Glial Cells as Central Players in Neuroinflammation and Neuronal Damage Caused by Bacterial Pneumonia

Abstract

Background

Pneumonia is an infection that affects the alveolar spaces of the lungs, associated with high global mortality, and remains a significant public health challenge worldwide. In a compromised immune system, the infection can progress, leading to the establishment of pneumonia. During this process, an intense inflammatory response is triggered in the lungs through the activation of resident immune cells, especially alveolar macrophages. This activation promotes the recruitment of neutrophils and the release of pro-inflammatory cytokines, ultimately resulting in the formation of exudative infiltrates within the alveoli. Pneumonia is a leading cause of sepsis, particularly among hospitalized patients and in intensive care units. Sepsis represents one of the most severe complications of pneumonia and is characterized by a dysregulated systemic inflammatory response to lung infection. Another critical challenge to treating clinical infectious conditions, which can lead to life-threatening sepsis, septic shock, and multiorgan dysfunction, is the continuous growth of antimicrobial resistance in bacteria.

Summary

Among the organ dysfunctions associated with sepsis, sepsis-associated encephalopathy (SAE) is the most frequent and constitutes a primary contributor to the neurological alterations observed in critically ill patients. Although SAE is often classified as a fully reversible pathophysiological process, increasing evidence suggests an association between sepsis, structural brain injury, and long-term neurological sequelae. The central nervous system (CNS) is one of the first regions exposed to peripheral inflammation during sepsis, allowing inflammatory mediators and immune cells to infiltrate the brain. This process activates microglia, the resident immune cells of the CNS, exposing neurons to an oxidative stress-rich environment that leads to neuronal dysfunction and apoptosis. A dysregulated pro-inflammatory microglial response plays a significant role in SAE, as microglia-derived cytokines are strongly associated with neuronal damage. Furthermore, activated microglia stimulate astrocytes to adopt a reactive inflammatory phenotype, thereby amplifying neuroinflammation.

Key Messages

Recent studies have demonstrated that regulating microglial and astrocytic hyperactivation can attenuate the inflammatory response. Therefore, targeting glial cells during SAE holds significant therapeutic potential, offering a promising avenue for the development of new strategies aimed at reversing the exacerbated CNS inflammatory response, mitigating neuronal damage, and ultimately reducing the long-term neurological sequelae observed in post-septic patients.

Introduction

Pneumonia is a disease that affects the alveolar spaces of the lungs, associated with high global mortality, particularly among children, the elderly, and immunocompromised individuals. It remains one of the major global public health challenges. Recent data indicate that pneumonia is the second leading cause of death in children under 5 years of age in the USA [1, 2].

Pneumonia is classified into three categories based on the setting of acquisition: community-acquired pneumonia (CAP), hospital-acquired pneumonia, and ventilator-associated pneumonia [3]. The most common symptoms of pneumonia include fever, productive or persistent cough, dyspnea, chest pain, and pulmonary crackles [4].

Different pathogens, including viruses, fungi, and bacteria, can cause CAP. Host-related factors such as immune status, age, quality of life, and comorbidities are also relevant to disease susceptibility and progression [5].

The leading etiological agent of CAP is Streptococcus pneumoniae, a trend observed in both adults and children [6]. Haemophilus influenzae is the second most common bacterial pathogen, particularly in patients with chronic pulmonary comorbidities [7]. Common CAP-related organisms include Klebsiella pneumoniae, Legionella spp., Mycoplasma pneumoniae, Chlamydia spp., Pseudomonas aeruginosa, and in some regions, Mycobacterium tuberculosis and Burkholderia pseudomallei. In aspiration pneumonia, Gram-negative enteric bacteria such as Escherichia coli and Proteus mirabilis are often identified. The spectrum of bacterial pathogens can vary significantly depending on the geographic region, vaccination patterns, and local epidemiological characteristics [6, 7].

In cases of hospital-acquired pneumonia, the primary bacterial pathogens include K. pneumoniae, E. coli, P. aeruginosa, Staphylococcus aureus, Enterobacter spp., and the Acinetobacter baumannii complex. These pathogens can also cause ventilator-associated pneumonia, which may involve both non-multidrug-resistant organisms, such as S. pneumoniae and H. influenzae, and multidrug-resistant strains, including P. aeruginosa and A. baumannii [8]. Those are World Health Organization (WHO)-listed critical pathogens due to their antimicrobial resistance (AMR) and global public health threat.

