Role of TREM2 in neuroinflammation regulation: mechanisms, disease associations, and therapeutic translation advances

Abstract

Background

Neuroinflammation underlies the pathogenesis of neurodegenerative diseases, CNS trauma, aging, and pain disorders, profoundly influencing disease progression and clinical outcomes. TREM2 is an innate immune transmembrane receptor predominantly expressed on microglia in the CNS, primarily for sensing tissue damage and coordinating immune remodeling.

Main body

TREM2, by binding to its ligands and subsequently activating DAP12/10-mediated signaling pathways, coordinates microglial phagocytosis, metabolism, phenotype polarization, and inflammatory responses, thereby regulating neuroinflammation in various neurological diseases. Contrary to the conventional view of its neuroinflammatory suppression and neuroprotection, growing evidence has suggested that TREM2 exhibits dual functional roles that correlate with disease context and stage. This review provides a detailed overview of TREM2, including its structural features and relevant ligands, and elucidates the pathological mechanisms underlying TREM2-mediated regulation at both molecular and disease-specific levels—with particular focus on its effects on neuroinflammation through the modulation of microglial function. Finally, this review also discusses the current advances in TREM2-oriented therapeutic strategies and clinical translation, highlighting their potential applications for neuroinflammatory diseases.

Conclusions

TREM2 acts as a central regulator of neuroinflammation by modulating microglial function, exhibiting context-dependent neuroprotective or pathological roles across neurological disorders. Current TREM2-targeted therapies show translational potential for neuroinflammatory diseases, while sTREM2 holds value as a clinical biomarker.

Introduction

Neuroinflammation refers to an inflammatory response within neural cells and their microenvironment in the central/peripheral nervous system. It is characterized by glial cell activation, release of pro-inflammatory cytokine, disruption of the blood-brain barrier (BBB), and infiltration of peripheral immune cells—all of which collectively drive the initiation, progression, and clinical outcomes [1, 2]. Emerging evidence highlights neuroinflammation as a common denominator across neurodegenerative disorders (e.g., Alzheimer’s disease [AD], Parkinson’s disease [PD]), central nervous system (CNS) trauma (including ischemic/hemorrhagic stroke, traumatic brain injury [TBI]), infections of the nervous system, and chronic pain. This shared feature exerts profound impacts on disease progression and long-term prognosis [3, 4]. Therefore, alleviating neuroinflammation is widely recognized as a promising therapeutic strategy for mitigating symptoms and delaying the progression of these neurological disorders.

The triggering receptor expressed on myeloid cells 2 (TREM2) has emerged as a focal point in contemporary neuroscience research. As a key immune regulator predominantly expressed on brain microglia, TREM2 occupies a central role in mediating core microglial functions, encompassing cell migration, phagocytosis, metabolic homeostasis regulation, and transduction of inflammatory signals [5–7]. Accumulating studies have demonstrated that TREM2 exerts critical effects on the pathogenesis of neurological disorders: pathogenic mutations in TREM2 impair its normal function, while reactive upregulation of TREM2 modulates its protective capacities—both scenarios ultimately shaping disease progression [8, 9]. Notably, TREM2 expression directly governs microglial reactivity; this, in turn, influences the dynamic progression of neuroinflammation and the ultimate outcomes of neurological diseases [10, 11]. However, the precise molecular mechanisms underlying the interplay between TREM2 and neuroinflammation remain incompletely elucidated, necessitating further investigation.

This review aims to comprehensively synthesize the current body of knowledge regarding TREM2’s role in regulating microglia-mediated neuroinflammation, elucidate its molecular mechanisms, and delineate its functions across diverse neurological conditions. A theoretical framework is being sought to facilitate the development of TREM2-targeted therapeutic strategies for the treatment of neuroinflammatory diseases.

Structure, characteristics, and regulatory mechanisms of TREM2

TREM2 is a transmembrane immunomodulatory receptor consisting of 230 amino acids (aa) with a molecular weight of 26 kDa. Its structure primarily comprises an extracellular domain (aa 1–172), a transmembrane domain (aa 173–195), and a short cytoplasmic tail (aa 196–230), with its extracellular domain rich in cysteine residues capable of binding to multiple ligands [12] (Fig. 1). Physiologically, TREM2 is primarily expressed in peripheral myeloid cells, including macrophages, dendritic cells, and granulocytes; within the CNS, however, its expression is nearly restricted exclusively to microglia [13]. Under specific pathological stimuli—such as oxygen-glucose deprivation (OGD), lipopolysaccharide (LPS), or high mobility group protein box 1 (HMGB-1)—TREM2 can also be induced in astrocytes [14]. TREM2 interacts with a broad spectrum of endogenous and exogenous ligands, including anionic molecules, neurotoxic proteins, apolipoproteins, as well as certain bacteria or bacterial components [15], as summarized in Table 1. These ligand-receptor interactions reveal TREM2’s pivotal role in regulating myeloid cell functions and mediating diverse biological processes.

Fig. 1.

TREM2 structure and expression regulation. TREM2 consists of 230 amino acids and comprises four structural domains, and can be activated by a variety of ligands, such as Aβ, APOE, and LPS. With the assistance of DAP12, TREM2 recruits and activates SYK intracellularly, triggering multiple downstream signaling cascades. The expression of TREM2 can be regulated from several aspects. Proteases, such as ADAM10/17, can cleave TREM2 at the stalk region, leading to the production of soluble TREM2 (sTREM2) and a subsequent reduction in membrane-bound TREM2 levels. Mutations at specific amino acid residues of TREM2 will severely affect its function. Despite minimal changes in expression levels, such mutations significantly weaken the binding ability of TREM2 to ligands, thereby compromising TREM2-mediated downstream signaling. Finally, it can be regulated at the transcriptional level. Changes in methylation levels at different sites will upregulate or downregulate TREM2 expression. TrKB-CREB1, mTOR, and multiple microRNAs also influence the expression level of TREM2

Table 1.

Summary of TREM2 ligands and corresponding receptor cells

Ligands	Receptor cells	Refs
Anionic molecules
Phospholipids & Sulfolipids	Microglia, Apoptotic Neuro2a cells, NFAT-GFP-2B4 cells	[16–20]
Nucleic acid	Microglia	[21]
Sulfavant A	Dendritic cells	[22]
Dextran sulfate	TREM2-transfected BWZ cells	[23]
Heparan sulfate	Microglia	[24]
Mammalian proteins
Aβ	Microglia	[25]
Apolipoproteins（A/B/E/J）	Microglia	[26, 27]
TDP-43	Microglia	[28]
S1P	Microglia	[29]
C1q	Microglia	[30]
IL-34	AML cells	[31]
IL-4	BMDMs	[32]
HSP60	N9 cells	[33]
Galectin-3	Microglia	[34]
Cyclophilin A	RAW264.7 cells	[35]
Plexin-A1	COS7 cells	[36]
TLT-1s	Osteoclast precursor cells	[37]
Bacteria/bacterial components
LPS	TREM2-transfected BWZ cells	[23]
Bacterias	BMDMs, BWZ cells	[23, 38, 39]
Peptidoglycan	TREM2-transfected BWZ cells	[23]
Lipoteichoic acid	TREM2-transfected BWZ cells	[23]
Cholera toxin B	NFAT-GFP-2B4 cells	[40]
Mammalian Cells
Apoptotic cells	Microglia	[17, 41]
Astrocytes	Microglia	[33]
BMDMs	BMDMs	[42]
BMDCs	Dendritic cells	[43]
Small molecules
Hecubine	BV2 cells	[44]

Abbreviations: Aβ, β-amyloid, TDP-43, TAR-DNA binding protein 43 kDa, S1P, Sphingosine-1-phosphate, IL-34, Interleukin-34, IL-4, Interleukin-4, HSP60, Heat shock protein 60, TLT-1s, The alternative transcripts of triggering receptor expressed on myeloid cell-like transcript-1, LPS, Lipopolysaccharide

Owing to the lack of signal motifs in its short cytoplasmic tail, TREM2 heavily relies on binding to the intracellular adaptor protein DNAX activator protein 12/10 (DAP12/10) for signal transduction. DAP12/10 contains an immunoreceptor tyrosine-based activation motif (ITAM) domain, which forms docking sites following tyrosine phosphorylation. Upon interaction of TREM2 with its ligands and resulting activation, the ITAM motif is phosphorylated by the members of SRC kinase family, initiating downstream signaling cascades and subsequent immune responses [45, 46].

