The PINK1/Parkin pathway is a highly intricate regulatory mechanism, and studies have shown that both PINK1 and Parkin are situated in the MAM under normal physiological conditions as well as during cellular stress (Barazzuol et al., 2020). Following an IS, detrimental stimuli such as ROS and hypoxia facilitate the initiation of PINK1 via phosphorylation. Phosphorylated PINK1 subsequently attaches to the surface of mitochondria, leading to the binding and phosphorylation of Parkin on the outer membrane of mitochondria. Upon activation, Parkin will partially attach ubiquitin to the OMM protein, thereby triggering mitophagy. Certain mitochondrial proteins undergo ubiquitination and subsequent degradation, a process crucial for mitophagy (Tanaka et al., 2010). Other ubiquitinated proteins are recruited by autophagy adaptor proteins, such as optineurin (OPTN) and nuclear dot protein 52 (NDP52), which anchor marked mitochondria to autophagosomes via their LC3-interacting region (LIR) motif, thereby initiating mitophagy (Lazarou et al., 2015). The main site where LC3-II is locally accumulated and the membrane origin of the mitochondria has been identified as MAM (Yang and Yang, 2013). Post-ischemic reperfusion therapy enhances mitochondrial autophagy through the induction of mitochondrial ubiquitination mediated by PINK1/Parkin, thereby mitigating transient global cerebral ischemia and ameliorating neuronal injury (Wen et al., 2021). Following mitochondrial depolarization, PINK1 and Parkin are recruited to the MAM, and it seems that the translocation of PINK1 to MAM is crucial for recruiting the autophagy machinery to this region (Gelmetti et al., 2017). When mitophagy is triggered, the amount of PINK1 rises in a manner dependent on calcium, and the expression of PINK1 is stimulated while mitophagy is initiated when there is a local increase in calcium levels during the exchange of calcium between cellular organelles in MAM (Gómez-Sánchez et al., 2014). Under typical conditions, there exists a delicate equilibrium between the generation and elimination of ROS within the human body. However, this balance is disrupted in the event of an IS. ROS serves as the primary initiator of mitochondrial autophagy, with excessive ROS triggering this process. The PINK1/Parkin pathway has the capacity to impede ROS accumulation and promptly eliminate damaged mitochondria, thereby regulating IS and substantially mitigating oxidative harm (Wen et al., 2023). IS-induced neuronal injury is partly due to the excitatory toxicity of glutamate (L-Glu), and it has been observed that the PINK1/Parkin pathway safeguards neurons against L-Glu-induced damage by maintaining mitochondrial function (Dou et al., 2022). It is important to mention that the recently identified extended form of PTEN (PTEN-L) has been demonstrated to function as a suppressor of mitophagy by obstructing Parkin’s movement to mitochondria and restraining its E3 ligase activity. Some PTEN has been identified to be localized on the ER membrane and in MAM (Wang et al., 2018a). Prior research has indicated that the activation of mitophagy occurs via the PINK1/Parkin pathway in instances of brain ischemia and reperfusion-induced brain injury (Lan et al., 2018). Therefore, the regulation of mitochondrial autophagy mediated by PINK1/Parkin in MAM may represent a promising approach for the treatment of IS.

