Astrocyte-driven multicellular mechanisms of CNS repair and cerebroprotection

Abstract

Astrocytes are central regulators of brain homeostasis, which respond dynamically to injury and disease by modulating neuronal activity, immune responses, and vascular function. Recent work has revealed that the diversity of astrocytic states and their context-dependent interactions with neighboring cells critically influence outcomes after central nervous system (CNS) injury and inflammation. This review synthesizes new insights into how astrocytes interact with microglia, neurons, and the neurovascular unit to shape neuroprotection and repair. We discuss how astrocyte-secreted proteins dynamically regulate synaptic remodeling following stroke; how regionally restricted lesion-remote astrocytes govern the specification of repair-associated microglia and promote white matter regeneration after trauma; and how signaling between astrocytes and microglia modulates neuronal survival in neurodegeneration. We further highlight emerging evidence linking disrupted astrocyte–endothelial cell interactions to neurodegenerative disease in humans. Together, these findings illustrate that astrocyte heterogeneity and multicellular communication define the balance between injury resolution and degeneration. Understanding these multidimensional astrocytic networks opens new avenues for therapeutic strategies aimed at promoting neuroprotection, restoring circuit function, and minimizing long-term neurological deficits.

Introduction

Almost every review in the last decade about astrocytes begins by stating that they were long regarded as support cells. Yet we now understand that they are dynamic regulators of central nervous system (CNS) function, with influence that extends to virtually every aspect of neural physiology. Astrocytes shape synaptic connectivity, modulate neurotransmission, maintain metabolic and ionic homeostasis, regulate blood flow and blood–brain barrier (BBB) integrity, and orchestrate immune responses within the CNS.^1–4 This broad functional repertoire is driven not only by their intrinsic complexity, but also by their remarkable capacity to establish highly specialized, and often regionally restricted, interactions with surrounding neurons, microglia, vascular cells, and other glial cells, including astrocytes themselves (Figure 1). Growing evidence suggests that these multicellular interactions are, in fact, the ones that determine whether the CNS engages successfully in repair processes after injury or on the contrary, leads it toward chronic dysfunction and degeneration.

Figure 1.

Astrocyte-mediated multicellular communication pathways involved in CNS repair and cerebroprotection. Astrocytes dynamically interact with neurons, microglia, and endothelial cells/NVU components to regulate synaptic remodeling, neuroinflammatory responses, BBB integrity, and vascular repair following CNS injury and during neurodegenerative disease. Astrocyte–neuron signaling pathways include secreted synaptogenic and plasticity-regulating proteins such as Chrdl1, TSP1/2, and SPARC, which modulate excitatory synapse formation, AMPAR regulation, and post-stroke plasticity. Astrocyte–microglia communication involves cytokine-mediated and inflammatory molecules, including CCN1, APOE, CLU, LRP1, SFRP1, CCL2, OXR1, CHL1, and reactive markers such as VIM, which influence neuroinflammation, repair-associated microglial states, and neurodegenerative progression. Astrocyte–endothelial/NVU interactions include pathways related to vascular remodeling, inflammatory signaling, and BBB regulation, including TNF–STAT3 signaling, Serpina3n, WNT10B/FZD7 signaling, and LAMC1-mediated vascular remodeling. Bidirectional communication between astrocytes and neighboring cell types ultimately determines the balance between neuroprotection, tissue repair, and chronic neurodegeneration in the injured CNS. The pathways and interactions shown are representative and not exhaustive; several proteins may participate in additional signaling networks and intracellular mechanisms not depicted here, including context-dependent functions of CCN1.

Created with BioRender.