TREM2 expression is dynamically modulated under neuroinflammation conditions (Fig. 1): it is upregulated in AD, PD, multiple sclerosis (MS), ischemic stroke, intracerebral hemorrhage (ICH), TBI, and neuropathic pain [21, 47–63], while downregulated in spinal cord injury (SCI), postoperative cognitive dysfunction (POCD), and in vitro neuroinflammation models [64–69]. Despite these variations across diseases, TREM2 consistently engages in pathological processes such as phagocytosis of neurotoxic proteins, inhibition of neuroinflammation, and protection of cognitive function. TREM2 expression is primarily modulated at the transcriptional and epigenetic levels. In AD patients, for example, the methylation status of TREM2 is significantly correlated with its expression level [70]. Specifically, hypomethylation at site cg25748868 and intron 1 of TREM2, along with enriched methylation at the transcription start site, particularly 5-hydroxymethylcytosine (5hmC), collectively upregulate TREM2 expression [70–73]. Additionally, activation of signaling pathways such as TrKB-CREB1, cGAMP-STING-IRF3, and mTOR also promotes TREM2 transcription [74–77]. Conversely, microRNAs (miRNAs), including miR-3473b, miR-665, miR-130a-3p, and miR-34a, inhibit TREM2 expression, potentially via negative regulation of the nuclear factor-κB (NF-κB) pathway [78–81]. Small molecules [82–93] and protein receptors [94–97] also exert regulatory effects on TREM2, though the mechanisms remain undefined.

Second, TREM2 expression is eliminated by the pro-inflammatory signaling triggered by LPS or IFN-γ [98, 99]. Multiple hydrolytic enzymes, such as α-secretase disintegrin and metalloproteinase domain-containing protein 17 (ADAM17) and ADAM10, can cleave TREM2 at the H157–S158 peptide bond within its extracellular domain, releasing soluble TREM2 (sTREM2) and subsequently decreasing TREM2 levels [100, 101]. In contrast, the extracellular domain of integral type II transmembrane protein (BRI2) can bind directly to TREM2, preventing the cleavage of the α-secretase-independent pathway, thereby enhancing steady-state levels of membrane surface TREM2 [102]. Finally, since microglia require intact TREM2 to respond to pathological stimuli, TREM2 mutations significantly impair microglial function [103]. Despite unaltered expression levels, mutations at the relevant amino acid sites, such as R47H, R62H, D87N, and Q33X, markedly reduce TREM2-ligand binding affinity, compromising the efficiency of receptor-mediated signaling and exacerbating the progression of neurological diseases [104, 105]. For instance, the most common variant, R47H, fails to initiate TREM2/DAP12-mediated phosphorylated Spleen tyrosine kinase (pSYK) signaling, resulting in diminished downstream events, including reduced ERK1/2-Akt phosphorylation, impaired NOD-like receptor family pyrin domain containing 3 (NLRP3) inflammasome activation, blunted mTOR responses, and compromised NF-κB inhibition. Collectively, these mutations desensitize microglia to a variety of endogenous “danger” signals and environmental stimuli [106, 107].

In summary, the activation and regulation of TREM2 represent a complex process that depends primarily on the tissue microenvironment and intracellular state. Further investigation is needed to clarify how TREM2 interacts with its diverse ligands and mediates signaling pathways in damaged tissues under different neuroinflammatory conditions.

Microglial TREM2 in neuroinflammation

Neurodegenerative diseases

Neurodegenerative diseases represent a class of neurological disorders defined by the selective neuronal loss in the CNS, including AD, PD, MS, and amyotrophic lateral sclerosis (ALS). Neurotoxic protein aggregation and genetic variations are their common pathological features. The accumulation of various neurotoxic proteins, coupled with the release of damage-associated molecular patterns (DAMPs, e.g., ATP, HMGB1, S100B, DNA) triggered by subsequent neuronal death, collectively constitute co-stimulatory signals in the inflammatory response. Once recognized by pattern-recognition receptors (PRRs) on microglia, these signals induce pro-inflammatory cascades, ultimately establishing a sustained chronic inflammatory microenvironment within the CNS [3]. Therefore, despite differences in their underlying pathogenesis, all these neurodegenerative diseases share the key feature of chronic neuroinflammation.

Alzheimer’s disease (AD)

AD is pathologically distinguished by two hallmark lesions: extracellular senile plaques composed of amyloid β (Aβ) peptide and intracellular neurofibrillary tangles formed by misfolded tau protein. Both lesions serve as potent inducers of neuroinflammation [108]. Studies have shown that TREM2 activation drives microglial aggregation around Aβ plaques, forming a neuroprotective microglial barrier that facilitates the compaction of amyloid deposits and their isolation from surrounding healthy neural tissue. In the early stages of AD, TREM2 suppressed Aβ-induced tau accumulation and spreading, while enhancing microglial phagocytosis of Aβ and tau to mitigate neuroinflammation [109] (Fig. 2).

Fig. 2.

Function of TREM2 in AD across disease stages. In the early stage, TREM2 initiates protective inflammatory activation, where microglia establish a restrictive barrier against Aβ oligomers and tau, thereby sustaining neurological function. However, in the late stage, two distinct scenarios arise: one entails the continuous chronic activation of microglia, with concomitant deposition of Aβ plaques and neurofibrillary tangles (NFTs), ultimately leading to axonal dystrophy; the other is marked by a reduction of plaque-associated microglia, which promotes the seeding and spread of Aβ and pTau, exacerbating neural axon dystrophy

However, the contribution of TREM2 in late-onset AD remains controversial, with debates focusing on its effects on Aβ plaques, microglial activation, and neuronal function. Some studies have reported that germline knockout of TREM2 or the TREM2 R47H variant reduces the proliferation of microglia surrounding Aβ plaques, which in turn exacerbates seeding and spread of tau aggregates [110]. Meanwhile, TREM2 knockdown impaired microglial phagocytic capacity, promoting more severe Aβ deposition [111, 112]. These observations suggest that TREM2 may limit Aβ plaque aggregation and mitigate tau pathology in late-stage AD, with TREM2 mutations linked to an increased risk of late-onset AD. In contrast, other studies using late-stage AD mouse models have observed the opposite: TREM2 deficiency actually reduces Aβ plaques deposition and tau protein phosphorylation in the cerebral cortex and hippocampus, while also alleviating microglial activation associated with Aβ plaques. Notably, however, TREM2 deficiency simultaneously altered the morphology and composition of Aβ plaques, exacerbated Aβ-induced neurotoxicity, and failed to halt the progression of late-onset AD [113, 114]. Overall, these conflicting findings highlight the paradoxical effects of TREM2 on microglial responses to Aβ and tau in late-onset AD. Throughout the entire progression of AD, neither TREM2 deficiency nor the R47H variant has been shown to alleviate neuroinflammation or restore cognitive function. Despite its context-dependent roles across different disease stages, TREM2 mediates time-dependent neuroprotection effects, with distinct functional impacts in early versus late AD (Fig. 2).

