Молекулярные модели, связанные с ущербом от митохондриала: нейроиммуномодуляторы в патофизиологии центральной нервной системы.

Introduction

Neuroinflammation plays a central role in the development and progression of several neuropathologies including Alzheimer’s disease (AD), Parkinson’s disease (PD), Huntington’s disease (HD), and multiple sclerosis (MS). It is primarily mediated by the persistent pro-inflammatory activity of the central nervous system (CNS) glial cells, microglia, and astrocytes. Microglia are the resident immune cells of the CNS, closely resembling peripheral macrophages in their function of immune surveillance, defense, and clearing debris by phagocytosis. Microglia can become reactive with disturbances in brain homeostasis, resulting in a shift in their morphology and the release of pro-inflammatory cytokines and cytotoxins that can cause neuronal death (reviewed by Guo et al., 2022). Astrocytes are another type of glia and are the most abundant CNS cell type. They primarily regulate the function and maturation of neurons, but also demonstrate phagocytic debris engulfment and can be activated to produce inflammatory cytokines and neurotoxic molecules (reviewed by Onyango et al., 2021).

Glial cells become reactive upon detection of certain endogenous intracellular molecules that can appear in the extracellular space. These immune mediators, known as damage-associated molecular patterns (DAMPs), are released by damaged or dying cells and act as ligands for specific pattern recognition receptors (PRRs) expressed by glia. Activation of PRRs in most cases leads to the onset of acute neuroinflammatory cascades or sustained, chronic neuroinflammation. For example, toll-like receptors (TLRs) represent a well-studied family of PRRs, which induce intracellular pro-inflammatory signaling cascades leading to the activation of nuclear factor (NF)-κB and upregulation of the components of the nucleotide-binding domain, leucine-rich–containing family, pyrin domain-containing-3 (NLRP3) inflammasome (Li and Wu, 2021). A large proportion of DAMPs reside within mitochondria and can be released into the cytosol by mitochondrial injury or as a part of regulated cell death and subsequently into the extracellular space due to membrane permeabilization (reviewed by Marchi et al., 2023). The endosymbiotic theory of mitochondrial origin, supported by the many shared features between bacteria and mitochondria, postulates that this organelle originates from an α-proteobacteria (reviewed by Bajwa et al., 2019). As a result, several distinct mitochondrial molecules act as DAMPs once they enter the extracellular space. This review considers the following mitochondrial DAMPs (mtDAMPs): heme or hemin, cytochrome c (CytC), cardiolipin (CL), adenosine triphosphate (ATP), mitochondrial DNA (mtDNA), mitochondrial transcription factor A (TFAM), N-formyl peptides (NFPs), succinate, fumarate, and itaconate. Table 1 illustrates the structural diversity of the above mtDAMPs, which are physiologically located within different compartments of mitochondria ( Figure 1 ) and contribute to specific functions of this organelle, for example, the tricarboxylic acid (TCA) cycle ( Figure 2 ). In this review, we first summarize our current understanding of the main physiological activities of the above mitochondrial molecules, followed by descriptions of their pro- and anti-inflammatory effects as mtDAMPs in both the periphery and the CNS. We demonstrate that the receptors these molecules interact with, the signaling pathways they activate, and their roles in chronic inflammation have been studied extensively in peripheral immune cells and tissues. We also assess the need for targeted research on the effects of mtDAMPs in the CNS where their pathophysiological roles are understudied. For example, mitochondrial dysfunction and cell death are common features of neurodegenerative diseases, suggesting that mtDAMPs, by acting as pro-inflammatory stimulants of glial cells, may contribute to the persistent neuroinflammation observed in these disorders. Research that focuses on CNS functions of mtDAMPs could identify novel treatment strategies for multiple neurodegenerative diseases, which currently lack effective therapies.

Table 1 Biochemical characteristics of mitochondrial DAMPs Molecule Alternative name Structure/Nature Location in functioning mitochondria Typical activity Heme Heme a, b, c Iron protoporphyrin IX Ferroprotoporphyrin Protoporphyrin molecule bound to a ferrous iron (Fe 2+ ) atom Synthesized in mitochondrial matrix Prosthetic group for hemoproteins, reversibly binds to oxygen Hemin Free heme Protoporphyrin molecule bound to a ferric iron (Fe 3+ ) atom Primarily in extracellular space Degraded by heme oxygenase CytC N/A 12 kDa protein containing a heme c prosthetic group Intermembrane space Facilitates exchange of electrons between complex III and IV CL 1,3-bis(sn-3'-phosphatidyl)-sn-glycerol Phospholipid dimer with a glycerol molecule linking two phosphatidyl moieties bound to fatty acyl chains Mitochondrial outer and inner membranes Provides membrane stability and structure ATP N/A Contains an adenine base, ribose sugar, and three phosphate groups All mitochondrial compartments Provides energy for metabolic and other cellular processes mtDNA N/A Circular double-stranded DNA molecules ranging from 15.7 to 19.5 kb Mitochondrial matrix Encodes essential mitochondrial proteins TFAM Mitochondrial transcription factor A 25 kDa protein belonging to the high mobility group family Synthesized in nucleus, located in mitochondrial matrix Regulates mtDNA activity and packaging NFPs Formylated peptides Amino acid chains of various lengths containing a fMET molecule Mitochondrial matrix By-product of mitochondrial protein translation Succinate N/A Dicarboxylic acid derivative Mitochondrial matrix TCA cycle metabolite, substrate for SDH Fumarate N/A Dicarboxylic acid derivative Mitochondrial matrix TCA cycle metabolite, product of succinate oxidation Itaconate N/A Dicarboxylic acid derivative Mitochondrial matrix Competitive inhibitor of SDH in the TCA cycle ATP: Adenosine triphosphate; CL: cardiolipin; CytC: cytochrome c; fMET: N-formylated methionine; mtDNA: mitochondrial DNA; NFPs: N-formyl peptides; SDH: succinate dehydrogenase; TCA: tricarboxylic acid; TFAM: mitochondrial transcription factor A.

Biochemical characteristics of mitochondrial DAMPs

ATP: Adenosine triphosphate; CL: cardiolipin; CytC: cytochrome c; fMET: N-formylated methionine; mtDNA: mitochondrial DNA; NFPs: N-formyl peptides; SDH: succinate dehydrogenase; TCA: tricarboxylic acid; TFAM: mitochondrial transcription factor A.

Mitochondrial damage-associated molecular patterns are linked to specific physiological functions of the organelle.

(1) Heme is synthesized from protoporphyrin and Fe 2+ in the mitochondria of all cells. (2) ATP is generated in mitochondria and acts as the primary biochemical fuel in cellular processes. (3) mtDNA encodes essential proteins needed for oxidative phosphorylation and peptide assembly. (4) TFAM is the main transcription factor controlling mtDNA packaging and gene expression. (5) NFPs are products of mitochondrial peptide synthesis. Created with BioRender.com. ADP: Adenosine diphosphate; ATP: adenosine triphosphate; IMM: inner mitochondrial membrane; mRNA: messenger RNA; NFPs: N-formyl peptides; OMM: outer mitochondrial membrane; Pi: inorganic phosphate; rRNA: ribosomal RNA; TFAM: mitochondrial transcription factor A; tRNA: transfer RNA.

Some mitochondrial damage-associated molecular patterns are essential for ETC activity, while others are directly associated with TCA cycle.

(1) CL provides structure to cristae and stabilizes the ETC. (2) CytC facilitates the efficient exchange of electrons between complex III and IV. (3) Succinate is a metabolite in the TCA cycle that is required for adenosine triphosphate production in the ETC. (4) Fumarate is also a TCA cycle metabolite and the direct product of succinate oxidation by succinate dehydrogenase. (5) Itaconate is a byproduct of the TCA cycle that inhibits the activity of succinate dehydrogenase, preventing the oxidation of succinate into fumarate. Created with BioRender.com. CL: Cardiolipin; CytC: cytochrome c; ETC: electron transport chain; IMM: inner mitochondrial membrane; TCA: tricarboxylic acid.

Search Strategy

Articles published between 2000 and 2024 are cited in this narrative review. The initial search was performed within OVID Medline (PubMed) databases using the following keywords: “damage-associated molecular pattern,” “DAMP,” “alarmin,” “mitochondria,” “mitochondrion.” Subsequently, the names of individual mtDAMPs (“heme or hemin,” “cytochrome,” “cardiolipin,” “adenosine triphosphate or ATP,” “mitochondrial DNA or mtDNA,” “mitochondrial transcription factor A or TFAM,” “N-formyl peptides or NFP,” “succinate,” “fumarate,” and “itaconate”) were used as inputs in combination with “brain,” “nervous system,” central nervous system,” “microglia,” “astrocyte,” “neuron,” “Alzheimer’s disease,” “Parkinson’s disease,” “Huntington’s disease,” and “multiple sclerosis.” The abstracts of the retrieved research articles were screened and full versions of pertinent articles were acquired. Additional applicable articles were identified by reviewing the references cited in the selected publications.

Heme (hemin)

Role in metabolic processes and hemoprotein activity

Heme is an essential iron-containing molecule that serves as a prosthetic group for all hemoproteins, including hemoglobins, myoglobins, cytochromes, catalases, and peroxidases. Heme biosynthesis, in which a protoporphyrin molecule and a ferrous iron atom (Fe 2+ ) are bound, is a highly conserved and regulated pathway occurring in the mitochondria of all cells ( Figure 1 ; reviewed by Yien and Perfetto, 2022). However, this process is mainly performed by bone marrow erythrocytes and liver hepatocytes in mammals. Multiple forms of heme are used by different hemoproteins to reversibly bind oxygen (Li et al., 2011). Heme b , which binds non-covalently to hemoproteins, is the most common and serves as the precursor to all other heme forms, acting as the prosthetic group for several prevalent hemoproteins (such as hemoglobins, myoglobins, and peroxidases). When oxidized, heme b transforms into protein-free hemin, also known as free heme, which contains a ferric iron atom (Fe 3+ ) that binds to halide ions such as chloride. Heme c , the second most common form of heme in nature, is found in CytC. It differs from heme b in that it covalently binds to protein residues within the heme binding site (reviewed by Ow et al., 2008; Phillips, 2019). Hemin is primarily degraded by heme oxygenases (HOs), of which multiple isoforms are highly expressed in immune cells of both the periphery and the CNS (reviewed by Che et al., 2021; Si and Wang, 2020). The HO-1 isoform is inducible, and its expression is significantly upregulated in response to noxious stimuli such as reactive oxygen species (ROS) and extracellular hemin itself (reviewed by Gáll et al., 2019). HO-1 degrades hemin into free ferrous iron (Fe 2+ ), carbon monoxide, and biliverdin, which is further reduced to bilirubin by biliverdin reductase. Carbon monoxide and bilirubin possess anti-inflammatory and antioxidant capabilities (reviewed by Gáll et al., 2019). These products of hemin degradation are Janus-faced, meaning at high concentrations they become toxic and cause mitochondrial damage or ferroptosis, a type of iron-dependent programmed cell death (Choi and Kim, 2022). Therefore, HO-1 and hemin play an important role in the immune responses and regulation of cell fate in peripheral tissues as well as the CNS.

Hemin intake

Peripheral immune cells can internalize extracellular hemin using multiple transporter proteins (reviewed by Donegan et al., 2019). Feline leukemia virus subgroup C cellular receptor 2 transports hemin across the plasma membrane into the cytosol in mammalian peripheral cells, including human muscle cells and Chinese hamster ovary cells (Duffy et al., 2010). Additionally, primary murine peripheral macrophages express and utilize the heme responsive gene 1 transporter, which is located within phagolysosomal membranes and traffics hemin into the cytosol (White et al., 2013). With regards to CNS cells, cultured murine glia have been shown to scavenge extracellular hemin, with microglia being especially adept at retaining the internalized hemin after in vitro washout (Chen-Roetling et al., 2014). Heme carrier protein 1, a known hemin transporter protein expressed by primary murine astrocytes, may contribute to this uptake of hemin by glial cells (Dang et al., 2010). Notably, glial cells in the human brain express Feline leukemia virus subgroup C cellular receptor 2 (Duffy et al., 2010).