In recent years, the field of astrocyte biology has gained unprecedented insight into how astrocytes respond to pathology, thanks to advances such as single-cell transcriptomics, spatial profiling, high-resolution imaging, and cell-type-specific genetic and proteomic tools. All these approaches have revealed that astrocytes are not the homogeneous cell population we once assumed, and that their response to injury is far from uniform.^5,6 Instead, astrocytes enter diverse, context-dependent reactive states shaped by local cues, developmental stage, timepoint after injury onset, and neuroanatomical location. All these factors not only influence the astrocyte response but also the response of neighboring cells to injury. Thus, if we want to understand CNS repair and recovery mechanisms in injury and disease, we must define and understand the multicellular signaling systems where astrocytes play crucial roles. In this review, we highlight three such systems: astrocyte–neuron interactions that regulate synaptic remodeling and restoration; astrocyte–microglia crosstalk involved in the neuroinflammatory response; and astrocyte–endothelial cell interplay as an example of astrocyte-mediated regulation of the neurovascular unit (NVU).

One of the most studied functions of astrocytes in the last years is the regulation of synaptic formation and remodeling. Much less well documented, however, is astrocyte-mediated synaptic regulation in the context of stroke or neurodegenerative diseases. Some work has examined this astrocytic function in response to stroke,^7,8 moving well beyond the common idea that they “only” become reactive and generate the formerly known “glial scar,” or act merely as targets of neurodegeneration. ⁹ Astrocyte-derived proteins shape excitatory and inhibitory synapses during development, and the list of these identified proteins has significantly grown in the last few years (Table 1). A recurring theme across these studies, and something we emphasize in this review, is that astrocyte-mediated synaptogenesis appears to be strongly dependent on the brain region considered. Astrocyte heterogeneity, once appreciated only at the level of morphology (white matter vs gray matter astrocytes), is now recognized as a fundamental biological principle defining circuit development, synaptic remodeling, and injury repair. ¹⁰

Table 1.

Identified astrocyte-secreted proteins involved in synaptic regulation.

Astrocyte-secreted protein	Synaptic target, effect	References
TSP1, TSP2	Excitatory (structural synapses, often silent)	Christopherson et al. 26
BDNF	Inhibitory, regulation	Elmariah et al. 29
TIMP-1	Modulator, synaptic stabilization (learning and memory)	Jourquin et al. 66
TNF-α	Excitatory and inhibitory, receptor trafficking	Stellwagen et al. 67
CSPG	Excitatory, stabilization	Pyka et al. 68
SPARC	Excitatory, antagonist	Kucukdereli et al. 14
Hevin (SPARCL1)	Excitatory, synaptogenic	Kucukdereli et al. 14
GPC4, GPC6	Excitatory, functional synapse formation	Allen et al. 27 and Farhy-Tselnicker et al. 69
TGF-β1	Inhibitory and excitatory, formation	Diniz et al.70,71
Complement C3	Excitatory, Ca2+ regulation	Lian et al. 72
Chrdl1	Excitatory, maturation/plasticity limitation	Blanco-Suarez et al. 15
PTX3	Excitatory, functional synapse formation	Fossati et al. 73
IL-33	Excitatory, homeostatic synaptic plasticity	Wang et al. 74
CLU/ApoJ	Excitatory (post- and presynaptic), structure and transmission	Chen et al. 75
PTN	Excitatory, rescue in Down’s Syndrome model	Caldwell et al. 41 and Brandebura et al. 76
NCAN C-terminal fragment	Inhibitory, SST+ interneurons formation	Irala et al. 30
GPC5	Excitatory, maturation/stabilization	Bosworth et al. 28
CCN1	Excitatory and inhibitory neurons. Interaction with oligodendrocytes and microglia.	McCallum et al. 18 and Sancho et al. 45

TSP1/2: thrombospondin-1 and -2; BDNF: brain-derived neurotrophic factor; TIMP-1: tissue inhibitor of metalloproteinase-1; TNF-α: tumor necrosis factor-α; CSPG: chondroitin sulphate proteoglycans; GPC4/5/6: Glypican-4, -5, and -6; TGF-β1: transforming growth factor-beta 1; Chrdl1: chordin-like 1; PTX3: pentraxin 3; IL-33: interleukin-33; CLU: clusterin; PTN: pleiotrophin; NCAN: neurocan; CCN1: cellular communication factor 1.