FUNDC1 functions as a newly discovered MAM protein essential for triggering mitochondrial autophagy under hypoxic conditions. Research has indicated that FUNDC1 has the ability to interact with MAM-specific proteins, facilitate the generation of MAM, and influence mitochondrial dynamics (Wu et al., 2017). Under typical oxygen levels, a minimal quantity of FUNDC1 can be found within the MAM (Wu et al., 2016a). In the absence of oxygen, FUNDC1 accumulates at the MAM through its interaction with the ER resident protein CANX (calreticulin). The regulation of FUNDC1 mitochondrial autophagy is influenced by a variety of stress factors and cellular proteins. The phosphorylation of FUNDC1 at Tyr18 and Ser13 is carried out by Src and casein kinase 2 (CK2), respectively, which serves to inhibit its interaction with LC3 and the initiation of autophagy (Chen et al., 2016). In neurons undergoing ischemic reperfusion, excessive activation of Src due to ischemic injury leads to the phosphorylation of Tyr18 on FUNDC1. This results in the impairment of FUNDC1’s ability to bind with LC3, ultimately hindering its participation in neuronal mitochondrial autophagy. Repressing Src activity can preserve the function of FUNDC1 and initiate neuroprotective effects mediated by mitochondrial autophagy through FUNDC1 (Tang et al., 2023). FUNDC1 has the ability to engage with OPA1, a mitochondrial fusion protein, as well as Drp1, a mitochondrial fission protein (Chen et al., 2016). The breakdown of OPA1 at the S1 site caused by cerebral ischemia worsens the splitting of mitochondria and reperfusion damage in neurons (Li et al., 2022). Inhibition of Drp1 may restrict brain injury following IS and avert neuronal impairment and mortality (Grohm et al., 2012). FUNDC1 may enhance mitophagy and ameliorate neuronal damage following IS through its interaction with OPA1 and Drp1. Prior research has indicated a clear link between MAM and FUNDC1 in facilitating the process of mitochondrial autophagy, FUNDC1 interacts with another MAM protein, IP3R2, to mediate IP3R2-dependent calcium signaling from the ER to the mitochondria (Wu et al., 2017), thereby enhancing calcium transport in MAM and impacting mitophagy. An increasing number of studies are demonstrating the potential for regulating FUNDC1-mediated mitochondrial autophagy to improve IS (refer to Table 2). These findings indicate that targeting FUNDC1-mediated mitochondrial autophagy in MAM may hold promise for the treatment of IS.

While Beclin 1 was initially investigated for its role in regulating autophagy, it also contributes to the regulation of mitochondrial autophagy. After stimulating mitochondrial autophagy, it was translocated to the MAM, facilitating the crosstalk between the ER and mitochondria and boosting the efficacy of mitochondrial autophagy (Gelmetti et al., 2017). Beclin1’s localization on MAM ensures that the initiation of autophagosome formation occurs in close proximity to impaired mitochondria, facilitating their efficient engulfment. Beclin 1 collaborates with different cofactors to regulate the function of the Vps-34 protein and promote the formation of the Beclin 1-Vps34-Vps15 core complex, thus initiating autophagy (Kang et al., 2011). Furthermore, Beclin1 initiates mitophagy by promoting the translocation of Parkin from the cytosol to the mitochondria. It is noteworthy that Beclin1 is significantly upregulated by ischemic preconditioning (Xie et al., 2018), whereas its expression is downregulated following reperfusion injury (Dai et al., 2017). After cerebral ischemia, the primary localization of Ulk1 is in the small glial cells within the infarcted region, where it facilitates the development of anti-inflammatory pathways in these cells (Xiong et al., 2024). Beclin1 ensures the accurate targeting of autophagosomes to mitochondria via ULK1-dependent serine phosphorylation (Quiles et al., 2023), and Beclin1 may facilitate mitochondrial autophagy, thereby promoting neuronal repair following ischemic brain injury. Mouse studies have demonstrated that Beclin1 safeguards mitochondria by preferentially triggering a distinct mitophagy pathway. The protective role of Beclin1 in this process may potentially extend to MAM involvement, possibly contributing to the formation or maintenance of MAM (Sun et al., 2021). A recent investigation has presented fresh findings indicating that Beclin1 and Beclin2 exhibit around 57% sequence similarity and possess several common structural domains. However, it was observed that Beclin2 does not contain the Ulk1 phosphorylation site found in Beclin1 and is not localized to MAM during mitophagy (Quiles et al., 2023). These results provide additional evidence that the distinctive N-terminal domain of Beclin1 and its precise positioning in MAM could account for its distinct role in mitophagy. Research has demonstrated that the downregulation of Beclin1 can attenuate cell death in the ischemic hemisphere, facilitate neurogenesis, and diminish infarct size (Zheng et al., 2009). Repressing Beclin1 can impede the formation of autophagosomes, leading to a notable decrease in neuronal loss, glial proliferation, and cell death (Xing et al., 2012). In conclusion, Beclin1-mediated mitochondrial autophagy in MAM represents a promising therapeutic target for the management of IS.