Bacteria can reach the lung parenchyma through several mechanisms. The primary routes of infection include microaspiration of colonized oropharyngeal secretions, often occurring silently during sleep; macroaspiration, typically associated with vomiting, gastroesophageal reflux, or neurological disorders that impair swallowing; and hematogenous spread, where pathogens reach the lungs via the bloodstream from a distant infectious focus. These mechanisms contribute to the initiation of the pulmonary inflammatory response characteristic of bacterial pneumonia [9].

Once they colonize the oropharynx, bacteria can be transmitted to other individuals. The infection may progress in susceptible hosts, particularly those with compromised immune systems, leading to the development of pneumonia. Pneumonia may also occur alarmingly as a secondary bacterial infection following a primary viral illness, such as influenza, respiratory syncytial virus disease, or COVID-19 [10, 11].

During pneumonia, an intense inflammatory response is triggered in the lungs, initiated by the activation of resident immune cells, particularly alveolar macrophages, in response to pathogen-associated molecular patterns. This process leads to the recruitment of neutrophils and the release of pro-inflammatory cytokines such as tumor necrosis factor-alpha (TNF-α), interleukin-1β (IL-1β), and interleukin-6 (IL-6), resulting in the formation of exudative infiltrates within the alveoli. These infiltrates impair gas exchange and contribute to hypoxemia [12].

Sepsis is one of the most severe complications of pneumonia, characterized by a dysregulated systemic inflammatory response to infection. Evidence indicates that pneumonia is a leading cause of sepsis, particularly among hospitalized patients and those in intensive care units [13]. The triggering event of sepsis is the colonization of an organ, compartment, or fluid of the organism by a pathogen, inducing an inflammatory reaction. The associated cytokine storm involves the excessive release of pro-inflammatory cytokines in response to bacteria, leading to intense immune activation [14]. This exaggerated response may trigger complement activation, coagulation disturbances, multiple organ failure, and ultimately death.

Physiopathology of Pneumonia and Sepsis

In the lungs, alveolar macrophages account for approximately 95% of the white blood cells among the alveolar cells, playing a crucial role in the pathogenesis of lung inflammation and serving as the primary initiating cell of the local inflammatory response. They recognize pathogen-associated molecular patterns using specific pathogen recognition receptors, namely, Toll-like receptors (TLRs). They are also able to identify host-derived debris released by damaged cells, known as damage (or danger)-associated molecular patterns (DAMPs) [15].

When activated, these macrophages synthesize and secrete a large number of inflammatory mediators, triggering the activation of other immunocompetent cells and enhancing the release of inflammatory factors, which leads to an uncontrolled local inflammatory response in the lungs and tissue destruction [16, 17]. Neutrophils are the first cells to migrate through the vascular epithelium and reach the site of infection, stimulated by the release of a concentration gradient of the chemokines CXCL1, CXCL2, leukotriene B4, and IL-8 (CXCL8), which acts on leukocyte rolling and induces stable binding between adhesion molecules, thereby favoring transmigration. Due to the progressive dysregulated inflammatory process, neutrophils exhibit low phagocytic capacity, delayed apoptosis, and increased formation of neutrophil extracellular traps during infection [18]. The large number of inflammatory mediators released during sepsis leads to disruptions in the endothelial layer, increasing its permeability, allowing plasma proteins and body fluids to cross the blood vessel wall, leading to edema formation and ultimately multiorgan dysfunction, also releasing inflammatory mediators into the plasma, leading to uncontrolled activation of the immune system and severe cytokine storm [19].

Cytokines such as IL-9 and IL-13 stimulate mucin production by lung epithelial cells, resulting in purulent secretions that may facilitate bacterial neutralization. These inflammatory mediators also promote disruption of the alveolar-capillary barrier, facilitating leukocyte migration to the site of infection, amplifying the inflammatory response. Although essential for host defense, this process can compromise pulmonary homeostasis, leading to decreased lung compliance, interstitial fibrosis, and progressive dyspnea. When the inflammatory process reaches the pleura, activation of somatic receptors of the phrenic nerve in the parietal pleura can cause pleuritic chest pain. In severe cases, parenchymal injury may progress to localized necrosis, tissue rupture, and bleeding, clinically manifesting as hemoptysis [9].