The region-dependent astrocyte heterogeneity becomes especially important after stroke, when the brain enters a temporally restricted phase of enhanced endogenous plasticity that typically spans from days 2 to 30 in rodent preclinical models, and within the first 2–3 months after stroke in human patients. ¹¹ During this recovery phase, astrocytes can promote or impede functional restoration depending on which molecules they secreted and how they interact with surrounding cells, including neurons (Table 1).^7,8,12 Several astrocyte-secreted proteins have been identified to be regulators of AMPA receptors (AMPARs), such as SPARC and Chrdl1.^13–15 These proteins display striking differences under physiological versus injury conditions, suggesting that injury modifies the astrocyte-to-neuron (and vice versa) signaling mechanisms in non-canonical ways. Manipulating these mechanisms in preclinical rodent models has yielded improvements in dendritic spine preservation, excitatory synapses stabilization, and ultimately behavioral recovery. These studies highlight astrocyte–neuron communication and the proteins responsible for this regulation as promising therapeutic targets under injury conditions.

Another important aspect in injury repair is astrocyte–microglia communication, a major player in the neuroinflammatory response in the CNS. Microglia rapidly sense and respond to the pathological cues after injury. Microglia-derived cytokines and extracellular vesicles (EVs) can modulate astrocyte phenotypic transitions after injury that dictate reactivity states, astrocyte border formation (formerly known as “glial scar”), and in some contexts drive neurotoxic or neuroprotective cascades. ¹⁶ These interactions are not unidirectional, and astrocytes simultaneously shape microglial behavior through contact-mediated cues, ¹⁷ which can have important functions in injury scenarios. New research employing transcriptional methods has revealed the heterogeneity of the astrocyte–microglia communication depending on their location relative to the core of the injury. ¹⁸ Beyond injury contexts, neurodegeneration has been shown to be strongly mediated by astrocyte–microglia crosstalk.^19,20 To study these mechanisms with enhanced translational potential to human disease, new approaches using neurospheres derived from human 3D induced pluripotent stem cells (iPSC) now allow researchers to dissect the molecular pathways between neurons, astrocytes, and microglia.^20,21 These new integrative approaches are crucial to advance our understanding of astrocyte-mediated multicellular cascades in human pathology. The use of iPSCs has also been very helpful in the study of other multicellular mechanisms, such as astrocyte-mediated inflammation that may contribute to BBB dysfunction. ²²

The interplay between astrocytes and the components of the NVU is increasingly recognized as a major determinant of injury and neurodegeneration progression or repair. Astrocyte endfeet envelop nearly all capillaries in the brain, positioning astrocytes as central integrators of vascular signals and regulators of BBB integrity. Pathological disruption of astrocyte–endothelial cell (EC) communication not only exacerbate BBB leakage after acute injuries such as stroke or TBI, but may also contribute to or accelerate chronic neurodegenerative disease.^19,23 New proteomic approaches, such as astrocyte-specific proximity biotinylation, have begun to map endfoot-endothelial signaling pathways with unprecedented resolution.^19,24 Moreover, recent studies demonstrate region-dependent roles of astrocytes in angiogenesis and vascular remodeling, further reinforcing the overarching theme of astrocyte heterogeneity.

In this review, we synthesize and contextualize recent work that was presented at a symposium titled Leveraging glia to improve CNS repair at the international meeting Brain and Brain PET 2025 in Seoul, South Korea. The symposium brought together worldwide experts with a common goal of presenting ground-breaking research and data supporting new strategies to improve CNS repair and function in the context of various conditions, including stroke, spinal cord injury, and neurovascular unit disruption. Taken together, these studies illuminate how astrocytes coordinate plasticity, immune responses, and vascular remodeling via multicellular communication (Figure 1).