Mfn2 can be found in the MAM and serves as a prototypical functional tethering protein within this region. Its presence in the MAM allows for interaction with mitochondria, facilitating the formation of an organelle bridge within the cell (McFie et al., 2016). In the context of cerebral ischemia–reperfusion, the lack of Mfn2 hinders platelet activation, formation of prothrombotic platelets, and diminishes the extent of the infarct (Jacob et al., 2024). Mfn2 is essential for the proper functioning of mitochondria, encompassing fusion, axonal transport, inter-organelle communication, and mitophagy (Stuppia et al., 2015). Mfn2’s regulation of mitochondrial function can attenuate hypoxia-induced cell apoptosis (Peng et al., 2015). Downregulation of Mfn2 results in impaired mitochondrial function, disrupting calcium homeostasis and ultimately leading to delayed neuronal cell death. Studies have indicated a notable decline in Mfn2 levels 90 min after 6 h of middle cerebral artery occlusion. As a result, the decrease in Mfn2 is considered to occur later during reperfusion. Targeting this reduction may contribute to minimizing ischemic damage and widening the current limited stroke treatment window (Martorell-Riera et al., 2014). In order for mitophagy to take place, mitochondria need to undergo fission initially. This process involves the separation of dysfunctional mitochondria through mitochondrial fission and selective fusion, enabling their removal via mitophagy (Twig et al., 2008). Mfn2 is involved in the regulation of the structure of MAM and mitochondrial function, which is essential for protecting neurons. Decreased expression of Mfn2 results in impaired MAM function and mitochondrial activity, ultimately leading to increased fragmentation of mitochondria (Zhu et al., 2024). Ubiquitination of Mfn2 could potentially inhibit mitochondrial fusion by facilitating the breakdown of these proteins through proteasomes or by obstructing the formation of MFN dimers between mitochondria (Joaquim and Escobar-Henriques, 2020). Cerebral ischemia–reperfusion injury results from inadequate autophagy, and Mfn2 has the potential to enhance ischemia–reperfusion injury by boosting the generation of autophagosomes and facilitating their fusion with lysosomes (Peng et al., 2018). Additionally, the interaction between the ER and mitochondria occurs as an initial stage in the process of mitophagy, preceding the engulfment of cellular organelles by the autophagosome. Mfn2, a protein that tethers mitochondria to the ER, is essential for the formation of autophagosomes during mammalian starvation (Naon et al., 2016). Knockout or silencing of Mfn2 results in the separation of ER and mitochondria (de Brito and Scorrano, 2008). Studies have indicated a connection between global cerebral ischemia and the swift migration of Mfn2 from the mitochondria to the cytosol (Klacanova et al., 2019). Additionally, Mfn2 serves as a substrate for parkin, and during the initial phases of mitophagy, parkin may attach to ubiquitinated Mfn2 by interacting with the pUb domain. The ubiquitination of Mfn2’s HR1 domain by parkin is essential for efficient mitophagy (McLelland et al., 2018). Recent research has shown that Mfn2 is directly involved in inhibiting mitochondrial autophagy by connecting mitochondria to the ER (Basso et al., 2018). These studies indicate that targeting Mfn2-mediated mitochondrial autophagy in MAM may be an effective approach for regulating IS.

Glucose regulation protein 78 (GRP78) functions as an ER chaperone located within the intraluminal space and is a significant constituent of MAM (Prasad et al., 2017; Mahamed et al., 2023). The upregulation of GRP78 serves as an indicator of ER stress, regulating the activation of transmembrane ER stress sensors by means of binding and release mechanisms (Lee, 2005). GRP78 serves as the primary controller of the UPR, facilitating proper protein folding and inhibiting the accumulation of protein aggregates within the ER. UPR is controlled by three nearby detectors IRE1, PERK, and ATF6. GRP78 interacts with these three sensors via its peptide-binding domain, maintaining them in a state of inactivity. When cells accumulate misfolded proteins, they attach to GRP78 and interfere with its communication with the stress sensors located upstream. Moreover, GRP78 is a protein that is activated under stress and has broad expression in the cells of animals (Chang et al., 1987). When cells experience ER stress, particularly during the depletion of stored calcium and the buildup of irregular proteins, there is an increase in the transcription rate of GRP78 (Li et al., 1994). IS can result in the disruption of ER calcium homeostasis, damage to UPR, and impairment of proteasome function, thereby leading to secondary ER dysfunction. The sole method to break free from this potentially life-threatening pattern is to initiate UPR, thus prompting the production of GRP78 at a degree adequate for reconfiguring misshapen proteins (Paschen, 2004). The serine protease tPA is considered the standard treatment for cerebral ischemia, and prior research has identified Grp78 as a novel neural receptor for tPA. Activation of tPA by Grp78 on cell surfaces triggers a negative feedback loop that inhibits the activation of the PERK branch, leading to decreased phosphorylation of eIF2α and ultimately reducing neuronal death in cerebral ischemia (Louessard et al., 2017). Moreover, a research study indicates that GRP78 plays a role in facilitating the movement of peptides through the ER membrane and serves to safeguard cells against cell death induced by ER stress (Rao et al., 2002). An increasing number of research findings indicate the essential involvement of GRP78 in ER stress (refer to Table 5). In conclusion, GRP78 located in the mitochondrial associated membrane (MAM) is capable of regulating neuronal death induced by ER stress. Targeting GRP78 in MAM may represent a promising therapeutic approach for managing ER stress following IS (Figure 2).