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

Hemoglobin released from lysed erythrocytes can rapidly oxidize, increasing the concentration of hemin in serum and other extracellular compartments of the periphery (reviewed by Gáll et al., 2019). Extracellular hemin may also originate from other hemoproteins, such as cytochromes, catalases, and peroxidases found in most peripheral and CNS cells. The effects of extracellular hemin in peripheral tissues are widely studied, as hemin is considered a DAMP due to its immunostimulatory effects (reviewed by Bozza and Jeney, 2020). Hemin is an agonist of multiple PRRs, with TLR4 being the most well defined. Activation of TLR4 by hemin increases the secretion of cytokines such as tumor necrosis factor (TNF) by primary murine peripheral macrophages and endothelial cells (Figueiredo et al., 2007; Belcher et al., 2014). The NLRP3 inflammasome can be activated in primary murine peritoneal macrophages by extracellular hemin, increasing their interleukin (IL)-1β secretion (Li et al., 2014). Hemin also induces primary murine neutrophil migration and ROS production by engaging protein kinase C and phosphatidylinositol 3-kinase (PI3K) signaling (Porto et al., 2007). Overall, the DAMP-like activity of hemin and the receptors it engages in the periphery are well-defined in the current literature (reviewed by Mendonça et al., 2016; Bozza and Jeney, 2020).

The CNS effects of hemin are primarily studied in neuroimmune processes associated with intracerebral hemorrhages (ICH). Extracellular hemin at high concentrations (> 10 μM), which is used to model ICH in vitro , has been shown to activate both microglia and astrocytes. At these high concentrations, it also significantly decreases the viability of primary murine microglia, astrocytes, and neurons in vitro , with microglia being the most vulnerable (Cai et al., 2011). BV-2 murine microglia treated with hemin produce significantly more ROS than untreated cells, contributing to oxidative damage and decreased cell viability (Hu et al., 2020). This is supported by another study in which the transfer of conditioned medium from hemin-treated HMC3 human microglia onto human SH-SY5Y neuroblastoma cells increases their ROS production (Jia et al., 2022). At high concentrations (> 10 μM), hemin activates multiple well-studied pro-inflammatory pathways. It potentiates TLR4 pro-inflammatory signaling, engaging the myeloid differentiation factor (MyD)88 and toll/interleukin-1 receptor-domain-containing adapter-inducing interferon (IFN)-β (TRIF) pathway. This causes NF-κB-dependent secretion of inflammatory cytokines including TNF and IL-1β by primary murine microglia and astrocytes (Lin et al., 2012). Hemin also acts as an agonist of TLR2, upregulating the activity of matrix metalloproteinase-9 in cultured primary murine astrocytes and increasing the permeability of the murine blood–brain barrier (BBB) in vivo (Min et al., 2017). Additionally, treatment of SIM-A9 murine microglia with hemin raises the production of nitric oxide (NO), a well-known pro-inflammatory molecule, in a dose-dependent manner (Sayeed et al., 2017).

A previous study focuses on the effects of lower concentrations of hemin (< 10 μM) on microglia and astrocytes, which are hypothesized to be more representative of the conditions seen in neurodegenerative diseases such as AD. Hemin concentrations as low as 3 μM are still sufficient to significantly decrease the viability of primary murine astrocytes and neurons in vitro (De Caro et al., 2023). Notably, hemin may display some protective activity in the CNS when it binds to the AD-associated amyloid-β (Aβ), reducing its pro-inflammatory activity. However, primary murine astrocytes and SIM-A9 murine microglia treated with 0.05 μM hemin demonstrate downregulated phagocytic activity, leading to decreased clearance of Aβ in the extracellular space. This is due to the hemin-induced activation of the PI3K/Akt pathway, which modulates inflammation and phagocytosis (Sankar et al., 2018). This lowered phagocytic activity is accompanied by a reduced expression of select cytokines in hemin- and Aβ-stimulated primary murine astrocytes, including IL-1β, C–C motif chemokine ligand (CCL)5, and granulocyte-macrophage colony-stimulating factor (Sankar et al., 2018). The inhibitory effects on cytokine secretion contradict the canonical activity of hemin as a pro-inflammatory mediator and complicate the interpretation of its role in AD-associated neuroinflammation, indicating the need for further investigation of the DAMP-like activity of hemin in the CNS.

Association with brain pathologies

Various conditions, including hemolytic anemias, ischemia/reperfusion, and hemorrhages, cause the release of free hemin from erythrocytes into the extracellular space, leading to oxidative damage and inflammation (reviewed by Bozza and Jeney, 2020). In the CNS, this is often caused by ICH, resulting in the release of hemoglobin, hemin, and iron from lysed erythrocytes into the brain parenchyma. This initiates an inflammatory response by both microglia and astrocytes and disrupts the BBB (reviewed by Puy et al., 2023). Similarly, in the AD brains, the total content of heme b is increased 2.5 fold compared to the control brains (Atamna and Frey, 2004). This likely results from a weakened BBB, a hallmark of AD, that facilitates the infiltration of plasma into the brain (Sankar et al., 2018). The levels of extracellular hemin in the AD brain are thought to be much lower than those observed after an ICH, but the exact pathophysiological range is not known. It is, however, hypothesized to correlate with the concentration of Aβ, which is supported by reports that Aβ deposits in AD brains also contain high concentrations of hemin. Hemin-Aβ complexes may be more harmful than Aβ deposits since they act as peroxidases by oxidizing and disrupting the activity of essential biomolecules (reviewed by Puy et al., 2023; Sankar et al., 2018).

Conclusions

Extracellular hemin acts as a DAMP-like molecule in the CNS, but the complexity of its interactions with other pathologically relevant molecules, such as Aβ, necessitates more detailed studies. Hemin likely contributes to the development and pathogenesis of neurodegenerative diseases, particularly AD. However, the reported effects of AD-relevant concentrations of extracellular hemin on CNS immune cells contradict its more established pro-inflammatory role, indicating the need for more research to clarify its contribution to AD-related inflammation. While some hemin transporters have already been identified, elucidating the exact mechanism of trafficking extracellular hemin into CNS cells is necessary, as well as improving our understanding of how glial cells remove hemin from the brain parenchyma. Additionally, the typical serum and brain parenchyma levels of hemin in AD are unknown, and determining these concentration ranges would facilitate more effective research into the role of hemin in AD-related inflammatory signaling. Given the role of hemin in essential physiological processes, treatment options aimed at modulating its distribution and tissue levels must be precisely targeted to be effective and safe. Future research should address these critical knowledge gaps to deepen our understanding of the role of hemin in neuroinflammation, enabling the discovery of novel therapies.

Cytochrome c

Role in metabolic processes and apoptosis

CytC is a highly conserved nuclear-encoded 12 kDa protein, best known as an essential electron carrier in the mitochondrial electron transport chain. CytC possesses a covalently bound iron-containing heme c prosthetic group, which facilitates the efficient exchange of electrons between complex III and IV (reviewed by Ow et al., 2008; Bajwa et al., 2019). Under physiological conditions, this hemoprotein is primarily localized to the mitochondrial intermembrane space and typically bound to inner mitochondrial membrane (IMM) lipids, predominantly CL. However, when apoptotic signals such as oxidative stress or damaged DNA are detected by the cell, CytC is translocated from mitochondria to the cytosol where it can activate intrinsic caspase-dependent apoptotic mechanisms. The peroxidase activity of CytC causes CL oxidization, leading to mitochondrial outer membrane permeabilization (MOMP) and the release of CytC and other pro-apoptotic factors into the cytosol. Released CytC then binds to and engages apoptotic protease-activating factor 1 to form the apoptosome, initiating programmed cell death (reviewed by Ow et al., 2008; Bajwa et al., 2019). Anti-apoptotic molecules can counteract CytC-mediated cell death, either by preventing CytC translocation to the cytosol in the case of B-cell lymphoma protein-(BCL)2 family members (reviewed by Qian et al., 2022), or by interfering with CytC-caspase binding, a functional characteristic of X-linked inhibitor of apoptosis protein (reviewed by Hanifeh and Ataei, 2022). Thus, CytC is a crucial protein in the initiation of programmed cell death and is the target of specific regulatory proteins controlling this pathway.

Cytosolic activity

The CytC-dependent apoptotic pathways can be activated in all cell types, including CNS cells. This mechanism is rapid and irreversible once a threshold level of cytosolic CytC is reached, which can vary between different cells. For example, up to 15% of mitochondria in CSM14.1 rat nigral neural cells can be damaged without activating apoptosis (Khodjakov et al., 2004). Several pathologically relevant molecules have demonstrated pro-apoptotic activity in neurons and glia. For example, the mutant huntingtin protein, responsible for HD-associated neurodegeneration, disrupts intracellular Ca 2+ homeostasis in medium spiny neurons of an HD mouse model, triggering MOMP and CytC-mediated apoptosis (reviewed by Czeredys, 2020). Treatment of primary rat cortical neurons with bilirubin causes the release of CytC into the cytosol by MOMP and the subsequent downstream activation of caspase-3 (Rodrigues et al., 2002). The activation of N13 murine microglia by Aβ, the plaque-forming peptide in AD, induces CytC release into the cytosol of these cells (Chiozzi et al., 2019). Stimulants commonly used to trigger glial immune responses, such as lipopolysaccharide (LPS) and IFN-γ, also increase the levels of cytosolic CytC in murine N9 microglia (Mayo and Stein, 2007). These findings underscore the significance of intracellular CytC-mediated cell death not only in neurons but also in microglia, particularly in response to pro-inflammatory molecules.

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

Cellular damage or death, including that induced by high levels of cytosolic CytC, ultimately leads to the release of CytC into the extracellular space. In the periphery, CytC is recognized as a DAMP and has been suggested to be a biomarker that can potentially be used to diagnose pathologies characterized by increased cell death, such as myocardial infarction, liver disease, pancreatitis, and sepsis (Eleftheriadis et al., 2016). The injection of CytC into the knee joint of mice leads to NF-кB-mediated activation of neutrophils and monocytes, causing arthritis (Pullerits et al., 2005). These data strongly support the role of extracellular CytC as a DAMP in peripheral inflammation, although more information regarding the specific receptors that recognize CytC would expand the understanding of its function in the periphery.

Extracellular release of CytC has been shown in several different CNS cell types, indicating this protein can act as a DAMP in neuroimmune responses. Increased concentrations of CytC are detected in cell culture medium after the treatment of primary human astrocytes (Wenzel et al., 2019) and primary chick neuronal cells (Ahlemeyer et al., 2002) with staurosporine, a known activator of the intrinsic apoptotic pathway. Likewise, BV-2 murine microglia cells treated with staurosporine, Aβ, or TNF release CytC into the cell supernatant (Wenzel et al., 2019). Once released, extracellular CytC may contribute to glia-mediated neuroinflammation. For example, CytC increases the secretion of NO by BV-2 microglia. Additionally, CytC acts as a priming agent for N-formyl-methionyl-leucyl-phenylalanine (fMLP)-induced ROS production in differentiated HL-60 promyelocytic cells used as human microglia models (Gouveia et al., 2017). CytC also decreases the viability of both murine and human microglia model cells and enhances the cytotoxicity of microglia towards SH-SY5Y human neuroblastoma cells. These effects are further amplified when CytC is combined with other immune stimulants such as LPS and IFN-γ, suggesting a synergistic relationship in microglial activation. It has also been demonstrated that this hemoprotein at least partially engages TLR4 and c-Jun N-terminal kinases (JNK) signaling in microglia (Gouveia et al., 2017). Similar stimulatory activity of CytC is observed in primary human astrocytes, where it significantly enhances their cytotoxicity towards neuronal cells and upregulates the secretion of granulocyte-macrophage colony-stimulating factor and IL-1β in a TLR4-dependent manner (Wenzel et al., 2019). The latter observation may indicate an assembly of astrocytic inflammasomes in response to CytC stimulation. The most extensively studied inflammasome is NLRP3, which is known to have a crucial role in AD (reviewed by Chiarini et al., 2020). Other mitochondria-derived DAMPs, such as ATP (Zhong et al., 2018) and heme (Li et al., 2014), are already known to act as NLRP3 activators. It can be concluded that extracellular CytC exhibits DAMP-like activity on glial cells, increasing their production and secretion of pro-inflammatory molecules that may contribute to glia-mediated neuronal death. Extracellular CytC-induced activation of glial cells may lead to further release of CytC from dying neurons, exacerbating chronic neuroinflammation (Ahlemeyer et al., 2002; Wenzel et al., 2019).