Astrocyte–neuron interactions in plasticity regulation after stroke

More than 20 years ago, it was demonstrated that neuronal function was enhanced in the presence of astrocytes. These experiments were initially conducted in vitro, demonstrating that synaptic efficacy was not an intrinsic property of the neuron, but is regulated by extrinsic signals, particularly from neighboring astrocytes. ²⁵ Follow-up studies demonstrated that astrocytes can regulate excitatory synaptic formation, maturation, and function in different brain regions by secreting a plethora of proteins.^{13–15,26–28} Likewise, astrocytes can also promote and enhance the formation of inhibitory synapses via protein secretion.^29–31 All these studies have contributed to a growing list of astrocyte-secreted proteins known to be crucial regulators of synaptic development, function and stabilization, as well as important effectors of intercellular communication (Table 1). A common thread in all these studies is that the synaptogenic properties of astrocytes appear to be region-dependent. This aligns with more recent findings in the astrocyte biology field showing that astrocytes are highly heterogeneous, and so are their functions, depending on the brain region in which they are located.^10,32

Astrocyte–neuron interactions have been key in understanding synaptic formation, maturation, and function in physiological conditions and during development, which is very important for understanding what goes awry during pathology ³³ and in response to injury, such as stroke, ⁷ and for identifying and designing new strategies that may stimulate neuronal plasticity and repair.^7,8 In particular, during the post-stroke phase commonly referred to as the recovery phase, we know that astrocytes can play important roles in either promoting or restricting plasticity and therefore aiding or hindering recovery, respectively. ⁸ One of the key functions of astrocytes that may be extremely beneficial during this phase is their ability to regulate plasticity by promoting the formation of new synapses, stabilizing dendritic spines, or allowing for the redistribution of synaptic receptors, all to compensate for the injury-induced loss.

One of the first studies that linked astrocyte-secreted proteins to neuronal protection and recovery after an ischemic insult identified thrombospondin-1 and -2 (TSP-1 and TSP-2) as important players in the recovery phase after stroke. ³⁴ They found that elimination of TSP-1 and TSP-2 in a mouse model worsened stroke outcomes, showing greater loss of excitatory synapses in the peri-infarct area, as well as impaired axonal sprouting.

Secreted protein acidic and cysteine rich (SPARC) is a matricellular protein that, under physiological conditions, is predominantly enriched in microglia but is also expressed in cortical astrocytes and it has been linked to the formation of excitatory synapses.^13,14,35 SPARC illustrates how the same astrocyte-secreted protein can produce divergent, and even seemingly contradictory, effects depending on the physiological or pathological context. SPARC was found to be upregulated in astrocytes in response to ischemic lesions, and in this context, GluA1-containing AMPARs were recruited to hippocampal synapses. Interestingly, exogenous SPARC was beneficial only when administered prior to induction of ischemic-like conditions. ³⁶ The authors of this study speculated that this may provide neurons with increased protection from excitotoxic damage that occurs in the early stages following ischemic injury. However, this finding is puzzling, as elevated levels of GluA1-containing AMPARs are normally linked to worse neuronal survival in response to excitotoxicity due to their higher Ca²⁺ permeability.^37–39 This research team also demonstrated that SPARC KO mice exhibited increased AMPARs levels and synaptic strength, which contradicts their results obtained in the stroke study.^13,36 These results raise the compelling possibility that astrocytes, via SPARC, may fulfill different roles in physiological versus pathological or injury contexts.