At the molecular level, TREM2 downregulates microglial Toll-like receptors (TLRs: TLR2, TLR4, TLR6), suppressing downstream NF-κB and mitogen-activated protein kinase (MAPK) signaling. This shifts the cytokine balance by reducing pro-inflammatory (IL-1β, TNF-α, IL-6) and elevating anti-inflammatory (IL-4, IL-10, IL-13) factors, thereby mitigating neuroinflammation [84, 90, 115]. Additionally, ligand-mediated TREM2 activation triggered immune responses through the TREM2/DAP12 complex. Specifically, SYK-PI3K-Akt-GSK-3b-mTOR signaling axis mediated both the transcriptional phenotypic conversion of “disease-associated microglia (DAM)” and the inhibition of tau protein hyperphosphorylation, effectively alleviating neuroinflammation in AD mice [10, 116]. Overexpression of TREM2 also regulated the microglial polarization phenotype (M1→M2) through either the PI3K/Akt/FoxO3a or JAK/STAT/SOCS signaling pathway, thereby inhibiting excessive inflammation [117, 118] (Fig. 3). It has been revealed that LPS/Aβ42 induced SYK-mediated inactivation of AMP-activated protein kinase (AMPK) and Akt activation, leading to mitochondrial fragmentation and subsequent activation of the NLRP3 inflammasome. TREM2 counteracted this pathological cascade by enhancing AMPK activation and inhibiting Akt phosphorylation, which in turn suppressed NLRP3 inflammasome activity [119, 120]. Overall, TREM2 coordinates microglial functions through multiple distinct signaling pathways, and activation or upregulation of TREM2 may promote pathological recovery in AD.

Fig. 3.

TREM2-mediated signaling network and regulation of neuroinflammation. Upon the activation of microglia, TLRs trigger the activation of MAPK and NF-κB via MyD88-dependent and -independent pathways. This leads to excessive microglial activation and the release of proinflammatory cytokines, thereby contributing substantially to the pathological progression of neuroinflammation. Following ligand binding, TREM2/DAP12 complexes mediate multiple SYK-centered signaling cascades, including PI3K-Akt, JAK-STAT, and PLCγ2. On one hand, TREM2/DAP12 directly restrains the overactivation of NF-κB/MAPK proinflammatory pathways by downregulating the activity of key molecules such as TRAF6 and IRAK3. On the other hand, by activating signaling pathways like mTOR and LXR, this receptor-adaptor complexs drive microglial polarization toward M2 or disease-associated microglia (DAM) phenotypes. This polarization, in turn, boosts microglial phagocytic ability and induces lipid metabolic reprogramming. Ultimately, this TREM2-mediated signaling network enables precise regulation of the neuroinflammatory microenvironment, mitigating neuronal damage and functional deficits

Parkinson’s disease (PD)

PD is a common neurodegenerative disorder characterized by progressive degeneration of dopaminergic neurons, Lewy body deposition, and local microglia activation in the substantia nigra. Neuroinflammation occupies a central role in PD pathogenesis, with microglial activation as a key mediator. Excessive activation leads to the release of pro-inflammatory cytokines, which in turn damage dopaminergic neurons and further drive PD progression [121]. Crucially, TREM2 counteracts neuroinflammation via multiple mechanisms. First, in 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine (MPTP)-induced PD mouse models, TREM2 suppressed the activation of NF-κB and MAPK signaling pathways, which are normally mediated by the TLR4/tumor necrosis factor receptor-associated factor 6 (TRAF6) axis [49]. Second, TREM2 inhibited NLRP3 inflammasome activation via the TLR4/myeloid differentiation primary response 88 (MyD88)/NF-κB axis, significantly reducing levels of pro-inflammatory markers (TNF-α, IL-1β, cyclooxygenase-2 [COX-2], and inducible nitric oxide synthase [iNOS]) and also dampening pro-inflammatory polarization of both microglia and astrocytes [48, 122]. Third, TREM2 activates autophagy by inhibiting the p38 MAPK/mammalian target of rapamycin (mTOR) pathway, thereby decreasing the accumulation of pathological aggregates like α-synuclein and protecting dopaminergic neurons from neuroinflammation-induced injury [123] (Fig. 3). In conclusion, TREM2 suppresses multiple overlapping neuroinflammatory pathways, thus ameliorating the key pathological features of PD.

Multiple sclerosis (MS)

MS is an inflammatory, demyelinating, and neurodegenerative disease of the CNS initiated by an autoimmune attack. Its key pathological features include multifocal demyelinating plaques in the white matter, accompanied by inflammatory cell infiltration, oligodendrocyte damage, and axonal lesions [124]. Neuroinflammation driven by persistent myelin debris accumulation [125] is mitigated through TREM2-dependent microglial phagocytosis, which clears myelin fragments and associated lipids to enable remyelination and neuroprotection [126, 127]. In the demyelinating mouse model, TREM2 deficiency both the activation of microglia and disrupts the phagocytic degradation of myelin debris, concurrently resulting in blocked cholesterol efflux from microglia and the intracellular accumulation of cholesteryl/oxidized cholesteryl esters [128, 129] (Fig. 3). These functional deficits may stem from TREM2-mediated transcriptional dysregulation of key effectors involved in phagocytosis and lipid metabolism (Lpl, Cd36, Axl, Cst7) [60, 130].

Postoperative cognitive dysfunction (POCD)

POCD is a common geriatric neurodegenerative disorder following major surgery and anesthesia, primarily attributed to the dysregulation between perioperative acute stress-induced inflammatory response and the inherent vulnerability. Neuroinflammation is widely recognized as the predominant pathological mechanism underlying POCD pathogenesis. Substantial studies have highlighted the crucial role of TREM2 in modulating neuroinflammation during POCD development. Specifically, aged mice subjected to surgery procedures exhibit downregulated TREM2 expression, whereas TREM2 overexpression ameliorates both neuroinflammatory responses and surgery-induced cognitive impairment [66, 67]. Mechanistically, TREM2 exerts anti-inflammatory and cognitive-protective effects through activating the PI3K-Akt signaling pathway, simultaneously suppressing NF-κB and NLRP3 inflammasome activity, and upregulating synaptic-associated proteins, such as brain-derived neurotrophic factor (BDNF), N-methyl-D-aspartate receptor 2B (NR2B), postsynaptic density protein 95 (PSD95), glial cell line-derived neurotrophic factor (GDNF), and synaptophysin [65–67, 131].

In summary, across a spectrum of neurodegenerative diseases, microglial TREM2 can regulate the initiation, progression, and resolution of neuroinflammation through multiple signaling pathways and cascades, thereby ameliorating pathological features and functional deficits associated with these neurological disorders.

CNS trauma

CNS trauma refers to injuries of the CNS caused by external mechanical forces or internal pathological processes, including ischemic stroke, ICH, subarachnoid hemorrhage (SAH), TBI, and SCI. Despite varying primary etiologies, cellular ischemia, hypoxia, and pathological damage consistently trigger sustained neuroinflammation, thus exacerbating neuronal death and dysfunction—particularly in ischemic stroke and TBI. Immune regulatory molecules like TREM2 play a critical role in modulating the progression of these injuries by regulating microglial function and behavior.