Association with brain pathologies

CytC activity in the intracellular and extracellular space is associated with several CNS pathologies. Mitochondrial dysfunction is known to cause the release of CytC from this organelle, triggering programmed cell death. Substantial evidence supports the hypothesis that this adverse event, a common trait observed in nearly every neurodegenerative disease, plays a causative role in their development (reviewed by Klemmensen et al., 2024). In PD, damaged mitochondria in dopaminergic neurons release increasing amounts of CytC into the cytosol (Kumar et al., 2016). Oligomeric conformations of the presynaptic protein α-synuclein are implicated in PD-associated neuronal death by inducing mitochondrial fragmentation, increasing oxidative stress, and forming Lewy bodies, which are abnormal masses of protein that accumulate in neurons (reviewed by Calabresi et al., 2023). A previous study using post-mortem PD brain tissues reported that 47% of α-synuclein in diseased neurons and in the intracellular Lewy bodies was co-localized with mitochondria-released CytC (Bayir et al., 2009). The peroxidase activity of CytC can cause α-synuclein radical formation and subsequent oligomerization of this protein facilitating PD-associated neuronal death (Kumar et al., 2016). Regarding AD, Aβ treatment induces intracellular CytC-mediated death in Ntera2 human teratocarcinoma cells, a model for early neuroepithelial progenitors. The ubiquitous association between mitochondrial dysfunction and CytC-mediated neuronal death is demonstrated in other neurodegenerative conditions, including HD (reviewed by Jurcau and Jurcau, 2023), amyotrophic lateral sclerosis (reviewed by Zhao et al., 2022), and MS (reviewed by de Barcelos et al., 2019). Consequently, CytC has become a target in the development of therapies for neurodegenerative diseases. One example of a drug targeting CytC is methazolamide, which inhibits the release of this hemoprotein from mitochondria and has been shown to delay disease onset in transgenic mouse models of HD (Wang et al., 2008).

It is likely that extracellular CytC also plays a pivotal role in inflammation associated with CNS injury. In this pathology, the widespread cell death resulting from CytC-mediated apoptosis leads to an increased concentration of CytC in the extracellular space (Au et al., 2012), where previous in vitro studies have demonstrated that it may act as a DAMP-like molecule and further exacerbate neuroinflammation in a TLR4/JNK-dependent manner (Gouveia et al., 2017; Wenzel et al., 2019). Thus, both cytoplasmic and extracellular CytC may contribute to the persistent inflammatory response in neurodegenerative diseases and injury.

Conclusions

Extracellular CytC may act as a DAMP-like molecule in the CNS, triggering glia-mediated neuroinflammation and facilitating the production of neurotoxic molecules and cytokines by both microglia and astrocytes. This process is partially mediated by TLR4/JNK signaling, but further research is necessary to determine other PRRs that CytC may interact with. Given its prosthetic heme group, CytC may bind the same or similar PRRs as extracellular heme which include other TLRs, such as TLR2, potentially contributing to CytC-induced glial activation. However, the covalent binding of heme to the CytC protein may interfere with the ability of the prosthetic heme group to bind to PRRs. This is supported by observations that other hemoproteins, such as hemoglobin, do not always act through the same PRRs as heme (reviewed by Bozza and Jeney, 2020). Overall, extracellular CytC is a prominent and relevant CNS pro-inflammatory signaling molecule that contributes to the pathogenesis of neurodegenerative diseases. Blocking or diminishing its DAMP-like activity in the CNS could be an effective therapeutic approach for these disorders. Identifying the signaling pathways responsible for the CytC-induced glia-mediated neuroinflammation could provide additional targets for drug development in numerous CNS disorders.

Cardiolipin

Role in physiological processes

CL is a phospholipid dimer composed of a glycerol molecule that links two phosphatidyl moieties, each bound to two fatty acyl chains. It is found primarily in mitochondrial membranes, where under physiological conditions, CL accounts for approximately 20% of all IMM lipids but only 3% of the outer mitochondrial membrane (OMM) total lipid content (reviewed by Dudek et al., 2019). CL synthesis is a multistep process occurring in the mitochondrial intermembrane space. Immature or saturated CL (sCL), which contains saturated fatty acyl chains is produced by cardiolipin synthase. The transacylase protein tafazzin (TAZ) enzymatically remodels sCL on the outer leaflet of the IMM to produce mature or unsaturated CL (uCL), which contains unsaturated fatty acyl chains (reviewed by Pizzuto and Pelegrin, 2020). Normally, very little sCL is found outside of mitochondria in mammalian cells due to the TAZ activity, but mutations in the enzyme can limit its ability to produce uCL (Pizzuto et al., 2019). The four fatty acyl chains create a large, bulky hydrophobic region, giving CL a conical shape, which is concentrated in the concave surfaces of the outer leaflet of the IMM cristae, where it provides stability and maintains cristae morphology (reviewed by Paradies et al., 2019). CL plays a crucial role in oxidative phosphorylation and the electron transport chain, as it is required for the activity and organization of complexes I, III, IV, and ATP synthase. CL also facilitates the transport of mitochondrial proteins, such as adenosine diphosphate (ADP)/ATP carrier and Pi carriers, across the IMM, further promoting ATP production (reviewed by Pizzuto and Pelegrin, 2020). Additionally, CL is involved in other processes, including mitochondrial fission and fusion, mitophagy, and apoptosis, underscoring its essential role in mitochondrial function.

Cytosolic activity

CL participates in pro-inflammatory and apoptotic signaling in both peripheral and CNS cells. Cytotoxic stimuli can induce mitochondrial oxidative stress, causing the oxidation of uCL by either ROS or the peroxidase activity of CytC (Jiang et al., 2022). Oxidized uCL disassociates from the IMM and is transported to the outer leaflet of the OMM by phospholipid scramblase 3 and nucleoside diphosphate kinase D (reviewed by Pizzuto and Pelegrin, 2020). Once in the OMM, oxidized uCL can recruit pro-caspase 8 to promote self-cleavage, facilitate the oligomerization of the pore-forming protein BCL-2-associated X protein (BAX), and stabilize pores formed by BAX oligomers, ultimately contributing to MOMP and the release of CytC into the cytosol. Additionally, oxidized uCL can be recognized by the microtubule-associated protein 1A/1B-light chain 3 (LC3), causing mitophagy of the damaged mitochondria (reviewed by Li et al., 2015). Thus, CL plays a crucial role in the immune response of peripheral and CNS cells by responding to pro-apoptotic stimuli and promoting apoptosis and mitophagy.

CL is also involved in the pro-inflammatory signaling of peripheral and CNS immune cells. In primary murine macrophages, CL binds directly to NLRP3, both activating the inflammasome and serving as a mitochondrial docking site for its assembly (Iyer et al., 2013). Treatment of J774A.1 murine macrophages with LPS and ATP in the presence of the saturated long-chain fatty acid palmitate, which decreases CL synthesis, and attenuates secretion of the NLRP3-associated cytokine IL-1β when compared to cells not exposed to palmitate. Additionally, the constitutively expressed precursor to the pro-inflammatory cytokine IL-1α, known as pro-IL-1α, binds to uCL in primary murine macrophages. LC3 and pro-IL-1α share similar binding motifs for uCL, suggesting that pro-IL-1α may compete with LC3 to prevent mitophagy (Dagvadorj et al., 2021). These findings underscore the crucial role of CL in NLRP3 inflammasome activation and function in the periphery.

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

CL released by damaged or dying cells primarily remains bound to the membranes of damaged mitochondria and acts as a DAMP-like molecule in the periphery. In RAW 264.7 murine macrophages, liposomes containing uCL are engulfed at much higher rates than liposomes with other major phospholipids, including phosphatidylserine. This suggests that uCL serves as a potent “eat me” signal, recognized by the macrophage scavenger receptor cluster of differentiation 36 (Balasubramanian et al., 2015). uCL-containing liposomes reduce the production of the pro-inflammatory cytokines IL-6, granulocyte-macrophage colony-stimulating factor, IL-12, and C-X-C motif chemokine ligand 10 (CXCL10, also known as IFN-γ-inducible protein (IP)-10) by LPS-stimulated RAW 264.7 macrophages. This inhibitory activity of uCL is due to its actions as a TLR4 antagonist. Meanwhile, sCL functions as a TLR4 agonist, mimicking the activity of LPS. Liposomes containing sCL induce NF-κB activation and TNF secretion by RAW 264.7 macrophages (Pizzuto et al., 2019). These studies highlight the ability of CL to behave as either a pro- or anti-inflammatory mediator in peripheral immune cell signaling.

uCL interacts with microglia and astrocytes in a manner similar to other DAMPs. uCL in a TLR4-dependent manner increases the phagocytic activity of primary murine microglia two fold compared to untreated cells (Pointer et al., 2019; Wenzel et al., 2021). However, it also reduces the production of neurotoxic molecules by reactive THP-1 monocytic cells used as human microglia models. Pre-treatment with uCL significantly decreases the LPS- and IFN-γ-induced secretion of TNF by these cells. A similar effect is observed in primary murine microglia, where uCL reduces the secretion of TNF and monocyte chemoattractant protein 1 (MCP-1, also known as CCL2), induced by LPS or Aβ stimulation (Wenzel et al., 2021). uCL lowers ROS production by fMLP-stimulated human HL-60 microglia-like cells primed with LPS and reduces concentrations of NO in supernatants from LPS-stimulated BV-2 murine microglia (Pointer et al., 2019). These studies indicate that, in the CNS, uCL acts in a manner similar to its actions in the periphery by inhibiting the activity of TLR4 agonists such as LPS. However, uCL alone induces the secretion of the pro-inflammatory MCP-1 by both BV-2 and THP-1 cells, which is attenuated by TLR4-blocking antibodies (Wenzel et al., 2021). This finding challenges the previous observations of the anti-inflammatory activity of uCL and is similar to the results obtained using U118 MG astrocytic cells, whereby uCL induces the secretion of MCP-1 in a TLR4-dependent manner (Murray et al., 2022). However, uCL pre-treatment reduces LPS-induced glial fibrillary acidic protein expression and IL-1β secretion by U118 MG astrocytic cells. Additionally, uCL significantly increases the phagocytic activity of both U118 MG astrocytic cells and primary murine astrocytes, which may have beneficial, inflammation-resolving effects (Murray et al., 2022). The fact that uCL can act as either a TLR4 antagonist or agonist in microglia and astrocytes suggests it may have a more complex role in glial inflammatory signaling in the CNS compared to other mtDAMPs. These discrepancies also indicate that CL may not align strictly with the established categories of pro- or anti-inflammatory molecules, potentially due to variability in its molecular composition and structure.

Association with peripheral and brain pathologies

CL is involved in the progression of several inflammatory diseases of the periphery and the CNS. Barth syndrome, a rare genetic condition, is characterized by mutations in the TAZ gene that impedes the remodeling of sCL into uCL in mitochondria. This disrupts the structure of cristae, reduces respiratory chain activity, and generally leads to mitochondrial dysfunction, manifesting in patients as cardiac and skeletal myopathy (reviewed by Dudek et al., 2019). The pro-inflammatory effects of sCL in peripheral immune cells provide additional context for this disease, highlighting the importance of CL maturation by TAZ and suggesting significant differences in the effects of uCL and sCL on immune signaling (Pizzuto et al., 2019).

CL also plays a role in age-related mitochondrial dysfunction in the brain, which can contribute to the development of neurodegenerative conditions. Older mice (17–24 months) exhibit a 20%–30% decrease in total brain CL levels, reflecting a progressive decline in all lipid content over time (reviewed by Falabella et al., 2021). Tau aggregates, a notable neurotoxic substance in AD, preferentially bind to CL-rich regions of the OMM in the mitochondria of SH-SY5Y cells, inducing mitochondrial swelling and apoptotic signaling (Cheng and Bai, 2018). Similarly, α-synuclein aggregates associated with PD pathology bind with CL to alter the membrane integrity of mitochondria in murine neuronal cell lines, subsequently causing MOMP and CytC release (reviewed by Ghio et al., 2019). Ultimately, CL significantly contributes to neuroinflammation due to its DAMP-like activity and because its association with pathogenic protein aggregates induces mitochondrial dysfunction and cell death.

Conclusions

Although CL is recognized as a DAMP-like molecule in the CNS, current literature suggests it does not have the same clear pro-inflammatory role that other mtDAMPs do. Both uCL and sCL act as mediators of the peripheral immune response, with the degree of saturation and their oxidation state appearing to affect the pro- or anti-inflammatory activity of these lipids. Similar studies using glial cells and uCL produce conflicting data, demonstrating its ability to act as either a TLR4 agonist or antagonist. Future investigations into the DAMP-like activity of CL should consider the degree of its fatty acyl chain saturation and its oxidation state, as these may be crucial factors influencing the binding potential of this lipid to receptors such as TLR4. Finally, investigating the interactions between pathological protein aggregates and CL holds promise for identifying new treatment strategies in aggregopathies, such as AD and PD, since the DAMP-like activity of these macromolecular complexes has not been investigated yet.