Other AMPAR-regulating astrocyte-secreted proteins have been identified, such as Glypican 4, 5, and 6 (Gpc4, 5, and 6),^27,28 and chordin-like protein 1 (Chrdl1). ¹⁵ Chrdl1 is a bone morphogenetic protein antagonist, a secreted glycoprotein highly expressed in astrocytes in the upper layers of the cortex. ¹⁵ It was initially identified in the CNS as the astrocyte-secreted factor that recruits GluA2-containing AMPARs to cortical synapses contributing to synaptic maturation and limiting experience-dependent plasticity. ¹⁵ It was found to be upregulated in peri-infarct astrocytes in response to ischemic conditions both in vitro and in vivo.^12,40 Elimination of Chrdl1 promoted enhanced experience-dependent plasticity which proved to be beneficial in the context of focal cortical strokes.^12,15,40 In fact, in the absence of Chrdl1, dendritic spines were protected from loss in the peri-infarct region, ¹² and Chrdl1 KO mice exhibited improved motor performance during the post-stroke recovery phase, indicating that absence of Chrdl1 could aid recovery in the context of focal ischemic strokes. ⁴⁰ Interestingly, stroke-induced loss of GluA2 was not observed during the recovery phase when Chrdl1 was absent, ⁴⁰ something that seems counterintuitive given the GluA2-recruiting function of this protein during synaptic development and maturation. ¹⁵ This is important, as it appears to be a similar situation found with SPARC, where both astrocyte-secreted proteins may play distinct functions during development, or under pathological conditions.

Overall, all these studies show that manipulating the post-stroke expression and function of these astrocyte-secreted proteins can produce effects that, although sometimes subtle, still impact overall functional performance after stroke. This indicates that targeting astrocyte-secreted proteins after stroke and manipulating them appropriately depending on whether they play beneficial or detrimental roles, can represent an important therapeutic strategy to strive for better outcomes following ischemic injuries. Even if these rodent preclinical models show modest improvements, they could ultimately represent important advancements in the development of therapeutics for stroke survivors. Beyond stroke, more studies are expanding on the communication between astrocytes and neurons in other CNS disorders,^33,41 with promising results highlighting these secreted proteins as potential and effective therapeutic targets.

Astrocyte–microglia crosstalk in repair and disease

Astrocytes and microglia are deeply involved in neuroinflammation, with important consequences in CNS repair. ⁴ This intercellular signaling between these two cell types has been characterized in previous studies, which found that, depending on the stimulus, microglia can trigger different phenotypes in astrocytes, leading them to exert different functions. ⁴² Microglia-derived cytokines have been identified as the signals that promote this phenotypic shift in astrocytes toward reactivity and the formation of the astrocytic border, that is, characteristic of acute injuries such as ischemic stroke or traumatic brain injury (TBI). ⁴³ Early after injurious signals, microglia were found to upregulate the expression of tumor necrosis factor-α (TNF-α), interleukin 1β (IL-1β), and IL-6, cytokines that further exacerbated the reactive phenotype of astrocytes. ⁴³ Interestingly, in this study they found that these pro-inflammatory signals actually promoted neuroprotection in response to TBI, ⁴³ while other earlier studies revealed that these pro-inflammatory cytokines tend to promote neurotoxic effects, together with additional molecules like IL-1α or complement component C1q.^42,44

Rather than being contradictory, these studies suggest that the role of microglia and astrocyte communication can lead to neuroprotection or neurodegeneration depending on multiple factors, such as the nature of the pathological stimulus. In addition, given that these studies were conducted in different brain regions and at different developmental stages, these are also factors that can possibly explain these differing results.

In response to injury, there are unaffected CNS regions that are known to be crucial in recovery, possibly due to compensatory mechanisms and the remodeling processes that they undergo which may have important implications for the repair of the affected areas. In a recent study in a spinal cord injury (SCI) model, researchers characterized the transcriptomic profile of astrocytes that entered reactive states in remote areas from the core of the lesion. ¹⁸ Remarkably, they found that CCN1, a secreted matricellular protein that has also been recently identified as a novel astrocyte-derived synaptic regulator, ⁴⁵ drove the phenotypic transformation of these lesion-remote astrocytes (LRA), driving aberrant microglial responses. Using single-nuclei RNA sequencing and spatial transcriptomics, the authors of this study revealed divergent reactive states of LRAs. Overall, this study provided important information about the neuroanatomical diversity of the LRAs, their heterogeneous phenotypes, and their underlying mechanisms in response to SCI, ¹⁸ further highlighting the region-specific roles of astrocytes in response to acute injuries.