Ischemic stroke

The main characteristic of ischemic stroke and cerebral ischemia-reperfusion injury is the immediate cessation of cerebral blood flow, causing hypoxia and glucose deprivation that damages multiple cell types within the ischemic region. Despite reperfusion, tissue damage persists and can even be exacerbated. This pathophysiological process involves complex interconnected mechanisms, including metabolic reprogramming and neuroinflammation, mediated by numerous signaling pathways [132]. Crucially, TREM2, a key regulator of microglial phenotype and function, exerts potent anti-inflammatory and neuroprotective effects in both ischemic stroke and cerebral ischemia-reperfusion injury. In animal models of both diseases, TREM2 overexpression significantly suppressed inflammatory responses and neuronal apoptosis in the ischemic regions. Mechanistically, TREM2 activation promoted a microglial shift from M1 to M2 phenotype, reducing local pro-inflammatory cytokine levels, and enhanced microglial phagocytic activity by specifically binding to apoptotic neurons, thereby interrupting the neuroinflammatory positive feedback loop [29]. Conversely, TREM2 silencing exacerbated neuroinflammation, augmented neuronal apoptosis and infarct volume, and worsened neurological deficits [21, 53]. Importantly, TREM2 upregulation by baicalin, electroacupuncture, or exercise interventions mitigates disease progression and confers neuroprotection in preclinical models [82, 133–137].

Traumatic brain injury (TBI)

TBI is defined as brain damage caused by external mechanical forces, featuring an orderly spatiotemporal pathological progression involving both primary and secondary injuries. Primary injury stems from direct mechanical impact and is exacerbated by acute systemic complications such as hypotension, hypoxia. Meanwhile, mechanical stress triggers a series of biochemical cascades that drive secondary injury, including excitotoxicity induced by excessive glutamate release, production of reactive oxygen species (ROS), and robust neuroinflammatory responses [138]. Recent research has underscored the role of TREM2 in TBI pathogenesis. For instance, COG1410, a well-characterized TREM2 agonist, can upregulate and activate TREM2, thereby preventing BBB disruption, alleviating brain edema progression, and restoring cerebral blood flow. In TBI mouse models, TREM2 activation inhibited neuroinflammation and re-established microglial homeostasis via the Akt/CREB/BDNF pathway, which suppressed the levels of pro-inflammatory cytokines (TNF-α, IL-1β) and pro-apoptotic molecules (Bax [BCL2-associated X protein], cleaved caspase-3 fragment) [59]. Additionally, the anti-inflammatory activity of TREM2 mitigated surgically induced brain injury by blocking nuclear translocation of NF-κB p65 [139]. Notably, white matter injury is a pivotal pathological process in TBI that strongly influences long-term neurological outcomes [140], and its repair has been confirmed to be associated with TREM2 [141]. Specifically, upregulation of TREM2 promoted microglial phagocytosis and cholesterol metabolism via the DHCR24/LXR axis, which synergistically facilitates white matter repair and cognitive recovery in TBI mice [142] (Fig. 3).

The role of TREM2 in TBI pathogenesis remains controversial, with studies reporting conflicting effects on macrophage activation and disease progression across acute (3 days) and chronic (120 days) phases. Saber et al. observed that in the acute phase of TBI, TREM2 knockout (TREM2^−/−) C57BL/6 mice exhibited enhanced activation of macrophage and astrocyte at the cortical injury site. In the chronic phase, however, macrophage activation in TREM2^−/−mice normalized to levels comparable to wild-type mice. 3D brain reconstruction and behavioral analyses revealed reduced hippocampal atrophy and improved cognitive function in TREM2^−/−mice, suggesting that TREM2 deficiency may facilitate functional recovery during the chronic phase of TBI [143]. In contrast, Katsumo et al. found that hTau TREM2^−/−mice exhibited suppressed activation of macrophage/microglia at white matter injury sites in the acute phase, but exaggerated activation in the chronic phase. Mechanistic investigations indicated that TREM2 deficiency accelerated inflammation and neurodegeneration, accompanied by impaired microglial phagocytosis and persistent BBB leakage. These changes exacerbated TBI pathology and underscored a protective role of TREM2 [144]. Collectively, the role of TREM2 in TBI is highly complex and context-dependent, with discrepancies likely attributable to variations in animal models, research methodologies, outcome metrics, or observation time points. The full spectrum of TREM2 mechanisms in TBI remains unelucidated, which necessitates further investigation.

Cerebral hemorrhage

SAH and ICH are severe hemorrhagic cerebrovascular disorders sharing intricate pathological mechanisms. Both of them primarily arise from intracranial vascular lesions, including ruptured aneurysms, hypertensive vasculopathy, and cerebral vascular malformations, which lead to abnormal blood accumulation in the CNS. Blood accumulation and its degradation products activate resident immune cells (e.g., microglia) and recruit peripheral immune cells, triggering robust neuroinflammation and oxidative stress. This disrupts the BBB integrity, exacerbating cerebral edema and intracranial hypertension, and ultimately compressing the surrounding brain parenchyma, resulting in neuronal injury and neurological deficits. Importantly, neuroinflammation further amplifies this pathological damage, forming a vicious cycle that drives secondary brain injury. Emerging evidence has highlighted that TREM2 is intricately associated with both short-term and long-term neurological outcomes in SAH and ICH. In a SAH rat model, TREM2 overexpression upregulated the inflammatory negative regulator interleukin-1 receptor associated kinase-3 (IRAK3) and anti-inflammatory cytokine IL-10, promoted M2 microglial polarization, and reduced myeloperoxidase-positive (MPO⁺) cell infiltration and pro-inflammatory mediator release. All of them mitigated neuronal apoptosis and improved neurological deficits [55]. Notably, during the acute phase of SAH, the neuroprotective effect of TREM2 can be antagonized by the TLR4 pathway, indicating a complex interaction of the TLR4/TREM2 axis in SAH pathogenesis [56]. Additionally, upregulation of TREM2 has been shown to ameliorate long-term cognitive impairment and promote the recovery of hippocampal neural activity following SAH, which is primarily via the activation of PI3K/Akt signaling pathway [54] (Fig. 3).

In ICH, TREM2 also plays a critical regulatory role. Knockout of endogenous TREM2 exacerbated neurological deficits in both naive and ICH mice. The TREM2 agonist COG1410 significantly upregulated TREM2, which correlated with enhanced PI3K activity, elevated phosphorylated Akt, and increased B-cell lymphoma-2 (Bcl-2), while reducing the production of TNF-α, IL-1β, and Bax. In addition, COG1410 administration reduced microglial/macrophage activation and neutrophil infiltration, mitigated peri-hematomal cerebral edema, and suppressed neuronal apoptosis. These combined effects ultimately improved both short-term and long-term neurological outcomes in ICH models [54, 145]. Transcriptional regulators like nuclear factor erythroid 2-related factor 2 (Nrf2) modulated microglial function and inflammatory responses by regulating TREM2 expression. This TREM2-Nrf2 axis promoted hematoma clearance and neurological recovery [146] (Fig. 3). Collectively, TREM2 exerts neuroprotection in hemorrhagic cerebrovascular diseases through multi-faceted regulation of neuroinflammation and apoptotic pathways, making it a promising therapeutic target for nerve damage associated with hemorrhagic cerebrovascular disorders.

Spinal cord injury (SCI)

SCI initiates a complex pathological cascade involving three core processes: neuroinflammation, neuronal injury, and dysregulated immune homeostasis. Within this cascade, TREM2 exerts pleiotropic effects by modulating microglial function, neuroinflammatory responses, and neuronal synaptic plasticity. Studies have demonstrated that TREM2 promoted microglial M2 polarization via the apolipoprotein E (APOE)/TREM2/NF-κB signaling axis, thereby attenuating neuroinflammation and curbing SCI progression [64]. Besides, TREM2 facilitated functional recovery after SCI by regulating neuronal synaptic plasticity and axonal regeneration. In TREM2 knockout mice, SCI led to reduced microglial activation, decreased excitatory synaptic density, and increased inhibitory synaptic density. Concurrently, calcitonin gene-related peptide-positive (CGRP⁺) axon sprouting toward the dorsal horn was impaired, indicating that TREM2 deficiency disrupted microglia-mediated synaptic remodeling and hindered neural regeneration and functional recovery [147]. Interestingly, TREM2 plays dichotomous roles in immune regulation and autonomic nervous system function. TREM2 knockout ameliorated post-SCI immune suppression, mitigated splenic lymphocyte depletion, and significantly reduced the incidence of autonomic dysreflexia, while concomitantly improving cardiovascular abnormalities (e.g., hypertension and arrhythmia) [148]. These findings suggested that TREM2 deficiency may exert protective effects during specific pathological phases of SCI. This functional heterogeneity potentially stems from dynamic microenvironmental changes in SCI, such as divergent demands for TREM2 during early-phase inflammatory regulation versus late-stage neural repair. Additionally, TREM2 has been implicated in the MiR-665 regulatory pathway, but its precise mechanism and crosstalk with other signaling networks (such as TLR4 and PI3K/Akt) remained undefined [149].