Adenosine Triphosphate

Role in metabolic processes

ATP is the primary biochemical fuel that powers cellular metabolism in organic life. ATP molecules contain an adenine base, a ribose sugar, and three phosphate groups linked by phosphoanhydride bonds (Dunn and Grider, 2023). This nucleotide is generated by the breakdown of glucose in mitochondria, which involves three key processes: glycolysis, the TCA cycle, and oxidative phosphorylation (reviewed by Bonora et al., 2012). The hydrolysis of one or two of the phosphate groups produces ADP or adenosine monophosphate (AMP), respectively, resulting in the release of energy. These molecules are reused by ATP synthases in the aforementioned mitochondrial processes to replenish ATP levels in cells (Neupane et al., 2019). Once ATP has been produced, the ADP/ATP carrier protein transports it across the mitochondrial membranes into the cytosol, where it acts as a coenzyme by providing phosphate groups to various reactions. ADP and AMP are then transported back into the mitochondria to maintain high cytosolic ATP levels (Ruprecht et al., 2019). AMP-activated protein kinase (AMPK) is sensitive to the relative levels of AMP and ATP and is responsible for maintaining sufficient ATP concentrations and altering metabolic activity and relevant gene expression accordingly. When ATP concentrations are low, AMPK upregulates metabolic pathways that generate ATP and inhibit ATP-utilizing processes such as protein and lipid synthesis (reviewed by Bonora et al., 2012). These mechanisms are present in all cell types and provide the basis for cellular functions and activity.

Adenosine triphosphate storage and metabolism

ATP is present in the cytosol of all cells at high millimolar concentrations to provide energy for diverse metabolic processes. Even though ATP can be kept in large dense core vesicles within multiple cell types, it is not stored in the cytoplasm long-term due to a constant demand for energy. Instead, carbon sources such as glycogen are accumulated as precursors for ATP production (reviewed by Bonora et al., 2012). The production of ATP is regulated in reactive immune cells of the periphery as an initial necessary step of the immune response. For example, J774.A1 murine macrophages treated with both LPS and IFN-γ experience NO-induced inhibition of oxidative phosphorylation. While these cells then upregulate glycolysis to compensate for the lack of ATP generation, the total cellular ATP concentration still decreases by 40% after 12 hours, maintaining the mitochondrial membrane potential and ultimately preventing apoptosis (Garedew et al., 2010). Glycolysis generates ATP at a faster rate than oxidative phosphorylation, but is markedly less efficient, producing only two ATP molecules for every glucose molecule compared to the 36 synthesized by oxidative phosphorylation (Shiratori et al., 2019). This allows cells to immediately initiate the production of nucleic acids required for pro-inflammatory gene transcription but lowers overall ATP concentrations in the cell due to high demand and lower efficiency of synthesis. Such a metabolic shift is observed in multiple peripheral immune cell types (reviewed by Hu et al., 2024a).

Reactive microglia and astrocytes demonstrate a similar shift in metabolic activity in response to pro-inflammatory mediators. BV-2 murine microglia stimulated with LPS display decreased oxidative phosphorylation and upregulated glycolysis to support the pro-inflammatory response (Voloboueva et al., 2013). Likewise, increased NO in primary murine astrocytes enhances glycolytic activity and decreases the concentration of ATP in the cytosol by around 25% over one hour (Almeida et al., 2001). In contrast, anti-inflammatory stimuli, such as IL-4, increase oxidative phosphorylation in BV-2 and primary murine microglia (Holland et al., 2018). As discussed above, this metabolic switch in reactive glial cells from oxidative phosphorylation to glycolysis allows for rapid production of cytokines and ROS while lowering the detectable intracellular concentrations of ATP (reviewed by Lauro and Limatola, 2020).

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

ATP can be released from both peripheral and CNS cells into the extracellular space passively or as a result of necrosis, or by active vesicular exocytosis. Typical extracellular concentrations of ATP range from 0.4 μM–1 μM, but this can be increased significantly if ATP is released from cells due to its high millimolar intracellular concentrations. Extracellular ATP can be degraded to ADP, AMP, and eventually adenosine by ecto-ATPases (reviewed by Lombardi et al., 2021). In multiple cell types, extracellular ATP and ADP act on P2X and P2Y receptors. ATP primarily binds to P2X receptors, which are ion channels that allow the transport of cations such as calcium (Ca 2+ ), potassium (K + ), and sodium (Na 2+ ). In primary murine macrophages, ATP treatment causes the influx of Na + /Ca 2+ and efflux of K + across the plasma membrane by the P2X 7 receptor channel (reviewed by Ren et al., 2022). This exchange of cations leads to membrane depolarization and increased intracellular calcium ([Ca 2+ ] i ), which acts as a secondary messenger, contributing to the upregulated expression and release of cytokines such as IL-1β. Conversely, P2Y receptors belong to a family of G protein-coupled receptors and are activated by various nucleotides, including ATP and ADP, depending on the specific P2Y isoform (Puchałowicz et al., 2014). For example, the stimulation of P2Y 2 by ATP potentiates IL-8-induced primary human neutrophil migration approximately fivefold (Kukulski et al., 2009). Thus, ATP is considered a neutrophil chemoattractant; however, it alone does not increase migration significantly, as this purinergic signaling molecule lacks the potency of other molecules such as fMLP. In peripheral murine macrophages, by engaging P2Y 2 and P2X 7 signaling, ATP at low micromolar concentrations induces the production of ROS and CXCL2 (also known as macrophage inflammatory protein 2), indicating ATP facilitates the recruitment of macrophages (Kawamura et al., 2012). Therefore, ATP in the extracellular space is an essential mediator of peripheral neutrophil and macrophage recruitment to sites of cell and tissue damage.

Extracellular ATP plays a similar role in the CNS, where its effects on microglia and astrocytes have been extensively studied. ATP and ADP trigger P2Y receptor-mediated chemotaxis in primary murine microglia (Honda et al., 2001). Additionally, the mRNA and protein expression of P2Y 12 is increased in microglia following nerve injury in mice, activating p38 MAPK (Kobayashi et al., 2008). ATP in the CNS parenchyma also acts on P2X 7 whereby it induces the secretion of TNF by primary rat microglia, which is dependent on increased [Ca 2+ ] i (Hide et al., 2000). Therefore, extracellular ATP can act as a potent activator of the microglial pro-inflammatory response. In fact, it is generally recognized that physiological levels of ATP in the intercellular space of the brain promote active microglial surveillance. This is supported by in vitro studies utilizing apyrase, which hydrolyzes ATP to ADP. Injecting apyrase into mice hippocampal brain slices causes an overall decrease in microglial process ramification and surveillance activity compared to the microglia in control tissue (Madry et al., 2018). Notably, exposure of primary human microglia to ATP at high concentrations decreases their phagocytic activity in a P2X 7 -dependent manner, interfering with the clearance mechanisms (Janks et al., 2018). ATP increases [Ca 2+ ] i by facilitating Ca 2+ ion entry from outside the cell or their release from intracellular stores, such as endoplasmic reticulum and mitochondria. These processes appear to be particularly important for astrocytic reactivity and immune responses. ATP signaling also induces the release of ATP stored in astrocytic vesicles into the extracellular space, which activates nearby cells to propagate immune signaling as demonstrated with primary murine astrocytes (reviewed by Anderson et al., 2004). Treatment of primary murine cortical and hippocampal astrocytes with ATP also causes an efflux of H + , increasing the acidity of the extracellular space and promoting the release of neurotransmitters from nearby neurons (Choi et al., 2021). Mechanical stress can release ATP from primary murine astrocytes, leading to auto-stimulation of astrocytic P2X 7 receptors and activation of NF-кB and the NLRP3 inflammasome, increasing the secretion of IL-1β (Albalawi et al., 2017). Taken together, ATP acts as a DAMP-like molecule on both microglia and astrocytes and is essential in glial signaling and CNS immune responses.

Association with peripheral and brain pathologies

A study using rats intravenously injected with LPS to model systemic inflammation finds that co-treatment with intraperitoneally administered apyrase significantly decreases the LPS-induced upregulation of TNF and IL-1β in plasma and strongly increases the animal survival rate over 48 hours compared to rats treated with LPS only. This suggests that ATP facilitates the progression of systemic inflammatory response syndrome that can occur in the periphery due to trauma or ischemia/reperfusion (Cauwels et al., 2014). Similarly, ATP is implicated as a key signaling molecule in chronic neuroinflammation. It is released following CNS injuries, such as nerve damage and hypoxia-ischemia, or as a result of proteopathies associated with neurodegeneration, causing glial activation. For example, P2X 7 receptor activity is upregulated by around 70% in microglia isolated from post-mortem AD brains compared to microglia from control subjects (McLarnon et al., 2006). Additionally, primary murine macrophages treated with Aβ release significant levels of ATP and IL-1β, but these responses are absent in macrophages isolated from P2X 7 knockout animals (Sanz et al., 2009). Therefore, extracellular ATP is considered a strong contributor to inflammation in peripheral and CNS conditions, prompting it to be targeted either by apyrase-like molecules, which reduce its concentration in intercellular space or by P2 receptor inhibitors blocking its DAMP-like activity (Cauwels et al., 2014).

Conclusions

ATP is widely considered a DAMP of the CNS, contributing significantly to chronic glia-mediated neuroinflammation. P2 receptor activation is recognized as the main mechanism of action of extracellular ATP, allowing immune cells to respond to changes in its concentration. While the effects of extracellular ATP on a broad range of microglial functions have been documented, its effects on astrocyte immune functions have not been studied sufficiently. For example, regulation of the phagocytic activity of astrocytes by ATP has not been explored even though its inhibitory effects on this function in microglia are well known. Determining the effects of extracellular ATP on the phagocytic activity of human and murine astrocytes represents a promising avenue for future research.

Mitochondrial DNA

Role in mitochondrial replication and protein expression

MtDNA is organized into a circular, double-stranded molecule, with a size typically ranging from 15.7 to 19.5 kilobases (kb) across different species. It is present in the mitochondria of all eukaryotic cells (Zanini et al., 2023). Human mtDNA is 16.6 kb and encodes 14 essential proteins involved in oxidative phosphorylation and peptide assembly. Molecules of mtDNA exist in multiple copies, usually around five to ten per mitochondrion in humans with the total number of copies within cells depending on the metabolic activity of the cell and the tissue it resides in (reviewed by Castellani et al., 2020, Habbane et al., 2021, Hu et al., 2024b). In humans, high ATP concentrations-producing heart and skeletal muscle cells contain approximately 7000 copies of mtDNA, while cells with lower ATP requirements, such as neutrophils, have around 150 copies (reviewed by Castellani et al., 2020; Picard, 2021). Fixed samples from the mouse brain contain different relative mtDNA copy numbers depending on the region, ranging from 0.09 to 2.5, where 1.0 represents a median value of 39 brain regions (Fuke et al., 2011). In mammalian cells, mitochondrial division and mtDNA replication proceed independently of the cell cycle and are regulated by nuclear-encoded proteins such as TFAM (reviewed by Castellani et al., 2020). The previously mentioned well-established endosymbiotic theory of the origin of mitochondria is supported by the similar size, structure, and method of replication observed in mtDNA and prokaryotic DNA (reviewed by Boguszewska et al., 2020). Consequently, eukaryotic cells recognize mtDNA as an invasive pathogenic signal when it is released into the cytosol or extracellularly, allowing for the binding of these molecules to their cognate PRRs (reviewed by Marchi et al., 2023).

Cytosolic activity

MOMP is a highly regulated event in apoptotic cell death that facilitates the release of mtDNA into the cytosol, leading to the activation of several pro-inflammatory pathways in both peripheral and CNS cells. MOMP is regulated by the proapoptotic pore­forming proteins BAX and BCL-2 antagonist/killer 1, which are key in initiating apoptosis. Most released mtDNA tends to associate with the permeabilized mitochondria and therefore does not freely disperse into the cytosol, meaning the removal of these damaged organelles by mitophagy is a viable method of reducing cytosolic mtDNA-related pro-inflammatory signaling (reviewed by Marchi et al., 2023). In fact, mtDNA-bound TFAM released from mitochondria has been shown to interact with LC3, promoting mitophagy and preventing mtDNA-induced immune activation (Liu et al., 2024). Molecules of mtDNA are detected by cyclic GMP-AMP synthase (cGAS), a cytosolic DNA sensor, triggering a signaling cascade that activates the stimulator of IFN genes (STING) pathway in peripheral immune cells. Primary murine peripheral macrophages harvested from mice with reperfusion-induced hepatic injury have increased levels of cytosolic mtDNA, which engages cGAS/STING signaling and significantly increases the secretion of pro-inflammatory TNF and IL-6 (Zhong et al., 2022). The activation of cGAS/STING by cytosolic mtDNA also facilitates the assembly of the NLRP3 inflammasome, upregulating the secretion of IL-1β (Shimada et al., 2012). Overall, the role of cytosolic mtDNA as a pro-inflammatory mediator associated with the initiation of programmed cell death in peripheral cell types is well defined in current literature.