Astrocyte–microglia crosstalk is also of emerging interest in neurodegenerative diseases such as Alzheimer’s disease (AD). One role that has been ascribed to microglia in AD models is the phagocytosis of amyloid beta (Aβ) plaques, ⁴⁶ that otherwise accumulate during AD pathology. Microglia also modify the neuronal and astrocytic transcriptional profiles, as shown in a recent study, using iPSC-derived 3D neurospheres (hiNS). ²⁰ The authors were able to determine the role of microglia in this system, finding that many AD-associated genes were altered in astrocytes, including APOE, CLU, LRP1, and VIM, only when microglia were present. The neuroprotective roles of microglia were time-dependent, including microglia-dependent reactive astrogliosis, as seen by increased GFAP expression, alongside genes related to oxidative stress regulation such as OXR1, CCL2, and CHL1. Cytokines such as IL-6 and IL-12 also increased in a time-dependent manner when the hiNS were exposed to Aβ. Overall, this study illustrated the importance of time-dependent multicellular crosstalk for understanding the molecular landscape of neurodegenerative diseases like AD.

One intriguing field of research is the study of EVs in multi- and intercellular communication. Previous studies have shown that EVs derived from other cell types, such as human bone marrow mesenchymal stem cells (hBM–MSCs), can play important roles in ameliorating motor deficits in response to stroke, and dampening reactive astrogliosis as well as immune response (including reduced Iba-1, TNF-α, IL-1β, and IL-6), indicating that astrocytes and microglia must be responding to EV-induced mechanisms.^47,48 Some studies have shown the critical role of astrocyte-derived EVs, ⁴⁹ or microglia-derived EVs in astrocyte–microglia crosstalk ⁵⁰ in response to stroke. For example, miRNA-124 was found in the microglia-derived EVs which inhibited the astrocyte-mediated border formation and astrocyte proliferation and migration due to reduction of STAT3 and p-STAT3, illustrating how astrocyte–microglia crosstalk could be mediated via EVs. EVs pose a very interesting mechanism that may be mediating more of the multicellular mechanisms where astrocytes are implicated in response to injury, not only with microglia but also with other cell types, and even astrocytes themselves.

Other recent studies also demonstrate that microglia closely interact with astrocytes to regulate both neuroinflammation and the clearance of toxic protein aggregates, including Aβ.^51–53 While cooperative interactions between astrocytes and microglia can enhance aggregate clearance,^51,52 for example, through direct membrane contacts and microglial uptake of astrocyte-associated protein deposits, chronic inflammatory signaling can also drive pathology. ⁵³ Astrocyte-derived factors such as secreted frizzled-related protein 1 (SFRP1) have been shown to sustain microglial activation and amplify inflammatory pathways, including hypoxia-inducible factor (HIF)-dependent signaling and NF-κB-associated responses. ⁵³ In parallel, spatial transcriptomic analyses of the amyloid plaque niche in AD mouse models revealed that plaque-associated microglial accumulation drives heterogeneous astrocytic responses, promoting neurotoxic astrocyte phenotypes and disrupting neuronal synaptic signaling through enhanced GABAergic and reduced glutamatergic activity. ⁵⁴ Additional evidence indicates that microglia directly modulate Aβ-dependent astrocyte reactivity in AD, further highlighting the reciprocal nature of glial communication in disease progression. ⁵⁵ Supporting this concept, human imaging and biomarker studies have demonstrated that Aβ pathology is associated with reactive astrogliosis primarily in the presence of activated microglia, suggesting that microglial activation is a critical mediator of Aβ-induced astrocyte reactivity. Moreover, these microglia-dependent astrocytic changes are linked to downstream tau phosphorylation, tau aggregation, and cognitive impairment, emphasizing the importance of glial crosstalk in the progression of AD pathology. ⁵⁵ Importantly, astrocyte–microglia interactions may also exert protective effects, as astrocyte-derived interleukin-3 (IL-3) can reprogram microglia toward neuroprotective states that limit AD pathology. ⁵¹

Together, these studies demonstrate that astrocyte–microglia interactions are highly context-dependent and critically influence neuroinflammation, protein clearance, synaptic homeostasis, and neuronal survival in injury and neurodegenerative disease.