Aging

As a defining hallmark of aging, neuroinflammation contributes to the pathogenesis of multiple age-related neurological diseases. Gene expression profiling studies have identified TREM2 as a positive regulator of cognitive integrity during aging. In late life, TREM2 expression was significantly upregulated at both the transcriptional and protein levels, ameliorating microglia-driven inflammatory milieu, reducing neuronal loss, and thereby delaying cognitive decline [61, 150]. Knockdown of TREM2 in SAMP8 mice elicited increased pro-inflammatory cytokines (TNF-α, IL-6) and decreased anti-inflammatory IL-10, accompanied by pronounced synaptic loss and overt cognitive deficits. These findings confirmed TREM2’s protective effects against age-related neuroinflammation and cognitive impairment [151]. Taurine, a semi-essential amino acid with anti-inflammatory, antioxidant, and neuroprotective properties, significantly upregulated TREM2 expression in the cortex and hippocampus of SAMP8 mice. This effect reduced the number of activated microglia, inhibited tau phosphorylation, and decreased Aβ deposition, thereby delaying the pathological progression of aging and neurodegenerative diseases [83]. Notably, comparative studies between young/elderly TREM2-deficient (TREM2^−/−) and wild-type mice have revealed that aged TREM2^−/− mice exhibited superior cognitive performance, enhanced long-term potentiation (LTP), and increased dendritic spine density with elevated synaptic markers. These differences were absent in young mice (regardless of TREM2 status) [152]. The results indicated that TREM2 deficiency modulated neuronal architecture to confer resistance against age-related synaptic and cognitive decline during non-pathogenic aging. This apparent discrepancy with prior results is likely attributed to differences in animal models (SAMP8 versus TREM2^−/−) and variations in research focus (pathological aging versus physiological aging).

Pain-associated disorders

Pain-associated disorders are a heterogeneous group of clinical syndromes with pain as the primary symptom, covering a variety of pathological conditions, including neuropathic pain (such as diabetes neuropathy, trigeminal neuralgia, and chemotherapy-induced peripheral neuropathy (CIPN), chronic pain (such as postoperative pain, cancer pain), and inflammatory pain (such as arthritis and lumbago). The pathogenesis commonly involves neuroinflammation, as well as peripheral and central sensitization, which synergistically initiate or exacerbate pain perception [153]. Recent investigations have identified TREM2 as a key regulator of neuroinflammation and pain sensitization in these disorders.

In animal models of nerve injury, TREM2/DAP12 is mainly expressed in microglia of the spinal dorsal horn, where it promotes pro-inflammatory cytokine secretion and exacerbates neuropathic pain. Administration of TREM2 agonistic antibodies or TREM2-overexpressing lentiviral vectors upregulated the production of inflammatory cytokines (TNF-α, IL-1β, IL-6) via the TREM2/DAP12 axis, ultimately inducing mechanical allodynia. In contrast, genetic deficiency of DAP12 significantly suppressed pro-inflammatory cytokine expression and attenuated pain-related behaviors [62, 154]. Similarly, in models of CIPN, TREM2/DAP12 signaling triggered neuroinflammation, which exacerbates cisplatin-induced mechanical pain sensitization, sensory deficits, and intraepidermal nerve fiber loss [155]. In short, these findings highlight the TREM2/DAP12 signaling pathways as a potential therapeutic target for neuropathic pain.

In a tibial fracture-induced chronic pain model, CCL21-dependent TREM2/DAP12 inflammatory signaling and resulting microglial activation mediated the initiation and maintenance of postoperative chronic pain. Pharmacological inhibition of either CCL21 or TREM2 remarkably ameliorated pain symptoms [63]. Likewise, in a femoral cancer pain model, TREM2/DAP12 signaling promoted microglial activation and mechanical hyperalgesia [94]. In a nitroglycerin-induced migraine model, the TREM2-SYK axis drove NLRP3 inflammasome activation in the trigeminal caudate nucleus, leading to neuroinflammation and central sensitization [156]. Unlike its role in neurodegenerative diseases, TREM2/DAP12 signaling exerts pathogenic significance in regulating neuroinflammation and pain sensitization across diverse pain-associated disorders. These findings not only provide a novel paradigm for understanding pain sensitization mechanisms but also lay a preclinical foundation for developing analgesics targeting the TREM2/DAP12 pathway.

TREM2-oriented therapeutic strategies and clinical translation

In recent years, TREM2 has emerged as a research hotspot owing to its pivotal role in regulating neuroinflammation. Intervention strategies targeting the TREM2 signaling pathway thus hold promise as potential therapeutic approaches for neuroinflammatory disorders. Given TREM2‘s neuroprotective functions in the CNS, activating its signaling pathway is considered a particularly effective therapeutic strategy.

TREM2-activating therapeutic agents

To date, various TREM2 agonistic antibodies have been developed, yielding certain therapeutic effects in preclinical and clinical trials. Zhao et al. screened a tetravalent TREM2 agonist antibody, which exhibited nearly 100-fold enhanced TREM2 activation efficiency, stronger Aβ phagocytic capacity, and effectively improved microglial survival [157]. To address efficient antibody transport across the BBB, TREM2 antibodies were conjugated with transferrin receptors (TfR). This engineered TfR-conjugated antibody achieved broader brain distribution and more robust TREM2-mediated signal transduction [158]. The teams of Wang and Long demonstrated, for the first time in a human phase I clinical trial, that the anti-human TREM2 agonist antibody AL002C was safe and well-tolerated in over 130 healthy subjects and patients with mild to moderate AD [159, 160]. A phase II clinical trial targeting patients with mild cognitive impairment and mild dementia due to AD is also evaluating the efficacy of the same antibody (NCT04592874). Another independently developed TREM2 agonistic antibody, DNL91940, has completed safety and tolerability testing in healthy adults (NCT04592874); however, no results from this study have been published yet (Fig. 4). Notably, current research and development of TREM2 agonistic antibodies are primarily focused on AD, but their efficacy and safety in other neurodegenerative diseases or CNS injuries remain unevaluated [161].

Fig. 4.