The signaling pathways for intracellular mtDNA in microglia and astrocytes are similar to those in the immune cells of peripheral tissues. Treating primary murine microglia with LPS increases the levels of cytosolic mtDNA fivefold, subsequently upregulating the expression of cGAS (Wei et al., 2024). Reactive primary murine astrocytes also demonstrate increased mtDNA-induced cGAS/STING signaling, which is inhibited by C-176, a STING antagonist, and cyclosporin A, which prevents the leakage of mtDNA from mitochondria (Chen et al., 2024). Notably, murine microglia appear to express higher levels of cGAS than astrocytes, suggesting that mtDNA-related activation of the immune response may be more significant in microglia (Huang et al., 2024). Furthermore, mtDNA-induced cGAS/STING signaling may be linked with the assembly of the NLRP3 inflammasome in microglia. For example, BV-2 murine microglia treated with sevoflurane, a common inhalation anesthetic, release mtDNA into the cytosol, increasing cGAS/STING signaling and subsequently activating NLRP3. This is confirmed by the pre-treatment of BV-2 cells with inhibitors of mtDNA release and cGAS/STING activation, which reduces sevoflurane-induced NLRP3 assembly (Yang et al., 2024). Astrocytes also express components of the NLRP3 inflammasome, which is activated by select DAMPs such as ATP. However, cytosolic mtDNA-dependent cGAS/STING activity may not induce or significantly affect the activation of NLRP3 in astrocytes (reviewed by Kodi et al., 2024). Overall, the available data indicate that the release of mtDNA from mitochondria plays an important role in glial apoptosis and pro-inflammatory signaling, but it could be less consequential in astrocytes.

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

MtDNA is released into the extracellular space by damaged and apoptotic cells of both the periphery and the CNS. These molecules stimulate the same PRRs that recognize bacterial DNA during an infection, namely TLR9 (reviewed by Marchi et al., 2023). This mainly intracellular receptor is located on endosomal membranes and specifically recognizes regions of unmethylated cytosine-phosphate-guanine– a feature of both bacterial and mtDNA. Peripheral immune cells can internalize extracellular mtDNA by endocytosis, facilitating the activation of TLR9. Some cell types, including B lymphocytes and peripheral blood monocytes, also express TLR9 on their surface, allowing for direct activation by extracellular mtDNA (Nakayama and Otsu, 2018). Human neutrophils incubated with mtDNA demonstrate increased phosphorylation of p38 MAPK and upregulated secretion of pro-inflammatory matrix metalloproteinase-8, -9. These processes are facilitated by TLR9 and the subsequent recruitment of MyD88 (Zhang et al., 2010). Additionally, by binding to TLR9, extracellular mtDNA can induce the secretion of type I IFNs in primary human plasmacytoid dendritic cells. The activation of advanced glycation end-product receptors (AGER, also known as RAGE) by TFAM is augmented by mtDNA, but this receptor does not appear to associate with mtDNA alone (Julian et al., 2012). The signaling pathways activated by extracellular mtDNA alone and in combination with other mitochondrial DAMPs such as NFPs and TFAM have been recently reviewed in detail by Marchi et al. (2023) and Tao et al. (2024).

In the context of the CNS, the accumulation of mtDNA in the extracellular space can occur due to neurodegenerative processes. These molecules are endocytosed by glia and contribute to the persistent immune response by reactive microglia and astrocytes in several neuropathologies (reviewed by Zhao et al., 2021). However, the effects of mtDNA in the CNS have not been explored as extensively as in the periphery. It is known that the cytosolic activity of mtDNA is relevant for the glial immune response to extracellular mtDNA since its transport into cells and PRR binding facilitates cGAS/STING-induced cytokine production (reviewed by Zhao et al., 2021). Matsuda et al. (2015) demonstrate that primary murine microglia cultured in the conditioned medium from overactivated primary murine hippocampal neurons, which release nuclear and mtDNA, show increased TLR9 activity, leading to the activation of NF-κB and three-fold increased secretion of TNF. Extracellular nuclear DNA may contribute to the activation of TLR9 in this case, but it is likely that mtDNA also participates significantly (Jovasevic et al., 2024). Only one study reports the effects of mtDNA on human microglia by showing that HMC3 cells treated with this extracellular DAMP increase the production of intracellular ROS (Nasi et al., 2020). Although extracellular mtDNA plays a key role in neuron and glial cell signaling, more research is needed to clarify the mechanisms that facilitate a pro-inflammatory response to mtDNA, especially in human glia.

Association with peripheral and brain pathologies

An increase in circulating mtDNA levels in serum has been identified in over 60 peripheral and CNS diseases in humans, suggesting the potential of mtDNA as a biomarker (reviewed by Nakayama and Otsu, 2018; Pinti et al., 2021). Some of these diseases, including ischemic heart disease, atherosclerosis, liver diseases, MS, AD, and PD, are characterized by chronic inflammation (reviewed by Nakayama and Otsu, 2018; Zhao et al., 2021; Zhong et al., 2022). Mitochondrial dysfunction and increased circulating mtDNA levels are implicated in aging processes and the latter positively correlates with serum concentrations of pro-inflammatory TNF and IL-6. Human peripheral monocytes can be activated by serum mtDNA, leading to a persistent pro-inflammatory response that contributes to multiple cardiovascular diseases (Nakayama and Otsu, 2018). MtDNA levels could become elevated as a result of cell death observed in some of these pathologies. For example, disease-associated cardiomyocyte necrosis leads to increased serum mtDNA levels and can subsequently exacerbate inflammation. Ingested mtDNA and macrophage-STING signaling are implicated in the progression of non-alcoholic and alcoholic fatty liver disease (Zhang et al., 2022).

In the CNS, the extracellular mtDNA levels in the brain may depend on the specific neurodegenerative condition. Patients with progressive MS have significantly elevated mtDNA concentrations in their cerebrospinal fluid compared to healthy subjects and individuals with other neurological conditions, such as AD or PD (Pinti et al., 2021; Zhao et al., 2021). This may be due to the pathophysiological differences between these diseases. AD and PD are characterized in part by neuronal mitochondrial dysfunction, leading to smaller and fewer mitochondria that are released upon neuronal death. Conversely, MS is an autoimmune disorder in which damaged neurons release larger amounts of mitochondrial DAMPs into the extracellular space (Cai and Tammineni, 2017). Consequently, serum mtDNA as an indicator of mitochondrial dysfunction could be used to diagnose or assess the severity of multiple peripheral and CNS diseases.

Conclusions

Extracellular mtDNA acts as a DAMP-like molecule in the CNS, primarily by activating TLR9-related signaling pathways. However, most of the existing research has focused on the effects of mtDNA on peripheral cell types, notably neutrophils, where the functions of mtDNA as an activator of the cGAS/STING pathway and in NLRP3 inflammasome assembly are well established. In the CNS, the activation of cGAS/STING also occurs after extracellular mtDNA is endocytosed by glial cells. Exploring the interactions of mtDNA with other PRRs, such as AGER, may provide a broader understanding of the mechanisms of action of extracellular mtDNA, especially in combination with other mitochondrial DAMPs such as TFAM and NFPs. Additionally, more studies are needed to determine whether high concentrations of extracellular mtDNA measured in plasma or CSF may serve as a biomarker of specific neurodegenerative diseases.

Mitochondrial Transcription Factor A

Role in mitochondrial DNA packaging and expression

TFAM is a 25 kDa nuclear-encoded protein that regulates mtDNA transcription, maintenance, and packaging. It belongs to the high-mobility group (HMG) protein family and displays approximately 76% sequence homology with HMG box (HMGB)1, a protein known primarily for its function as a nuclear transcription cofactor (reviewed by Bajwa et al., 2019). For every molecule of mtDNA, approximately 1000 molecules of TFAM are present in mitochondria. TFAM coats mtDNA strands to provide stability and regulate gene expression by facilitating the packaging of nucleoids, which are DNA-protein complexes located in the mitochondrial matrix (reviewed by Kang et al., 2018; Lin et al., 2022). The expression of TFAM is regulated by nuclear respiratory factor (NRF)1, while its binding affinity to mtDNA is controlled by acetylation or phosphorylation of its HMG1 domain (reviewed by Lin et al., 2022). TFAM facilitates transcription by binding in a sequence-dependent manner to the mtDNA promoter regions. It also binds to and coats mtDNA molecules in a sequence-independent manner for packaging (reviewed by Kang et al., 2018). Therefore, TFAM regulates the switch from mtDNA replication to transcription, as the two processes are mutually exclusive in mitochondria. Thus, increasing concentrations of intramitochondrial TFAM decrease the availability of mtDNA for replication (reviewed by Fontanesi et al., 2020). Due to its role in mtDNA replication and transcription, TFAM is likely a major regulator of mtDNA copy number (reviewed by Kang et al., 2018).

Intramitochondrial and intracellular activity

Lowered concentrations of intramitochondrial TFAM are linked to mitochondrial dysfunction, which leads to the unpacking of mitochondrial nucleoids in peripheral and CNS cells. This allows both mtDNA molecules and TFAM to leak into the cytosol, where the former is fragmented by DNase (reviewed by Harrington et al., 2017). As mentioned previously, mtDNA-TFAM complexes interact with LC3 inducing mitophagy (Liu et al., 2024). The released mtDNA and TFAM are also detected by cGAS, which triggers a signaling cascade that activates the STING pathway. This molecular mechanism is described for both macrophages and T lymphocytes in the periphery (reviewed by Ou et al., 2021) and microglia in the CNS (reviewed by Lin et al., 2022). These cell types then launch an immune response, initiating the release of type I IFNs, such as IFN-α and IFN-β (reviewed by Ou et al., 2021; Paul et al., 2021). The immune response of peripheral macrophages and microglia is enhanced by the intracytosolic mtDNA, which engages TLR9 and NLRP3 inflammasome signaling pathways (reviewed by Lin et al., 2022). Consequently, any processes or mechanisms that cause TFAM to leak into the cytosol of peripheral macrophages or microglia may contribute to a pro-inflammatory response.

In contrast to the pro-inflammatory events caused by decreased intramitochondrial TFAM levels, overexpression of this transcription factor exerts a protective effect in both peripheral and CNS cells by promoting mitochondrial biogenesis and enhancing overall cell survival (reviewed by Kang et al., 2018). In the periphery, human cardiomyocytes exposed to hyperglycemic conditions have their mitochondrial ATP production restored to normal levels when TFAM is overexpressed (Suarez et al., 2008). Moreover, transgenic mice overexpressing human TFAM are protected from memory impairment and cardiac damage following experimental myocardial infarction compared to wild-type animals (DeBalsi et al., 2017). TFAM overexpression in the mitochondria of CNS cells also provides protection, as SH-SY5Y human neuroblastoma cells treated with Aβ or the neurotoxin 1-methyl-4-phenylpyridinium (MPP + ) are significantly shielded from the toxic effects of these molecules when TFAM is overexpressed (Xu et al., 2009). However, pro-inflammatory outcomes of elevated mitochondrial TFAM levels have also been reported in the CNS. For instance, hypoxia-induced overactivation of primary murine microglia causes the upregulation of NRF1 and subsequently increases intramitochondrial TFAM concentration, promoting mtDNA transcription, which is necessary for the formation of the NLRP3 inflammasome (Wang et al., 2022). More research is needed to confirm the effects of overexpressed TFAM on the immune response of CNS cells, specifically microglia and astrocytes.

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

Damaged cells and tissues can release TFAM into the extracellular space, where it acts as a DAMP. This activity is consistent with that of HMGB1, the homolog of TFAM, and a well-established DAMP that affects multiple cell types in both the periphery and CNS (reviewed by Bajwa et al., 2019). The role of TFAM as a DAMP in peripheral cells and tissues has been extensively studied. Investigations with RAW 264.7 murine macrophage-like cells and primary rat peritoneal macrophages following hemorrhagic shock reveal that extracellular TFAM upregulates the production of pro-inflammatory cytokines such as TNF and IL-1β (Chaung et al., 2012). Similarly, in murine splenocytes, TFAM-mtDNA complexes interact with the AGER and TLR9, triggering the release of TNF (Julian et al., 2013). Human monocytes exposed to a combination of mitochondrial NFPs and TFAM increase the secretion of IL-8 threefold compared to cells treated with TFAM, which does not change it significantly when added alone (Crouser et al., 2009). These studies underscore the impact of extracellular TFAM as a mtDAMP in peripheral tissues, particularly when acting in conjunction with other established immune stimulants such as mtDNA and NFPs.