Astrocyte role in the blood–brain barrier under pathological conditions

Acute injuries, like stroke and TBI, normally involve blood–brain barrier (BBB) disruption.^56,57 However, growing evidence points to BBB disruption and dysregulation of the NVU also as potential triggers for neurodegenerative disease, including Alzheimer’s and Parkinson’s disease.^58,59 Astrocytes are an important component of the neurovascular unit, a term that was coined in 2001 to refer to the cellular interactions, both molecular and physical, that comprise the BBB. ⁵⁶ These interactions are crucial for maintaining BBB integrity, and any impairments or alterations may lead to BBB instability and pathological responses. ³ It is increasingly recognized how understanding intercellular communication at the NVU level is essential to preserve brain health and identify new therapeutic targets. ⁶⁰

At the capillary level, the NVU is comprised of endothelial cells, pericytes, astrocytic endfeet, and the basement membrane, with additional contributions from neurons, microglia, and other extracellular matrix components. ⁵⁶ Astrocyte endfeet are in close proximity to the ECs; however, studies on how these cells communicate in both physiological and pathological conditions are rather scarce. This situates astrocytes at the interface between synaptic activity and the vasculature, allowing astrocytes to participate in blood flow regulation and neurovascular coupling through Ca²⁺-dependent signaling, K⁺ buffering, vasoactive lipid mediators, and interactions with pericytes, smooth muscle cells, and ECs. Disruption of these pathways may contribute to cerebral hypoperfusion and impaired metabolic support in AD, while in stroke, regional failure of neurovascular coupling and microvascular remodeling can influence peri-infarct survival, reperfusion, and plasticity during recovery. ⁶¹ It was previously shown that reactive astrocytes may influence vascular repair and remodeling in response to stroke by stabilizing multiple components of the NVU, and restoring blood flow and BBB integrity.^62,63

Astrocyte–EC crosstalk has been recognized as crucial for maintaining BBB integrity. For example, a recent study utilizing iPSCs described the TNF–STAT3 signaling pathway as a key cascade in BBB inflammation, with the limitation, as recognized by the authors, that did not distinguish different reactive astrocyte subtypes. ²²

Another recently published study that sheds light on this issue examined astrocyte–EC interactions in depth and characterized various astrocyte endfoot-to-EC pathways. ¹⁹ With further studies aimed at defining the transcriptional and proteomic profiles of reactive astrocyte subtypes, we will be able to understand the region-specific contributions of reactive astrocyte subtypes to BBB integrity in response to injury and disease. In this study, ¹⁹ the authors employed a method to combine astrocyte-specific proximity in vivo biotinylation using TurboID with vessel purification. Their approach was based in previous studies where endfoot-specific proteins (e.g. aquaporin4 or AQP4 ⁶⁴ ) were used to reveal the interactions mediated by the specific astrocytic protein located in the endfoot. The difference here is that this study expressed TurboID in the astrocytic cytoplasm and combined it with capillary purification to obtain a broader readout of the astrocyte endfoot proteins that may be interacting with ECs. ¹⁹ The pathways identified using this method were mostly related to inflammation and immune responses, as well as vascular function, and the authors found that most of these pathways were altered in response to inflammatory responses such as those triggered by LPS. For example, Wnt10b-FDZ7 was identified as an important pathway to translate signals from the blood via communication of astrocyte–ECs. ¹⁹ The authors also explored pathways that may be present in neurodegenerative disease in humans, specifically multiple sclerosis (MS), and AD, finding that, perhaps unsurprisingly, the BBB responses to acute inflammation differ from those in human chronic disease. They concluded that the mouse astrocyte–EC interactome was largely conserved in human samples, highlighting its translational potential. ¹⁹