TREM2-oriented therapeutic strategies and clinical translation. This pie chart systematically illustrates a multidimensional translational framework centered on TREM2. (1) Agonists: diverse TREM2-targeting agents are utilized to activate or modulate TREM2-mediated function, such as small molecule drugs, antibodies, proteins, or peptides. They are undergoing rigorous clinical validation to evaluate their safety and efficacy, while exploring their application scope across the spectrum of diseases; (2) Nano-delivery systems: engineered nanoplatforms facilitate the targeted delivery of TREM2-associated therapeutic payloads, such as nucleic acids, small molecule agonists, antibodies, and peptide-based drugs. These systems prioritize precise localization to target tissues and ensuring biosafety, thereby overcoming the limitations of traditional administration methods (e.g., poor bioavailability, off-target effects); and (3) Prediction, diagnosis, and prognosis: sTREM2, the remaining extracellular fragment of TREM2 cleavage, is a key biomarker for immune cells, including microglia, and encapsulates critical biological information regarding systemic and tissue-specific immune activation. Plasma and CSF sTREM2 levels exhibit disease-specific dynamic fluctuations. For instance, elevated CSF sTREM2 correlates with an increased risk of early onset of ad. Quantitative detection of sTREM2 in these body fluids enables non-invasive prediction of disease risk and helps stratify prognosis by linking its expression to disease progression, thus providing actionable insights for clinical management

As discussed in Section “Structure, characteristics, and regulatory mechanisms of TREM2”, various small-molecule compounds can exert anti-inflammatory activity by activating TREM2 and related signaling pathways, thereby conferring neuroprotective benefits across multiple disease models. For example, neuroactive small molecules such as dihydroquercetin and edaravone have been proven to promote TREM2 signaling activation in microglia, reduce the expression of pro-inflammatory cytokines, and effectively alleviate neuroinflammation in AD or PD models [84, 162]. Furthermore, certain TREM2 agonists—including heat shock protein 60 (HSP60), COG1410 (an APOE-mimetic peptide), and IL-4—exerted similar effects [59, 65, 95]. However, a critical consideration is the dualistic nature of TREM2 activation: while protective in neurodegeneration and CNS trauma, its activation may exacerbate pathology in pain-related diseases and cancer [163]. The underlying mechanisms remain incompletely elucidated, and simply activating TREM2 signaling may not constitute a universal treatment for neurological disorders. Moreover, TREM2 activation-based therapeutic strategies are still in early stages, and their long-term efficacy and safety require further validation through additional clinical trials (Fig. 4).

TREM2-based delivery systems

The widespread peripheral expression of TREM2 poses inherent safety risks for systemic agonist administration, due to insufficient tissue specificity and brain targeting. In this context, TREM2-based nano-delivery systems represent promising therapeutic prospects owing to their non-invasiveness, targeted delivery, and low toxicity. The gene delivery strategy developed by Wang and Jiang’s team—using TREM2 complementary DNA (cDNA)—specifically upregulated TREM2 expression in microglia, and thus reshaped the AD inflammatory microenvironment by inhibiting M1 polarization of microglia and enhancing Aβ phagocytosis [164, 165]. Boudesco et al. identified a TREM2 liposomal agonist with excellent TREM2 selectivity, mimicking antibody-mediated TREM2 activation while offering superior safety and stability [166]. Quercetin-betaine nanoparticle systems were designed to overcome the bioavailability and BBB permeability limitations of small-molecule drugs, delaying brain disease progression via precise TREM2 signaling modulation [167, 168]. Li et al. designed TREM2-affinitive peptides (HLRKLRKR and LRKLRLRL) conjugated to polyethylene glycol liposomes, enabling targeted recognition and uptake by glioma cells, which provided new insights for the development of precision therapy for glioma [169]. TREM2-based chimeric antigen receptor T (CAR-T) therapy and engineered Escherichia coli vesicles have also shown promise beyond neurological disorders—specifically in colorectal cancer treatment [170, 171]. By engineering CEA-specific CAR-T cell linked to anti-PD-1-TREM2 single-chain antibodies (PD-1-TREM2 scFv), this approach reduced the number of tumor-associated macrophages (TAMs) and myeloid-derived suppressor cells (MDSCs) in the tumor microenvironment, while simultaneously enhancing CD8⁺ T cell infiltration and cytotoxicity, thereby improving antitumor efficacy [170]. Undoubtedly, these advancements highlight the broad, cross-disease therapeutic potential of TREM2-based nano-delivery systems, particularly in neurodegenerative diseases and cancer treatment, and offer new possibilities for neuroinflammatory disorder treatments (Fig. 4).

sTREM2 as a biomarker for neurological disorders

Accurately predicting the progression of neurological diseases is critical for enabling early diagnosis and timely intervention. sTREM2 in cerebrospinal fluid (CSF) and plasma, as a biomarker reflecting central TREM2 expression, has emerged as a promising predictive indicator for neurological disorders. Studies have demonstrated that CSF/plasma sTREM2 levels were significantly elevated in patients with AD [172, 173], PD [174], MS [175], ischemic stroke [176], and dementia [177] compared to healthy controls, indicating its potential value in risk assessment, progression monitoring, and prognosis prediction. A systematic review and meta-analysis integrating 36 studies further confirmed that plasma sTREM2 levels were markedly higher in AD patients than in healthy subjects, supporting its utility as a diagnostic biomarker for AD [178]. Moreover, sTREM2 levels in CSF or plasma exhibit dynamic fluctuations across disease stages, showing a close correlation with clinical severity. For instance, a longitudinal study involving 1017 AD patients revealed that CSF sTREM2 levels gradually increased with age, particularly in individuals over 75 years old, highlighting the potential impact of aging on sTREM2 dynamics [179]. Analysis of AD-related databases further indicated that sTREM2 may serve as a predictive marker for the transition from mild cognitive impairment (MCI) to AD, underscoring its role in disease progression [180]. In acute ischemic stroke research, plasma sTREM2 levels during the hyperacute phase positively correlate with both the National Institutes of Health Stroke Scale (NIHSS) score at admission (reflecting stroke severity) and the modified Rankin Scale (mRS) score at 3 months (assessing neurological recovery), hinting that plasma sTREM2 not only reflects acute severity but also predicts 3-month prognosis of diseases [181] (Fig. 4). CSF/plasma sTREM2 has also demonstrated positive predictive value in various clinical outcomes, including cognitive function in AD patients [182–184], depressive symptoms [185], sleep quality in PD patients [174], treatment response in MS patients [175], and mortality/cardiovascular events in acute ischemic stroke patients [186], as summarized in Table 2.

Table 2.

Predictive value of sTREM2 in neurological disorders: CSF/plasma levels and clinical outcomes