Extracellular TFAM may also act as a DAMP in the CNS; however, only limited studies report the effects of this protein on the brain cell types. When combined with IFN-γ, TFAM induces a cytotoxic response in THP-1 human monocytic cells, a commonly used microglial model, leading to significantly enhanced production of IL-1β, IL-6, and IL-8 (Little et al., 2014). The supernatant of TFAM-exposed monocytic cells becomes cytotoxic, reducing the viability of human SH-SY5Y neuroblastoma cells by approximately 50% compared to supernatants from unstimulated THP-1 cells. However, when TFAM or IFN-γ is used alone, microglial cytotoxic response is not observed. Interestingly, the combination of TFAM and NFPs does not facilitate the activation of human microglia-like cells (Little et al., 2014). The same study shows that treatment of primary human microglia with TFAM plus IFN-γ yields similar results and leads to a threefold increase in IL-6 production compared to microglia treated with IFN-γ alone. Since antibodies against the AGER and macrophage antigen complex (Mac)-1 receptor reduce TFAM plus IFN-γ-induced cytotoxicity, both these receptors are suggested as the microglial molecular targets of TFAM (Schindler et al., 2018). However, the effects of these blocking antibodies are only partial, indicating the engagement of additional microglial signaling by extracellular TFAM. These in vitro findings are augmented by in vivo experiments that demonstrate injecting TFAM alone into the cisterna magna of rats causes significantly increased expression of MCP-1, IL-1β, IL-6, TNF, and NF-κB inhibitor alpha in the frontal cortex and hippocampus (Schindler et al., 2018). Taken together, these studies indicate that the DAMP-like activity of TFAM in the CNS is comparable to its actions in the peripheral tissues.

Association with brain pathologies

Decreased levels of TFAM protein expression are linked to mitochondrial dysfunction and are associated with CNS pathologies such as AD, PD, and HD, in which TFAM protein expression in the brain is reduced on average by 52%, 36%, and 31%, respectively (reviewed by Kang et al., 2018). The exact reduction in TFAM expression varies between individual studies across neurodegenerative diseases, ranging from 88% (Petrov et al., 2016) to a 20% (Pedrós et al., 2014) decrease compared to study-specific control values. The mechanism of reduced TFAM protein expression in neurodegenerative disorders is not well understood but is thought to be linked to mitochondrial dysfunction, a process also associated with natural aging (DeBalsi et al., 2017). Thomas et al. (2012) report a 60% reduction in TFAM protein expression in the brains of 20-month-old C57Bl/6 mice compared to younger adult mice between 8 and 10 weeks old. Therefore, age-associated decreases in TFAM protein expression may be a risk factor in the progression of neurodegenerative conditions rather than a consequence of disease processes. Conversely, as discussed previously, high mitochondrial TFAM levels are protective in peripheral conditions such as type 2 diabetes mellitus (Gauthier et al., 2009) and heart failure (DeBalsi et al., 2017), as well as in neuropathologies including AD (Xu et al., 2009).

Conclusions

Overall, TFAM acts as a DAMP-like molecule of the CNS. It does this by activating microglia by engaging their AGER and Mac-1 receptors, but it remains unclear which other receptors and signaling pathways TFAM activate to upregulate the production of cytokines and cytotoxins. The complete spectrum of molecules released by reactive microglia in response to TFAM and its other co-stimulants is also unknown, as distinct sets of neurotoxins can be released by microglia stimulated by a diverse range of mediators (Lindhout et al., 2021; Bernath et al., 2023). Since HMGB1 acts on TLRs such as TLR2, TLR4, and TLR9, it is plausible that TFAM may also interact with these receptors given the structural and biochemical similarities between these transcription factors (reviewed by Yuan et al., 2020); however, this has not yet been confirmed experimentally. Further investigation into the activity of TFAM as a DAMP when administered alone or in conjunction with other stimulants such as mtDNA, NFPs, or IFN-γ is also necessary. Exploring the DAMP-like activity of TFAM-mtDNA complexes could provide especially valuable insights considering the strong association between these two molecules. Additionally, astrocytes have been largely overlooked as potential cellular mediators of the DAMP-like effects of extracellular TFAM. A better understanding of the impact that TFAM has on all glial cell types could help clarify its role in modulating CNS inflammation. Lastly, the consequences of intramitochondrial TFAM protein overexpression on its extracellular activity as a DAMP have gone largely unaddressed. The protective role that intramitochondrial TFAM appears to have in multiple models of diseases is contradicted by its pro-inflammatory activity when released extracellularly. While this paradox is not unique to this protein, a study exploring the effects of altered TFAM expression on its extracellular levels and activity is warranted. A more in-depth understanding of this relationship may shed additional light on its importance in inflammatory processes.

N-Formyl Peptides

Role in mitochondrial DNA-encoded protein translation and cell signaling

NFPs are amino acid chains containing an N-formylated methionine. These peptides can vary in length and originate from bacterial or mitochondrial protein cleavage (reviewed by Trojan et al., 2020). Similar to bacterial peptide synthesis, mitochondrial protein translation requires the recognition of an N-terminal N-formylated methionine-transfer RNA by mitochondrial translational initiator protein 2 (Escobar-Alvarez et al., 2009); therefore, in eukaryotes, NFPs are generated within mitochondria. The most characterized NFP is fMLP, a three amino acid chemotactic factor extensively used in studying peptide-receptor interactions due to the ease of synthetic preparation owing to its short length (reviewed by Yang et al., 2008). In the extracellular space, NFPs act as agonists of G protein-coupled formyl peptide receptors (FPRs), three of which are identified in humans: FPR1, FPR2, and FPR3. FPR1 and FPR2 are both expressed in a wide range of peripheral and CNS immune cell types, including monocytes, neutrophils, dendritic cells, microglia, and astrocytes, whereas FPR3 expression is more limited (reviewed by Trojan et al., 2020). FPR1 exhibits 100-fold higher selectivity towards NFPs compared to their non-formylated counterparts (Wenzel-Seifert and Seifert, 2003). Conversely, FPR2 is a non-specific, extremely promiscuous receptor that can bind to a wide variety of molecules including lipids, peptides, and proteins, inducing either pro- or anti-inflammatory effects depending on the ligand (reviewed by Dorward et al., 2015). Both receptors are linked to cellular immune responses, with FPR1 activation playing a key role in neutrophil chemotaxis to sites of tissue damage or infection and FPR2 being implicated in the pathogenesis of various inflammatory diseases, including systemic lupus erythematosus and AD (reviewed by Trojan et al., 2020). Thus, NFPs and their receptors play an important role in immune cell signaling in both the periphery and the CNS.

Intramitochondrial and cytosolic activity

NFPs are present in the mitochondria of all cells, although most research focuses on peripheral cell types. While still in mitochondria, NFPs can undergo deformylation by peptide deformylase (PDF), a nuclear-encoded mitochondrial metalloprotease that removes the formyl moiety from NFPs (Lee et al., 2004). PDF may have anti-inflammatory activity, as extracellular deformylase enzymes isolated from the intestinal mucosal layer of Sprague−Dawley rats can cleave bacterial NFPs, restricting their ability to act as ligands for FPR1 (Nguyen and Pei, 2005). The human homolog of PDF, known as HsPDF, can deformylate 10 out of the 14 known human mitochondria-encoded peptides (Escobar-Alvarez et al., 2009, 2010; Hu et al., 2024b). Interestingly, the observed coexistence of both formylated and deformylated forms of the same peptides in the mitochondria of peripheral human cells indicates that HsPDF exhibits poor catalytic activity, meaning NFPs are always present in mitochondria (Nguyen and Pei, 2005; Escobar-Alvarez et al., 2010). Despite these studies, the exact role of PDF and the effects of changing ratios of formylated and deformylated peptides remain poorly understood. In peripheral human cells, the inhibition of HsPDF activity leads to a global reduction in mtDNA-encoded protein translation and subsequently lowered mitochondrial energy production, indicating that HsPDF is crucial for normal mitochondrial activity (Escobar-Alvarez et al., 2010). Further research is necessary to characterize the roles that PDF plays in mitochondrial function and immune cell signaling.

Low levels of NFPs are detected in the cytosol of HeLa cells, although their function in this cellular compartment remains unclear (Seo et al., 2020). Cytosolic NFP concentrations can be elevated by oxidative stress, such as exposure to ultraviolet radiation, or by silencing of the methionyl-transfer RNA formyl transferase gene, which disrupts mitochondrial activity. However, the transient transfection of synthetic mitochondrial NFPs into cells does not affect their viability or the activity of NF-кB. The only reported effect of cytosolic NFPs is to enhance the survival of bacteria during an infection (Seo et al., 2020). Investigations into the effects of cytosolic NFPs may offer further insights into their role in the cellular immune response, particularly in the CNS, as research in this area has not been performed yet.

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

In contrast to bacteria, no evidence indicates that mitochondrial NFPs are actively secreted by live eukaryotic cells; rather, they are released as a result of plasmalemma rupture caused by necrotic cell death (reviewed by Dorward et al., 2015). Upon entering the extracellular space, mitochondrial NFPs act as DAMP-like molecules in both the periphery and the CNS by interacting with FPR1 to induce an immune response. For example, peripheral human neutrophils treated with fMLP display chemotactic migration, upregulated phagocytic activity, and the respiratory burst response (Herrmann et al., 2007). The binding of fMLP to FPR1 on these cells initiates a complex signaling cascade that recruits multiple secondary messengers, including inositol trisphosphate (IP3), which regulates the mobilization of Ca 2+ and protein kinase C, leading to activation of MAPKs including p38 and extracellular signal-regulated kinase (ERK)1/2 (reviewed by Yang et al., 2008; Hazeldine et al., 2015). Ultimately, this induces secretion of the pro-inflammatory cytokine IL-8 and assembly of nicotinamide adenine dinucleotide phosphate (NADPH)-oxidase complexes that produce ROS (Hidalgo et al., 2015). NFPs appear to be the primary mtDAMP driving neutrophil activation, as the inhibition of FPR1 and MAPK signaling in primary human neutrophils treated with a mixture of mtDNA and NFPs results in significantly decreased L-selectin shedding and IL-8 secretion, both of which are associated with upregulated cell migration (Hazeldine et al., 2015). The effect of fMLP on primary human monocytes is similar: they are activated when exposed to fMLP in combination with TFAM, causing a significant increase in IL-8 production (Crouser et al., 2009). Neither fMLP nor TFAM alone induces monocytic IL-8 production. Peripheral human macrophage cell lines are also activated by fMLP, leading to the secretion of TNF, IL-1, and NO, which can be reduced by a pretreatment with protein kinase C and protein tyrosine kinase inhibitors (Shrivastava, 2007). It can be concluded that NFPs play a pivotal role in the activation of peripheral immune cells, with much of the research focused on the regulation of neutrophil immune functions and the respiratory burst response of phagocytes.

The effects of fMLP and, by extension, other NFPs, on microglia and astrocytes are not as widely studied as in peripheral cells. As mentioned previously, both these CNS cells express FPRs (reviewed by Trojan et al., 2020). However, microglia do not exhibit migratory responses to fMLP treatment, potentially indicating lower levels or differing functions of FPRs in microglia compared to neutrophils (reviewed by Trojan et al., 2020). The treatment of BV-2 murine microglia with fMLP also does not induce a detectable secretion of the cytokine CXCL10 (Bernath et al., 2023). Nevertheless, even sub-picomolar concentrations of fMLP can induce oxidative stress by activating NADPH oxidase of microglia in a primary rat and mouse midbrain cell culture model of PD, leading to the death of dopaminergic neurons (Gao et al., 2008). Previous studies have used HL-60 myelomonocytic cells, differentiated with dimethyl sulfoxide, to model microglial respiratory bursts. After these cells are primed with immunomodulators, such as LPS, their exposure to fMLP results in a rapid increase in ROS production. This assay has been used to demonstrate that the microglial fMLP-induced respiratory burst response can be inhibited by short-chain fatty acids (Wenzel et al., 2020) and primed by not only LPS but also another mtDAMP, CytC (Gouveia et al., 2017). Mitochondrial NFPs likely play a role in the immune response of microglia and astrocytes in the CNS, but further studies are required to advance the understanding of these mechanisms.

Role in infection and systemic inflammation

The DAMP-like activity of NFPs is associated with various pathological conditions in both the periphery and the CNS. During bacterial infections, pathogens release NFPs while also causing the destruction of host tissues, leading to the release of host-derived mitochondrial formylated peptides. This sequence of events triggers an immune response including the recruitment of neutrophils and other immune cells to the site of infection (reviewed by Dorward et al., 2015). In addition to bacterial infections, mitochondrial NFPs are implicated in tissue injury and post-trauma inflammation. Plasma levels of mtDNA, serving as an indicator of released mitochondrial contents including NFPs, are markedly elevated in patients with trauma-induced systemic inflammatory response syndrome compared to healthy subjects. Injection of mtDAMPs in the form of sonicated murine mitochondria into mice peritoneum results in FPR1-dependent neutrophil recruitment (Zhang et al., 2010). This is consistent with traumatic brain injury, where immune cells such as neutrophils can be recruited to the injury site by releasing pro-inflammatory molecules. In vivo research with traumatic brain injury model mice reveals that the knockout of the FPR1 gene has beneficial short-term effects, reducing acute inflammation by decreasing the activity of pro-inflammatory MAPK signaling (Fusco et al., 2020). However, after four weeks, FPR1 knockout mice demonstrate lower levels of neurogenesis and self-renewal of neurons due to a lack of neuronal differentiation, which is typically promoted by FPR1 activity. These findings suggest an important role for mitochondrial NFPs in post-trauma CNS inflammation and recovery. However, further studies are necessary to better define the role of NFPs within these processes given the limited research performed to date (Lubkin et al., 2018).

