. The elevated mitochondrial Ca²⁺ flux triggers mitoBK_Ca-mediated mitochondrial bulging and depolarization during CoMIC. OPA1 collaboratively regulates CoMIC through proteolytic processing by stabilized OMA1, resulting in the accumulation of S-Opa1, which disrupts the stability of Mic60-mediated OMM–IMM tethering. (C) The actin−NMII-dependent constriction provides the pre-constricted site at the OMM. NMII filaments are present near the constriction site along with actin filaments, both of which ensure the constriction of the actin cable, exerting pressure on mitochondria, bringing about an indentation on the surface of the OMM. DRP1 then assemblies at these pre-constricted sites by binding to its receptors, forming a ringlike structure to further constrict the membrane. (D,E) Finally, DNM2 cooperates with DRP1 to drive complete fission before disassembling of the fission machinery.

FIGURE 6

Schematic diagrams of mitochondrial fusion at MERCSs. MERCSs define the position of the OMM and IMM fusion, both of which are coordinated at the nodes of MERCSs. Mitochondrial fusion consists of five distinct steps. (A) OMM tethering: MFNs tether the OMM of two opposing mitochondria through their interactions in trans of the HR2 and/or GTPase domains. (B) OMM docking: GTP hydrolysis induces conformational changes in MFNs, which trigger docking of the OMM. Subsequently, the GTPase-dependent MFNs oligomerization completes fusion of the OMM. (C) L-OPA1-CL tethering: the IMM fusion begins with the interaction between L-OPA1 and CL. (D,E) IMM fusion: the L-OPA1-CL interaction is enhanced by S-OPA1, and the OPA1-dependent GTP hydrolysis ensures fusion of two IMMs.

FIGURE 7

Schematic diagrams of the molecular mechanism of mitochondrial dynamics, mitophagy, and MERCSs crosstalk under hypoxia. FUNDC1 acts as a key point in the molecular machinery that regulates the close link between mitochondrial dynamics and mitophagy at MERCSs under hypoxia. (A) FUNDC1 is an outer membrane protein of mitochondria. Under normal conditions, only a small amount of FUNDC1 is located on MERCSs, phosphorylated by CK2 or SRC kinase, interacting with OPA1 and indirectly binding to CNX via an unknown protein. MARCH5 is a mitochondrial E3 ubiquitin ligase that localizes on the OMM, usually forming dimers or oligomers. (B) At the initial stage of hypoxia, the oligomer of MARCH5 disassembles, and it directly binds to FUNDC1, mediating the ubiquitin-proteasome proteolytic pathway that degrades FUNDC1. Meanwhile, the inhibitory phosphorylation by SRC and CK2 kinase suppresses the activity of FUNDC1, both of which hinder FUNDC1-mediated mitochondrial fission and mitophagy, thereby avoiding inappropriate clearance of undamaged mitochondria. (C) As hypoxia progresses, activities of SRC and CK2 kinase are inhibited, while PGAM5 phosphatase is activated, leading to dephosphorylation of FUNDC1 dissociated from OPA1. Moreover, USP19 significantly accumulates at MERCSs and strongly binds to FUNDC1, which is deubiquitinated and stabilized at MERCSs, where FUNDC1 indirectly interacts with CNX. The FUNDC1-CNX association contributes to the formation of a close connection between the ER and mitochondria. (D) Subsequently, FUNDC1 dissociates from CNX while interacting with DRP1, and USP19-mediated FUNDC1 stabilization promotes DRP1 oligomerization at MERCSs, leading to hypoxia-induced mitochondrial fission. (E) At the late stage of hypoxia, FUNDC1 binds to LC3 and forms an mitophagosome, which engulfs and selectively removes impaired mitochondria. Meanwhile, the upstream ULK1 complex phosphorylates FUNDC1 to promote the process of mitophagy.