Sources	Diseases	Study population	Results	Conclusion	Refs
CSF	AD	Autosomal-dominant AD mutation carriers	There was a high amyloid burden at baseline, reduction of cortical atrophy, and alleviation of cognitive function.	CSF sTREM2 can identify carriers of pathogenic variants and their disease outcomes.	[182]
CSF	AD	Autosomal-dominant AD mutation carriers	The increase of CSF sTREM2 occurred 5 years before the expected onset of symptoms, but this increase occurred after amyloid deposition and neuronal damage.	CSF sTREM2 can identify carriers of pathogenic AD variants and predict disease outcomes.	[187]
CSF	AD	AD and mild cognitive impairment (MCI)	Higher CSF sTREM2 concentrations at baseline were associated with slower decline in memory and cognition.	A higher ratio of CSF sTREM2 to CSF ptau181 concentrations can predict slower conversion from cognitively normal to symptomatic stages or from MCI to AD dementia.	[183]
CSF	AD	Preclinical AD	According to the NIA-AA criteria, CSF sTREM2 levels decreased in stage 1 compared to stage 0, and then increased in stage 2.	CSF sTREM2 levels are dynamic in preclinical AD.	[173]
CSF	AD	AD	There was a higher sTREM2 concentration in AD and significant correlations between CSF sTREM2 and ttau and p-tau181.	CSF concentrations of sTREM2 correlate with markers of neurodegeneration, and may be used to quantify glial activation in AD.	[172]
CSF	AD	Cognitively unimpaired individuals at risk of AD	Higher baseline CSF sTREM2 was associated with a positive global cognition rate of change, and better memory and executive outcomes.	CSF sTREM2 can predict longitudinal cognition in AD patients with cognitively unimpaired.	[184]
CSF	AD	AD	The baseline CSF sTREM2 was significantly associated with baseline tau-PET and Aβ-PET rate of change only in the A+/TN+ group. A significant association was found between the rate of change of CSF sTREM2 and the tau- and Aβ-PET rate of change only in the A+/TN− group.	A faster increase in the level of CSF sTREM2 might attenuate future Aβ plaque formation and tau aggregate accumulation only in the presence of Aβ pathology.	[188]
CSF	AD	MCI	The level of CSF sTREM2 is independently correlated with subtypes of β-amyloid protein (such as Aβx-40), tau pathological markers, and BBB integrity markers(CSF/plasma albumin ratio).	CSF sTREM2 is independently associated with Aβ metabolism, tau pathology, and BBB function, and is a potential biomarker for disease progression.	[189]
CSF	AD	AD	Higher CSF sTREM2 was associated with slow clinical progression. The slow and medium progressing groups had higher CSF sTREM2 than the cognitively healthy, who had a similar level to patients with rapid clinical progression.	CSF sTREM2 levels were associated with clinical progression in AD, regardless of core biomarkers.	[190]
CSF	Depressio n and AD	Non-demented individuals at risk of AD	Minimal depressive symptoms (MDSs) individuals had lower CSF sTREM2 levels, and CSF sTREM2 mediated the association between MDSs and amyloid pathology.	CSF sTREM2 is associated with MDSs in non-demented patients.	[185]
CSF	PD	Newly diagnosed PD	The CSF sTREM2 did not differ between healthy controls and patients with PD or between PD clinical subgroups. However, higher baseline CSF sTREM2 predicted greater global cognitive decline in patients with PD.	CSF sTREM2 may be a promising predictor for the cognitive decline in PD rather than a diagnostic biomarker.	[191]
CSF	PD	Sporadic PD patients	The concentration of CSF sTREM2 had a positive correlation with CSF α-syn. The concentration of CSF sTREM2 of PD patients with sleep disorders was significantly increased and negatively correlated with the PDSS score.	CSF sTREM2 has significant diagnostic value for PD and its sleep disorders.	[174]
CSF	MS	Subjects with relapsing-remitting MS and primary progressive MS	CSF sTREM2 levels were significantly higher in relapsing-remitting MS and primary progressive MS subjects. Levels of sTREM2 in blood did not differ among the groups.	CSF sTREM2 may help diagnose MS.	[192]
CSF	MS	MS (from GWAS)	Genetically predicted per 1 pg/dL increase of CSF sTREM2 levels was associated with higher risk of multiple sclerosis.	Genetic predisposition to higher CSF sTREM2 levels is associated with higher risk of multiple sclerosis.	[193]
CSF	MS	MS patients at various stages	CSF sTREM2 levels in MS patients at various stages were significantly elevated, but decreased significantly after treatment with natalizumab or mitoxantrone, even reaching normal levels.	CSF sTREM2 can be used to assist in the diagnosis of MS and assess therapeutic efficacy.	[175]
CSF	ALS	Sporadic ALS patients	The CSF sTREM2 level in ALS patients was significantly higher than that in the control group. Its level was positively correlated with the UMN score, disease progression rate (ΔFS), and serum NFL.	CSF sTREM2 is elevated in ALS patients and may be a novel marker, probably reflecting upper motor unit severity and prognosis.	[194]
CSF	Delirium and dementia	Patients with or without pre-existing dementia who underwent acute hip fracture surgery	Delirium was associated with a higher level of CSF sTREM2 only among those without pre-existing dementia. Among patients with dementia, the level of Aβ38 and Aβ40 also correlated positively with sTREM2 in CSF.	Increased CSF sTREM2 prior to delirium could possibly serve as a delirium risk indicator.	[195]
CSF	Postoperat ive delirium (POD)	Patients undergoing total knee or hip arthroplasty	sTREM2 and CSF levels of tau and ptau in the POD group were higher than those in the NPOD group. The relationship between sTREM2 and POD was partially mediated by tau and ptau, with the mediation proportion of 17.91% and 22.09%, respectively.	Elevated CSF sTREM2 is a preoperative risk factor for POD, which is partially mediated by tau and ptau. Age ≥ 80 plus sTREM2 ≥ 20,000 pg/ml could increase 3-year mortality in POD cases.	[196]
CSF and plasma	AD	AD and MCI	MCI individuals with low levels of CSF sTREM2 and Aβ1–42 were more likely to develop AD, whereas with high plasma sTREM2 were at a greater risk for AD.	sTREM2 may be useful as a potential predictive biomarker of MCI-to-AD conversion.	[180]
CSF and plasma	AD	AD	There was a significant association between CSF and plasma sTREM2 levels only in the AD group, not evidenced in iNPH subjects suffering from a potentially reversible cognitive impairment.	Plasma sTREM2 may be a useful marker to identify cognitive impairment due to neurodegenerative processes.	[197]
CSF and plasma	AD	AD and MCI	CSF sTREM2 levels were significantly correlated with neurofibrillary degeneration, cognitive decline, and inflammasome activity in AD patients. Compared with plasma sTREM2, the concentration of CSF sTREM2.	CSF sTREM2 levels reliably predict neurofibrillary degeneration, cognitive decline, and inflammasome activation, and also have a high diagnostic potential for distinguishing diseased from healthy individuals.	[198]
CSF and plasma	Primary angiitis (PA)	Patients diagnosed with PACNS	sTREM2 levels in serum and CSF were significantly elevated in PA patients and significantly associated with the mRS, NIHSS and ADL scores as well as inflammatory cytokine levels.	sTREM2 represents a potential biomarker for monitoring disease and potentially predicting the prognosis of PA patients.	[199]
Plasma	AD	AD and MCI	The level of plasma sTREM2 in MCI patients was significantly higher than that in AD patients; the levels of inflammatory factors such as FGF-2, GM-CSF, and IL-1β in AD were lower than those in MCI, and their association with sTREM2 changes with disease stages.	The alterations in plasma sTREM2-related inflammatory activity are AD stagespecific, and are particularly critical in the early stages (MCI and A+T−N−).	[200]
Plasma	Ischemic stroke	Acute ischemic stroke	Higher plasma sTREM2 in the acute phase of ischemic stroke were associated with greater risk of death and cardiovascular events.	Plasma sTREM2 can predict death and cardiovascular events in patients with acute ischemic stroke one year later.	[186]
Plasma	Ischemic stroke	Patients within 4.5 h from the onset of ischemic stroke	Plasma sTREM2 concentrations at admission were associated with both the NIHSS at admission and the mRS at three months.	Plasma sTREM2 may serve as a biomarker for stratifying high-risk patients with ischemic stroke.	[181]
Plasma	Poststroke cognitive impairme nt (PSCI)	Acute ischemic stroke patients	The risk of PSCI elevated significantly with higher plasma sTREM2 levels.	Elevated level of plasma sTREM2 may be associated with PSCI, and sTREM2 has potential value in predicting PSCI.	[176]
Plasma	Ischemic stroke	Patients diagnosed with noncardioembolic Ischemic stroke	Higher initial NIHSS score, early increment of sTREM2, and late decrement of sTREM2, were more common in patients with poor outcome.	Increment of sTREM2 level at the early phase was a predictor of poor outcome.	[201]
Plasma	Small vessel disease	AD and spontaneous intracerebral haemorrhage with cerebral amyloid angiopathy or hypertensive small vessel disease.	Plasma sTREM2 levels were comparable between patients with AD and small vessel disease. In patients with small vessel disease, plasma sTREM2 was significantly associated with white matter hyperintensity volume.	Plasma sTREM2 can serve as a strong predictive marker for small vessel diseaserelated white matter injury.	[202]
Plasma	Dementia	Residents aged 60 and older without dementia	There is a significant association between elevated serum sTREM2 levels and the risk of all-cause dementia, AD, and VaD in the general elderly population in Japan.	High serum sTREM2 levels increase the risk of dementia in elderly ordinary residents.	[177]