Conclusions

Mitochondrial NFPs play a pivotal role in systemic inflammation by being released extracellularly following tissue damage, triggering an immune response and initiating chemotaxis, particularly in peripheral neutrophils. This role likely extends to brain inflammation; however, there is limited research exploring the effects of NFPs on CNS cells. It remains unclear whether intracellular NFPs and the deformylating enzyme PDF are involved in glial cell signaling, as limited previous studies with other cell types have suggested. Future research should focus on investigating the effects of NFPs on CNS immune cells, potentially in conjunction with other mtDAMPs that may further facilitate their pro-inflammatory signaling. Examination of the biological activity of these molecules, once they are released into the CNS extracellular space, may offer additional context for their role in neuroimmune signaling and may highlight these molecules as potential targets for treating neuroinflammation in neurotrauma and neurodegenerative diseases.

Succinate

Role in the tricarboxylic acid cycle

Succinate is a key metabolic intermediate in the TCA cycle that also contributes to several other important cellular pathways such as heme synthesis, the gamma-aminobutyric acid (GABA) shunt, ketone body utilization, branched-chain amino acid metabolism, and macrophage-specific itaconate metabolism (reviewed by Wang et al., 2021). Succinate is synthesized in the mitochondrial matrix and serves as the substrate for mitochondrial complex II, also known as succinate dehydrogenase (SDH). This enzyme drives ATP production from the electron transport chain by supplying electrons to ATP synthase with the oxidization of succinate to fumarate (Harber et al., 2020). The GABA shunt is a process that provides carboxylic acids to the TCA cycle and enhances energy production when other sources, such as the glucose/pyruvate pathway, are compromised (reviewed by Salminen et al., 2016). GABA is transaminated to form glutamate and succinate semialdehyde, which is then oxidized into succinate to enter the TCA cycle. In astrocytes and microglia, succinate semialdehyde reductase can convert succinate semialdehyde in a context dependent manner to gamma-hydroxybutyrate, an inhibitory neurotransmitter known to have neuroprotective properties (reviewed by Salminen et al., 2016; Tretter et al., 2016). The contribution of itaconate to succinate metabolism is also noteworthy, as itaconate inhibits SDH activity and thereby interferes with succinate catabolism (reviewed by Tretter et al., 2016). In addition to its essential metabolic role, succinate is also involved in both intra- and extracellular immune signaling.

Intramitochondrial and cytosolic activity

Succinate accumulates in the intracellular compartments of peripheral macrophages during an immune response and plays a crucial role in inflammatory signaling. Treatment of primary murine peripheral macrophages with LPS causes a significant increase in the concentration of intramitochondrial succinate, which is then transported into the cytosol by the dicarboxylic acid transporter, also known as solute carrier family 25 member 10 (SLC25A10) in mammalian cells (Tannahill et al., 2013; Murphy and Chouchani, 2022). This accumulation of succinate results from increased LPS-induced glutamine import, leading to the process of glutamine-dependent TCA cycle-intermediate replenishment known as anaplerosis (Tannahill et al., 2013). Cytosolic succinate suppresses prolyl hydroxylase activity, allowing for the stabilization and activation of hypoxia-inducible factor (HIF)-1α (Tannahill et al., 2013; Harber et al., 2020). Activated HIF-1α increases primary murine macrophage IL-1β secretion, contributing to the associated pro-inflammatory response (Harber et al., 2020). As discussed previously, the means of ATP synthesis are shifted from oxidative phosphorylation to glycolysis in LPS-stimulated primary murine peripheral macrophages. This allows mitochondria to maintain a higher membrane potential and increases mitochondrial ROS production by the inverse catalysis of SDH and reverse electron transport through mitochondrial complex I (Mills et al., 2016; Zhang et al., 2020). Overall, accumulated intramitochondrial and intracellular succinate acts as a mediator of the pro-inflammatory response to LPS in peripheral macrophages.

In the CNS, the transition of microglia into a pro-inflammatory state causes intracellular succinate accumulation as well as the metabolic shift similar to that described in the periphery, increasing ROS production. An inhibitor of mitochondrial fission, Mdivi-1, reduces this immune response and normalizes ROS production and succinate accumulation accordingly (Nair et al., 2019). However, the pre-treatment of primary murine microglia with cell-membrane permeable diethyl succinate is demonstrated to have an anti-inflammatory effect on LPS-induced microglia activation, since it lowers ROS production by 20% and reduces the upregulation of mitochondrial fission, both of which are associated with the typical inflammatory response (Wang et al., 2021). Intracellular succinate is certainly a critical modulator of microglial immune responses but there is an apparent lack of research on the intracellular effects of succinate in astrocytes. Regardless, its role in glial cells is undoubtedly complex and likely dependent on various external factors that might regulate the normal functions of these cell types.

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

Succinate can be released into the extracellular space following cell necrosis or tissue damage, where it acts as a DAMP and plays an established role in the inflammatory signaling of peripheral immune cells. In primary murine peripheral macrophages, the GPR91 receptor, also referred to as succinate receptor (SUCNR)1, binds extracellular succinate to upregulate HIF-1α-dependent IL-1β production (Littlewood-Evans et al., 2016). SUCNR1 is expressed in numerous cell types across multiple peripheral tissues in both mice and humans, including human embryonic kidney (HEK)293 cells, dendritic cells, T cells, cardiomyocytes, and adipocytes (reviewed by Deen and Robben, 2011). Although SUCNR1 is not usually referred to as a PRR in the literature, it exhibits several properties that suggest it acts in a PRR-like manner. The signaling cascade for SUCNR1 is not fully understood, but succinate-induced activation in HEK293 and human monocyte-derived dendritic cells involves IP3 accumulation, Ca 2+ mobilization, and ERK1/2 phosphorylation, pathways also associated with inflammatory responses (Grazioli and Pugin, 2018). Succinate-dependent ERK1/2 activation also appears to synergize with the activity of TLR ligands such as polyinosinic-polycytidylic acid (polyI:C) in primary human monocyte-derived dendritic cells, as the combination of succinate and polyI:C increases ERK1/2 phosphorylation almost two-fold at certain concentrations of succinate compared to when succinate is used alone (Rubic et al., 2008). Extracellular succinate and SUCNR1 are therefore notable regulators of immune cell activation in multiple peripheral cell types.

The effects of extracellular succinate on microglia and astrocytes are understudied in comparison to the peripheral cell types. Primary mouse microglia treated with cell-membrane impermeable disodium succinate alone or in combination with LPS do not display any significant changes in cytokine secretion or ROS production compared to control cells (Wang et al., 2021). Conversely, the injection of ethylmethylhydroxypyridine succinate into mice peritoneum leads to a shift of microglia to a reactive state in the brains of 18-month-old mice. This form of succinate crosses the BBB and the cell membrane and therefore is not confined exclusively to the extracellular space. Succinate injection causes an approximate 8% decrease in TNF and IL-1β secretion in 18-month-old mice (Kirova et al., 2021). In primary murine hypoxia-ischemia astrocyte models, succinate-SUCNR1 signaling increases expression of ERK1/2 and vascular endothelial growth factor, demonstrating the integral role that succinate and SUCNR1 play in post-hypoxia-ischemia revascularization and recovery (Hamel et al., 2014). These studies suggest the contribution of succinate to immune cell signaling may be similar in peripheral cells and glia. However, it is still unclear whether extracellular succinate induces an overall pro- or anti-inflammatory response in glial cells.

Association with brain pathologies

Succinate is involved in several metabolic pathways associated with CNS pathologies, acting as either a protective or harmful molecule. Mitochondrial complex I deficiency represents a condition in which succinate acts as a protective metabolite. The direct interaction of SDH with succinate can bypass the highly damage-prone mitochondrial complex I, also known as NAD hydride dehydrogenase. Mitochondrial complex I deficiency is the most prevalent defect within the respiratory chain and causes several disorders that often lead to neurological deficits (Ehinger et al., 2016). Leigh syndrome is one such example, characterized by pathological lesions within the CNS (reviewed by Bakare et al., 2021). Succinate supplementation can ensure reliable ATP synthesis, helping to prevent cell death when mitochondrial complex I is inhibited (Giorgi-Coll et al., 2017). In contrast, as discussed previously, oxidative stress can induce excessive accumulation of intracellular succinate, exerting potential pro-inflammatory effects on the cells or tissues involved (reviewed by Salminen et al., 2016). Instances of ischemia/reperfusion in CNS tissues can cause oxidative cell damage prompted by succinate-driven ROS production (Huang et al., 2023). The role of succinate in neurodegenerative diseases, such as AD, has not been studied in detail; however, cell-membrane permeable succinate demonstrates anti-inflammatory activity in reactive microglia, suggesting it may act as a regulator of chronic neuroinflammation that is associated with neurodegenerative diseases (Wang et al., 2021). More research is needed to explore the role that succinate has in the development of neurodegenerative processes.

Conclusions

Succinate and SUCNR1 play important roles in inflammatory signaling, but continued research is needed to clarify this relationship. Previous reports indicate that extracellular succinate has DAMP-like activity in the periphery but whether it performs this role in the CNS is unclear. The current understanding of the SUCNR1 signaling cascade and its contribution to glial cell-mediated inflammatory responses is conflicting and incomplete. Cell-membrane permeable forms of succinate used in experimental studies, such as diethyl or ethylmethylhydroxypyridine succinate, act on glial cells both intra- and extracellularly, potentially regulating their immune response. Future studies should use cell-membrane impermeable derivatives of succinate, such as disodium succinate, which may clarify the effects of extracellular succinate on the neuroimmune responses.

Fumarate and Itaconate

Role in the tricarboxylic acid cycle

Some TCA metabolites other than succinate, such as fumarate and itaconate, serve as notable regulators of metabolic and immune processes. The oxidation of succinate by SDH in the TCA cycle forms fumarate, which is then converted into L-malate by fumarate hydratase (FH). This enzyme also plays a role in the cellular response to DNA damage in both eukaryotic and prokaryotic organisms, where it localizes to the nucleus and catalyzes the reverse-conversion of L-malate into fumarate to facilitate the repair of double-stranded DNA breaks (reviewed by Álvarez-González and Sierra, 2024). Itaconate is a product of cis-aconitate decarboxylation by immune-responsive gene 1, a key regulatory enzyme of immunometabolism in peripheral macrophages (Noe and Mitchell, 2019). Itaconate acts as a competitive inhibitor of SDH, preventing the continued oxidation of succinate to fumarate as well as SDH-mediated reverse electron transport (reviewed by Li et al., 2023). Fumarate and itaconate are involved in immune cell signaling in both the periphery and the CNS.

Cytosolic activity of fumarate

Fumarate exhibits anti-inflammatory activity when localized to the cytosol of peripheral macrophages and CNS immune cells. Dimethyl fumarate (DMF) is the cell membrane-permeable methyl ester of fumaric acid and is widely used for in vitro and in vivo studies. In primary murine bone marrow-derived peripheral macrophages, intracellular DMF deactivates the nuclear factor erythroid 2-related 2 (NRF2) inhibitor Kelch-like ECH-associated protein (KEAP)1 by a process known as succination, wherein fumarate interacts with thiol groups of cysteine residues on the protein (Hoyle et al., 2022). This allows for the nuclear localization of NRF2, subsequently preventing NF-κB-dependent activation of the NLRP3 inflammasome and reducing the secretion of pro-inflammatory cytokines IL-1β and IL-6 (Hoyle et al., 2022). DMF pre-treatment also directly inhibits the activation of NLRP3 by the antibiotic nigericin in an NRF2-independent manner. These findings identify fumarate as an anti-inflammatory mediator of the immune response in peripheral cells. However, as discussed previously with succinate, the stimulation of primary murine macrophages with LPS causes glutamine-dependent anaplerosis, leading to the accumulation of fumarate in the cytosol due to the upregulated cytosolic aspartate–argininosuccinate shunt, which produces fumarate as a by-product (Hooftman et al., 2023). FH activity contributes to anti-inflammatory processes, as its inhibition in primary murine and human macrophages decreases the secretion of IL-10 and increases TNF production when cells are stimulated with LPS. This effect is also replicated by increasing cytosolic fumarate concentrations using DMF. Inhibition of FH enzymatic activity, but not treatment with DMF, also decreases the mtDNA-dependent production of the immunomodulatory cytokine IFNβ (Hooftman et al., 2023). These findings support the anti-inflammatory role of FH by regulating cytokine production in macrophages, although they suggest that increased accumulation of fumarate may result in the shift of macrophages to a pro-inflammatory phenotype. Previous research has also found that similar to succinate, fumarate suppresses prolyl hydroxylase activity to stabilize HIF-1α, a pro-inflammatory mediator, although the role of fumarate in this process may not be as prominent as other molecules including succinate and NO (reviewed by Knight and Stanley, 2019). Overall, the available evidence points to cytosolic fumarate as mainly an anti-inflammatory mediator, but more research is needed to determine whether under specific circumstances it can display pro-inflammatory activity in the periphery.