The transmembrane protein IRE1, which is involved in the UPR, possesses both a kinase domain and an RNAse domain (RD) and is highly conserved across species (Nikawa and Yamashita, 1992). As part of the MAM complex, the accumulation of IRE1α in MAM leads to either cell survival or cell death through the facilitation of excessive mitochondrial calcium levels (Shuda et al., 2003). Under normal conditions, the ER resident chaperone Bip is attached to the luminal domain of IRE1α, exerting negative regulation on it. During ER stress, Bip separates from IRE1, leading to the homodimerization and self-phosphorylation of IRE1. This is followed by the activation of its RNase domain. The activated IRE1 can then cleave the XBP1 precursor mRNA to generate the active spliced form of XBP1 (sXBP1). Once inside the nucleus, the sXBP1 mRNA is translated into the mature protein, which in turn stimulates the expression of genes related to protein folding and ultimately helps alleviate ER stress (Wang et al., 2022a). The likelihood of patients surviving IS is associated with the promotion of angiogenesis, and XBP1 plays a positive role in controlling the growth and development of brain microvascular endothelial cells following brain ischemia (Shi et al., 2019). A different research indicates that the IRE1/XBP1 pathway of the UPR becomes active during IS, and the lack of Xbp1 results in deteriorating stroke consequences (Jiang et al., 2017). Hence, IRE1 may modulate the occurrence of IS through the mediation of XBP1. Nevertheless, persistent and elevated levels of self-phosphorylation result in the formation of higher-order IRE1α oligomers, leading to a relaxation in the specificity of IRE1α. This subsequently triggers the degradation of numerous mRNAs encoding secretory proteins on the ER membrane through an intramolecular process known as regulated IRE1α-dependent decay (RIDD) (Oakes, 2017). RIDD plays a role in the degradation of various mRNAs and microRNAs, thereby regulating pathological processes such as inflammation and cell apoptosis (Wang et al., 2022a). Studies have demonstrated that continuous activation of IRE1α leads to neuronal death by reducing the expression of 14-3-3θ mRNA through IRE1α RIDD activity. Furthermore, IRE1α also serves as a scaffold to regulate the localization of inositol 1,4,5-trisphosphate receptors (IP3Rs) on MAM (Carreras-Sureda et al., 2019). Following an IS, the maintenance of intracellular calcium balance is essential for regulating neuronal function, with inositol 1,4,5-trisphosphate receptors (IP3Rs) serving as crucial channels for this purpose (Kesherwani and Agrawal, 2012). During cerebral ischemia, the ATP consumption is also linked to the continuous decrease in levels of inositol 1,4,5-trisphosphate, potentially resulting in neuronal function impairment and influencing cerebral ischemic injury (Sun and Hsu, 1996). IRE1α’s regulation of IP3Rs receptors has the potential to decrease neuronal mortality and subsequently control IS. In conclusion, the involvement of IRE1 in MAM is crucial for the regulation of ER stress, suggesting that targeting IRE1 in MAM could be a promising therapeutic strategy for IS (Table 6).