Abbreviations: NIA-AA, National Institute on Aging-Alzheimer’s Association, BBB, blood-brain barrier, PDSS, Parkinson’s disease sleep scale, GWAS, genomewide association study, UMN, upper motor neuron, NFL, neurofilament light chain, POD, postoperative delirium, iNPH, idiopathic normal pressure hydrocephalus, PACNS, Primary angiitis of the central nervous system, mRS, modified Rankin scale, NIHSS, National Institutes of Health Stroke Scale, ADL, activities of daily living, FGF-2, Fibroblast Growth Factor 2, GM-CSF, Granulocyte-Macrophage Colony Stimulating Factor, VaD, vascular dementia

Notably, while sTREM2 holds unique potential as a predictor for neurological disorders, current research still confronts several challenges. First, some studies suffer from small sample sizes, which compromises the generalizability of results, and discrepancies exist across studies [203]. Second, the origin and metabolic mechanism of sTREM2 in CSF and blood remain unclear. CSF sTREM2 is widely regarded as a specific marker of microglial proliferation, whereas blood sTREM2 likely originates from various myeloid cells (such as circulating monocytes and macrophages), reflecting peripheral inflammation and correlating with C-reactive protein (CRP)—indicating distinct biological signatures for the two matrices [204, 205]. Existing immunoassays cannot discriminate sTREM2 sources, potentially diluting CNS-derived sTREM2 with peripheral counterparts and affecting its predictive accuracy for neurological diseases [206]. In conclusion, sTREM2 demonstrates substantial promise as a predictor of neurological disorders, offering valuable insights into disease risk, progression, and prognosis.

Summary and outlook

As a pivotal immune regulatory receptor, TREM2 plays a central role in neuroinflammation modulation. By interacting with diverse ligands and binding to DAP12/10, it governs microglial immune responses, thereby participating in neuroinflammatory pathologies such as neurodegenerative diseases, neural trauma, aging, and pain. TREM2 exhibits dynamic functional shifts, exerting both neuroprotective effects and potential pathological exacerbation under specific contexts, highlighting its complex bidirectional regulatory nature. Ongoing exploration of TREM2-targeted therapies shows promise, with agonistic antibodies and small-molecule agonists demonstrating efficacy in animal models and early clinical trials. However, the safety and efficacy of these approaches require further clinical validation. Additionally, sTREM2 holds considerable potential as a predictive biomarker for neurological diseases, enabling assessments of disease risk, progression, and prognosis. Nevertheless, its clinical translation is constrained by limitations in small sample size, heterogeneous origins of sTREM2.

Interestingly, the effect of TREM2 across distinct cell types and pathological contexts is inherently complex, resisting oversimplification. This is particularly evident in neurodegenerative diseases, where its efficacy is tightly coupled to disease stage. While direct activation or inhibition of TREM2 has not yet demonstrated universal therapeutic applicability, ongoing research aims to decipher its specific immune-regulatory mechanisms and dynamic interplay with diverse pathological states. They will undoubtedly fuel the development of novel immune-guided therapeutic strategies.

In summary, the role of TREM2 in neuroinflammation remains a frontier of active investigation. Future efforts must focus on clarifying the mechanistic nuances and validating the therapeutic potential of TREM2 targeting, which will provide a robust theoretical foundation for immune-based interventions in neurological disorders.

Abbreviations

BBB

Blood-Brain Barrier

AD

Alzheimer’s Disease

PD

Parkinson’s Disease

CNS

Central Nervous System

TBI

Traumatic Brain Injury

TREM2

Triggering Receptor Expressed on Myeloid Cells 2

aa

Amino Acids

OGD

Oxygen-Glucose Deprivation

LPS

Lipopolysaccharide

HMGB-1

High Mobility Group Protein Box 1

DAP12/10

DNAX Activator Protein 12/10

ITAM

Immunoreceptor Tyrosine-Based Activation Motif

MS

Multiple Sclerosis

ICH

Intracerebral Hemorrhage

SCI

Spinal Cord Injury

POCD

Postoperative Cognitive Dysfunction

5hmC

5-Hydroxymethylcytosine

miRNAs

MicroRNAs

NF-κB

Nuclear Factor-κB

ADAM17/10

A Disintegrin And Metalloproteinase Domain-Containing Protein 17/10

sTREM2

Soluble TREM2

BRI2

Integral Type II Transmembrane Protein

pSYK

Phosphorylated Spleen Tyrosine Kinase

NLRP3

NOD-Like Receptor Pyrin Domain-Containing Protein 3

ALS

Amyotrophic Lateral Sclerosis

DAMPs

Damage-Associated Molecular Patterns

PRRs

Pattern-Recognition Receptors

Aβ

Amyloid β

TLRs

Toll-Like Receptors

MAPK

Mitogen-Activated Protein Kinase

SYK

Spleen Tyrosine Kinase

PI3K

Phosphatidylinositol 3-Kinase

Akt

Protein Kinase B

GSK-3β

Glycogen Synthase Kinase-3β

mTOR

Mammalian Target of Rapamycin

DAM

Disease-Associated Microglia

FoxO3a

Forkhead Box O3a

JAK

Janus Kinase

STAT

Signal Transducer and Activator of Transcription

SOCS

Suppressor of Cytokine Signaling

AMPK

AMP-Activated Protein Kinase

MPTP

1-Methyl-4-Phenyl-1,2,3,6-Tetrahydropyridine

TRAF6

Tumor Necrosis Factor Receptor-Associated Factor 6

MyD88

Myeloid Differentiation Primary Response 88

COX-2

Cyclooxygenase-2

iNOS

Inducible Nitric Oxide Synthase

BDNF

Brain-Derived Neurotrophic Factor

NR2B

N-Methyl-D-Aspartate Receptor 2B

PSD95

Postsynaptic Density Protein 95

GDNF

Glial Cell Line-Derived Neurotrophic Factor

SAH

Subarachnoid Hemorrhage

ROS

Reactive Oxygen Species

CREB

cAMP-Responsive Element-Binding Protein

Bax

BCL2-Associated X Protein

DHCR24

24-Dehydrocholesterol Reductase

LXR

Liver X Receptor

IRAK3

Interleukin-1 Receptor Associated Kinase-3

MPO⁺

Myeloperoxidase-Positive

Bcl-2

B-Cell Lymphoma-2

Nrf2

Nuclear Factor Erythroid 2-Related Factor 2

APOE

Apolipoprotein E

CGRP⁺

Calcitonin Gene-Related Peptide-Positive

GEO

Gene Expression Omnibus

SAMP8

Senescence-Accelerated Mouse Prone 8

LTP

Long-Term Potentiation

CIPN

Chemotherapy-Induced Peripheral Neuropathy

CCL21

Chemokine (C-C Motif) Ligand 21

TfR

Transferrin Receptors

HSP60

Heat Shock Protein 60

cDNA

complementary DNA

CAR-T

Chimeric Antigen Receptor T

CEA

Carcinoembryonic Antigen

PD-1

Programmed Death-1

scFv

Single-Chain Variable Fragment

TAMs

Tumor-Associated Macrophages

MDSCs

Myeloid-Derived Suppressor Cells

CSF

Cerebrospinal Fluid

MCI

Mild Cognitive Impairment

NIHSS

National Institutes of Health Stroke Scale

mRS

Modified Rankin Scale

CRP

C-Reactive Protein

Authors contributions

Yinsheng Liao and Guo Mu: Conceptualization, literature search, writing; Shengfeng Deng: Writing and editing; Bin Lu and Maoyao Zheng: Conceptualization, supervision. All authors reviewed the manuscript.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors agreed to publish this study.

Competing interests

The authors declare no competing interests.