In another recent study using a preclinical model of ischemia, researchers found that cortical astrocytes possess an angiogenic function via LAMC1, contributing to microvessel remodeling during stroke recovery. ⁶⁵ While intercellular crosstalk, and specifically the communication between astrocytes and other cellular components of the NVU, was not explored in this study, it nevertheless demonstrates that astrocytes have a heterogeneous response under pathological conditions and that astrocyte–vasculature interactions may also be region-dependent as we often saw in the studies presented in this review.

Discussion

The studies highlighted throughout this review reinforce a central emerging concept in neurobiology: astrocytes are not passive responders to CNS injury, but dynamic coordinators of multicellular repair programs whose functions are highly dependent on spatial context, temporal stage, and disease environment. Across acute injuries such as stroke and spinal cord injury, as well as chronic neurodegenerative disorders, including AD, astrocytes integrate neuronal, immune, and vascular signals to influence whether the CNS progresses toward recovery or sustained dysfunction. Importantly, these outcomes are not determined by astrocytes alone, but rather by the complex signaling networks established between astrocytes and neighboring neurons, microglia, endothelial cells, and other components of the neurovascular unit.

Studies discussed here demonstrate that astrocyte responses differ depending on proximity to lesions, white versus gray matter localization, developmental origin, and local inflammatory cues. These distinctions are particularly important when considering astrocyte-mediated regulation of synaptic remodeling, inflammatory signaling, and vascular repair. For example, astrocyte-secreted proteins that regulate synapse formation and AMPAR trafficking during development can exert markedly different, and sometimes opposing, effects after ischemic injury.^12,34,36,40 Similarly, lesion-remote astrocytes can acquire specialized transcriptional states that influence microglial phenotypes and white matter regeneration after spinal cord injury. ¹⁸ These findings collectively argue against viewing reactive astrocytes as a single unified state and instead support a model in which multiple astrocyte subpopulations exert context-dependent reparative or maladaptive functions.

The studies discussed here also underscore the importance of emerging technologies in redefining our understanding of astrocyte biology. Advances in single-cell and spatial transcriptomics, in vivo proteomics, proximity labeling, human iPSC-derived multicellular systems, and high-resolution imaging are beginning to reveal previously inaccessible aspects of astrocyte communication networks.^19–21,54 These approaches are moving the field beyond simplified binary classifications of reactive astrocytes and toward a more nuanced understanding of dynamic astrocyte states across time and disease contexts. Importantly, the increasing use of human-derived systems offers new opportunities to bridge longstanding translational gaps between rodent models and human pathology.

A major challenge moving forward will be translating our growing understanding of the spatiotemporally heterogeneous roles of astrocytes into therapeutic strategies that are precise, context-dependent, and targeted to specific reactive states, signaling pathways, or neuroanatomical regions. Many important questions remain unanswered: How do distinct astrocyte subpopulations coordinate multicellular repair programs? What molecular and environmental cues determine whether astrocyte-driven responses promote neuroprotection and regeneration versus chronic inflammation and degeneration? As emerging technologies continue to refine our ability to define astrocyte subtypes, circuit-specific functions, and dynamic intercellular signaling networks, it is becoming increasingly evident that targeting astrocyte-mediated multicellular communication holds substantial promise for improving outcomes across a broad spectrum of CNS disorders. Ultimately, understanding how astrocyte-driven signaling networks are coordinated across space and time will be essential for overcoming current translational barriers and for developing therapeutics capable of restoring circuit function, promoting cerebroprotection, and enhancing recovery in human neurological disease.