The ER stress sensor PERK, a protein kinase R-like kinase, plays a crucial role in the UPR and is specifically abundant in MAM (Verfaillie et al., 2012). PERK plays a crucial role in MAM and has the ability to create a compound with S1R and Mfn2, contributing to the formation of MAM and enhancing the stability of ER-mitochondria interaction (Mao et al., 2022). One key sign of UPR activation following cerebral ischemia reperfusion involves the detection of PERK activation, which becomes fully active in the early stage of cerebral reperfusion (DeGracia and Montie, 2004). As a result of intracellular stress, unstructured or incorrectly folded proteins vie with GRP78 for binding, resulting in the dissociation of GRP78 from PERK. This leads to the release of inhibition on PERK and its activation through dimerization and self-phosphorylation (McQuiston and Diehl, 2017). Upon activation, PERK phosphorylates eukaryotic translation initiation factor 2 (eIF2α), leading to the inhibition of protein translation and a reduction in the aggregation of unfolded proteins within the ER lumen (Harding et al., 2000). PERK, a protein, is accountable for reducing mRNA translation during ER stress. This action helps to block the flow of newly produced proteins into the already stressed ER compartment. The process of translation attenuation is facilitated by the phosphorylation of eIF2α (Sano and Reed, 2013). During the early ischemia/reperfusion period, PERK significantly enhances the phosphorylation of eIF2α (Owen et al., 2005). In addition to its role in inhibiting protein translation, phosphorylated eIF2α is also capable of activating the expression of activated transcription factor 4 (ATF4) (Harding et al., 2000). ATF4 has the potential to influence cerebral ischemic injury through the regulation of cell apoptosis and neuronal function (Wu et al., 2022). Therefore, the PERK-ATF4 signaling pathway may play a role in neuronal apoptosis subsequent to IS. Apart from elF2α, PERK has the ability to phosphorylate nuclear red cell 2 p45-related factor 2 (NRF2) as well. The activation of NRF2 by PERK contributes to the maintenance of oxidation–reduction balance and promotes cell survival following ER stress (Cullinan and Diehl, 2004). During the occurrence of IS, NRF2 is involved in reducing oxidative stress, combating inflammation, regulating mitochondrial balance, and safeguarding the integrity of the blood–brain barrier to mitigate cerebral ischemic damage (Wang et al., 2022b). PERK may modulate IS through the phosphorylation of NRF2. Currently, there is a growing body of research investigating the role of PERK in the regulation of IS (refer to Table 7), and targeting PERK in MAM may hold promise as an approach for treating IS.

The sigma-1 receptor (Sig-1R) is a widely distributed co-factor found in the ER, specifically located in the MAM of the central nervous system. It plays a role in mediating signal transduction between the ER and mitochondria. Under normal conditions, Sig-1R is in a dormant state in MAM and forms a complex with another ER partner, BiP. During periods of ER stress, Sig-1R separates from BiP and controls the function of the three primary pathways of the UPR (PERK, IRE1a, ATF6) (Penke et al., 2018). The expression of Sig-1R is elevated in response to ER stress (Mitsuda et al., 2011). Overexpression of Sig-1R suppresses the activation of PERK and ATF6 signals, leading to enhanced cell survival (Hall et al., 2009). The decreased activity of Sig-1R destabilizes the conformation of IRE1a and diminishes cell viability (Mori et al., 2013). Agonists targeting Sig-1R have demonstrated protective effects on cells during stroke, and these effects are associated with the alleviation of ER stress (Morihara et al., 2018). It is evident that the upregulation of Sig-1R serves as a defensive mechanism for cell survival in response to ER stress. Moreover, Sig-1R has the capability to engage with inositol 1,4,5-trisphosphate receptor 3 (IP3R3) in order to stabilize its association with voltage-dependent anion channel 1 (VDAC1), thereby ensuring appropriate calcium transfer between the two cellular organelles (Morihara et al., 2018). After an IS, the downregulation of VDAC1 expression leads to intracellular calcium overload. Stabilizing the expression of VDAC1 can alleviate neuronal loss following cerebral ischemia (Yao et al., 2018). Sig-1R may stabilize VDAC1 through its interaction with IP3R3, thereby modulating the regulation of IS. It is important to mention that Sig-1R shows significant expression in the central nervous system, leading to its crucial involvement in functions such as cell differentiation, synapse growth, and the stimulation of microglia cells (Wu N.H. et al., 2021). The polarization of microglia/macrophages is a critical factor in the damage to tissues and the recovery of function following an IS. Additionally, Sig-1R serves as a crucial regulatory element in the macrophage-mediated clearance of deceased cells. The complete transplantation of Sig-1R macrophages has been shown to significantly decrease tissue damage and neurological impairments associated with IS (Zhang et al., 2023). In conclusion, targeting Sig-1R in MAM shows great promise for the treatment of IS (Table 8).