In primary murine microglia and astrocytes, DMF inhibits the activation of NLRP3 and reduces the secretion of IL-1β by degrading KEAP1 and activating NRF2 (reviewed by Hoyle et al., 2022; Rosito et al., 2020). NRF2 can also be activated by a KEAP1-independent pathway in both primary murine microglia and astrocytes. In this case, DMF facilitates the phosphorylation of glycogen synthase kinase-3β, triggering the PI3K/AKT pathway and promoting NRF2 activity (Cuadrado et al., 2018). However, it is unknown whether FH functions as an anti-inflammatory mediator in the CNS. Evidently, fumarate operates as a relevant anti-inflammatory molecule in microglia and astrocytes, but further research is needed to determine if it can also function in a pro-inflammatory manner.

Cytosolic activity of itaconate

Itaconate acts as an anti-inflammatory mediator in the periphery and the glial cells of the CNS. Immune-responsive gene 1 is highly upregulated in reactive macrophages, leading to increased itaconate concentrations and inhibition of SDH activity (reviewed by Li et al., 2023). Primary peripheral murine macrophages pre-treated with the cell membrane-permeable dimethyl itaconate (DMI) accumulate intracellular succinate and produce lower levels of ROS after stimulation with LPS compared to cells not exposed to DMI (Lampropoulou et al., 2016). Similar to fumarate, DMI and another cell membrane-permeable derivative, 4-octyl itaconate, activate NRF2 in a KEAP1-dependent manner to inhibit NLRP3 assembly and also directly inhibit NLRP3, reducing cytokine production in primary murine peripheral macrophages (Hoyle et al., 2022). Additionally, derivatives of itaconate modulate other intracellular pathways such as the Janus kinase 1/signal transducer and activator of transcription 6, tet methylcytosine dioxygenase 2, and transcription factor EB to reduce inflammation (reviewed by Shi et al., 2022).

Within the CNS, itaconate is a well-recognized anti-inflammatory molecule that regulates the neuroimmune functions of microglia and astrocytes. Pre-treatment with DMI upregulates NRF2 and inhibits LPS- and ATP-induced NLRP3 activity, significantly lowering IL-1β secretion in BV-2 murine microglia and primary murine microglia. This is accompanied by a decrease in ROS production due to lower SDH activity (Yang et al., 2021). Additionally, conditioned media from 4-octyl itaconate-treated BV-2 microglia exhibit neuroprotective effects on the murine Neuro-2a neural crest-derived cell line treated with the neurotoxins rotenone and MPP + (Xia et al., 2023). Both these substances induce dopaminergic neurodegeneration similar to that observed in PD; therefore, itaconate and its derivatives could serve as anti-inflammatory and neuroprotective mediators regulating neuroinflammation in this disease. In an in vivo model of spinal cord inflammation caused by subcutaneous injections of complete Freund’s adjuvant into the left hind paw, intraperitoneal administration of DMI reduces microglial activation in the spinal cord assessed by immunostaining for the microglial marker IBA-1. A decrease in spinal cord IL-1β and TNF mRNA expression and increased IL-10 mRNA levels are observed, with these changes being measured using real-time PCR (Lin et al., 2022). The effects of itaconate derivatives on primary murine astrocytes are similar to those on microglia, inhibiting SDH and modulating NLRP3 assembly by activating NRF2 (Cordes et al., 2020; Khadem et al., 2022). These studies collectively define itaconate and its derivatives as neuroprotective and anti-inflammatory molecules in the CNS, reducing the pro-inflammatory responses of glial cells to stimulants.

Extracellular damage-associated molecular pattern-like activity in the periphery and the brain

Both fumarate and itaconate can be secreted or released into the extracellular space, but their extracellular roles in immune cell signaling are only beginning to emerge. Fumarate accumulates in reactive macrophages due to the aspartate–argininosuccinate shunt and can be released upon cell death (reviewed by Liu et al., 2021). Itaconate is also secreted from reactive primary murine macrophages into the extracellular space (Lampropoulou et al., 2016). Other forms of fumarate, such as monomethyl fumarate (MMF) which crosses the BBB, are shown to have anti-inflammatory properties. Exposure of LPS-stimulated N9 murine microglia to extracellular MMF leads to an eight-fold decrease in TNF secretion and halved production of IL-1β. Both these effects are blocked by co-administration of antibodies against hydroxycarboxylic acid receptor (HCAR) 2 (Parodi et al., 2015). This G-protein coupled receptor, which is localized on the plasma membrane, activates AMPK/sirtuin-1 resulting in increased [Ca 2+ ] i , and inhibition of NF-κB. In addition to its HCAR2-mediated effects, MMF could reduce inflammation by acting on intracellular pathways due to its ability to freely cross membranes, but this hypothesis will need to be tested experimentally. In the case of itaconate, it is unknown whether this metabolite participates in extracellular signaling since its commonly used derivatives are cell membrane-permeable making their extracellular actions difficult to evaluate. Ultimately, more research is needed on the extracellular effects of fumarate and itaconate in the CNS.

Association with brain pathologies

DMF is a well-established drug used for the treatment of several autoimmune disorders, including psoriasis and relapsing forms of MS (reviewed by Saidu et al., 2019). The previously mentioned ability of DMF to activate NRF2 and inhibit NLRP3 is beneficial in the latter condition, where it has been shown to reduce the progression of brain lesions. DMF may also be protective in neurodegenerative diseases due to its anti-inflammatory activity (reviewed by Rosito et al., 2020), as a recent in vivo study using AD model mice confirms its effectiveness in reducing disease-associated neuroinflammation (Sun et al., 2022; Xia et al., 2023). Derivatives of itaconate are not currently recognized as viable treatment options in neuropathologies, but a series of recent studies have identified it as a potential candidate for the treatment of PD, as previously discussed (reviewed by Kong et al., 2024).

Conclusions

Intracellular signaling by fumarate and itaconate leads to anti-neuroinflammatory effects in the CNS. However, it is not known whether these TCA metabolites have extracellular DAMP-like activity. Although MMF is shown to act on the plasma membrane receptor HCAR2 in microglia, it is unknown if fumarate facilitates immune cell signaling by extracellular means. This is also the case with itaconate, as it is not clear whether the potential exogenous release of this molecule leads to its interaction with extracellular receptors on glial cells. Continued research using cell membrane-impermeable derivatives of these metabolites may shed new light on their potential extracellular DAMP-like effects. Additionally, other TCA metabolites, such as α-ketoglutarate, citrate, and malate have all been studied in the context of peripheral immune signaling and inflammation (reviewed by Noe and Mitchell, 2019); therefore, comprehensive studies involving glial cells and all key TCA metabolites may enhance the understanding of the interaction between mitochondrial metabolism and inflammatory processes in the CNS driven by these potential mtDAMPs.

Summary

Chronic neuroinflammation plays a central role in the progression of neurodegenerative diseases. Mitochondrial dysfunction is a common feature in many of these CNS pathologies, leading to the release of mitochondria and their components into the extracellular space. As a result, mtDAMPs acting on various PRRs of microglia ( Figure 3 ) and astrocytes ( Figure 4 ), and their partially overlapping intracellular signaling pathways likely have a significant impact on the persistent glia-driven immune response in these diseases. While some mtDAMPs, such as hemin, CytC, ATP, and TFAM, have been thoroughly investigated and are already identified as key players in glia-mediated pro-inflammatory processes, research on the CNS effects of others, namely CL, mtDNA, NFPs, succinate, fumarate, and itaconate, remains limited. In addition to outlining the already identified pro- and anti-inflammatory effects of known mtDAMPs, this review at the end of each section highlights areas for future research that may uncover crucial additional insights into the roles these molecules play in neuroinflammatory responses. An additional overarching observation that can be drawn is a shortage of mtDAMPs-focused studies using human glial cells and human tissue models. Given the increasing recognition of interspecies differences in glial cell pathophysiology and neuroimmune mechanisms (reviewed by Geirsdottir et al., 2019; Cuni-Lopez et al., 2022; Falcone, 2022), studies using human-specific experimental models such as induced pluripotent stem cell-derived glial cell cultures or brain organoids should be prioritized when exploring various biological effects of mtDAMPs. Determining the contributions of these molecules to disease processes and the mechanisms by which they operate could enhance our understanding of neuroinflammation and could potentially lead to the development of effective treatments for neurodegenerative disorders.

Microglial PRRs and associated signaling pathways activated by mitochondrial damage–associated molecular patterns.

Arrowheads indicate activation and blunt ends indicate inhibition. Receptors, signaling molecules, and pathways that have not been identified in microglia but are present in astrocytes are depicted in a lighter or transparent style. Created with BioRender.com. AGER: Advanced glycation end-product receptor; AMPK: AMP-activated protein kinase; ATP: adenosine triphosphate; cGAS: cyclic GMP-AMP synthase; CytC: cytochrome c; ERK1/2: extracellular signal-regulated kinase 1/2; FPR1: formyl peptide receptor 1; HCAR2: hydroxycarboxylic acid receptor 2; IP3: inositol triphosphate; JNK: c-Jun N-terminal kinases; KEAP1: kelch-like ECH-associated protein 1; Mac-1: macrophage antigen complex 1; MAPK: mitogen-activated protein kinase; MMP: matrix metalloproteinase; mtDNA: mitochondrial DNA; MyD88: myeloid differentiation factor 88; NFPs: n-formyl peptides; NF-κB: nuclear factor κB; NLRP: nucleotide-binding domain, leucine-rich–containing family, pyrin domain-containing; NRF2: nuclear factor erythroid 2-related 2; PLC: phospholipase C; PRRs: pattern recognition receptors; ROS: reactive oxygen species; STING: stimulator of interferon genes; SUCNR1: succinate receptor 1; TFAM: mitochondrial transcription factor A; TLR: toll-like receptor; TRIF: toll/interleukin-1 receptor-domain-containing adapter-inducing interferon-β; uCL: unsaturated cardiolipin; VEGF: vascular endothelial growth factor.

Astrocyte PRRs and associated signaling pathways activated by mitochondrial damage-associated molecular patterns.

Arrowheads indicate activation and blunt ends indicate inhibition. Receptors, signaling molecules, and pathways that have not been identified in astrocytes but are present in microglia are depicted in a lighter or transparent style. Created with BioRender.com. AGER: advanced glycation end-product receptor; AMPK: AMP-activated protein kinase; ATP: adenosine triphosphate; cGAS: cyclic GMP-AMP synthase; CytC: cytochrome c; ERK1/2: extracellular signal-regulated kinase 1/2; FPR1: formyl peptide receptor 1; HCAR2: hydroxycarboxylic acid receptor 2; IP3: inositol triphosphate; JNK: c-Jun N-terminal kinases; KEAP1: kelch-like ECH-associated protein 1; Mac-1: macrophage antigen complex 1; MAPK: mitogen-activated protein kinase; MMP: matrix metalloproteinase; mtDNA: mitochondrial DNA; MyD88: myeloid differentiation factor 88; NFPs: n-formyl peptides; NF-κB: nuclear factor κB; NLRP: nucleotide-binding domain, leucine-rich–containing family, pyrin domain-containing; NRF2: nuclear factor erythroid 2-related 2; PLC: phospholipase C; PRRs: pattern recognition receptors; ROS: reactive oxygen species; STING: stimulator of interferon genes; SUCNR1: succinate receptor 1; TFAM: mitochondrial transcription factor A; TLR: toll-like receptor; TRIF: toll/interleukin-1 receptor-domain-containing adapter-inducing interferon-β; uCL: unsaturated cardiolipin; VEGF: vascular endothelial growth factor.

Acknowledgments:

The authors would like to thank all members of the Laboratory of Cellular and Molecular Pharmacology for the helpful discussions and comments on the manuscript.

Funding Statement

Funding:
This work was supported by grants from the Jack Brown and Family Alzheimer’s Disease Research Foundation, the Natural Sciences and Engineering Research Council of Canada (No. 2020-04407), and the University of British Columbia Okanagan Campus. The authors also thank Gentmed SIA for financial assistance.

Footnotes

Conflicts of interest:
The authors declare no conflicts of interest. No conflicts of interest exist between Gentmed SIA and publication of this paper.

C-Editors: Zhao M, Sun Y, Qiu Y; T-Editor: Zou JP

Data availability statement:

Not applicable .

References

Associated Data

Data Availability Statement

Not applicable .