The reactive nitrogen species (RNS) such as nitric oxide (NO), and reactive oxygen species (ROS), released by the immune cells, play a significant role in vascular endothelial cell dysfunction during sepsis by direct damage or modulating innate immune by activating the transcription factor nuclear factor-kappa B (NF-κB), thus enhancing TNF-α, IL-1β, IL-6, IL-8, IL-17, and IL-18 secretion. The production of matrix metalloproteinases, which is mediated by various factors, including vascular growth factors such as angiopoietin-2, is increased during sepsis and is associated with a poor prognosis and damage to vascular endothelial cells. This process leads to the disruption of tight junction proteins, resulting in increased vascular permeability and edema formation [20]. Pulmonary infection can lead to sepsis, and in turn, acute respiratory distress syndrome (ARDS) is the most common consequence of sepsis of different origins, affecting transepithelial neutrophil migration, fibroproliferation, and activation of apoptosis, resulting in the loss of alveolar-capillary, compromising membrane integrity [21]. Mice with ARDS exhibit neuroinflammation, with an increase of microglial cells with fewer branches and expressing CD68 protein [22], indicating that sepsis may lead to alterations in different systems, having consequences on the central nervous system (CNS).

Sepsis-Associated Encephalopathy

Sepsis-associated encephalopathy (SAE) is the most frequent organ dysfunction. It is a primary contributor to neurological alterations observed in critically ill patients, affecting up to 70% of those with sepsis, with patients also presenting increased mortality, morbidity, and cognitive disability, with symptoms ranging from delirium and seizures to coma [23]. Even though SAE is usually categorized as a fully reversible pathophysiological process, there is increasing evidence suggesting an association between sepsis, structural brain injury, and long-term neurologic sequelae [24]. Inflammatory signaling reaches the CNS through three distinct pathways: (1) activation of afferent neural routes, such as the vagus nerve, in response to peripheral inflammatory stimuli; (2) humoral signaling, in which circulating cytokines act on the brain, choroid plexus, and circumventricular organs, including the blood-brain barrier (BBB); and (3) structural and functional alterations in the BBB caused by endothelial activation and the release of inflammatory mediators, which increase its permeability [25]. Understanding the different ways peripheral inflammation reaches the CNS is essential to fully comprehend and distinguish SAE from other neuroinflammatory or neurodegenerative diseases, since SAE necessarily follows a strong peripheral inflammatory response [26].

It has also been demonstrated that sepsis leads to bacterial dissemination to the brain for several days, suggesting that viable bacteria, rather than only peripheral inflammatory mediators, may contribute to the pathogenesis of brain dysfunction following sepsis. However, the precise nature of this relationship, whether causative or not, remains to be fully explained [27].

The CNS is one of the first regions exposed to an inflammatory episode due to the progression of peripheral inflammation caused by sepsis, allowing peripheral inflammatory mediators and immune cells into the brain. Basically, once inflammation in the periphery is established, it may become a neuroinflammation when the components of the inflammatory process gain access to the CNS [28, 29]. As introduced, despite the BBB, cells and cytokines can infiltrate the brain, initiating a neuroinflammatory process that results in alterations in neurons, astrocytes, and microglial cells [30, 31], and the consequences can include morphological alterations [32]. Neuronal cells and other glial cells, derived from the ectoderm, and microglial cells, which are derived from the mesoderm, share characteristics with immune cells. These cells sense their brain environment and respond by changing their phenotype or releasing molecules and vesicles. They also exhibit rapid clonal proliferation in response to insult. In disease, as well as in homeostasis, microglial cells initiate the process of synaptic pruning by phagocytosing synaptic dendrites that are no longer functional. This process activates microglia, the resident immune cells of the CNS, exposing the neurons to an environment of oxidative stress, leading to neuronal dysfunction and apoptosis. Dysregulated pro-inflammatory microglial responses play a significant role in SAE, with pro-inflammatory cytokines secreted by microglia and associated with brain damage, the first evidence in a sequence of phenomena that gives the microglia a central role in SAE [23]. Microglial interaction and co-activation with astrocytes also play a significant role in SAE, influencing the progression and severity of neurological disease by releasing large amounts of inflammatory cytokines that exacerbate the inflammatory response and brain damage. Recent studies have demonstrated that controlling the hyperactivation of microglia and astrocytes with orexin-A attenuates the inflammatory process and neuronal damage in the hippocampus during polymicrobial infection in mice [33]. Although inflammation is triggered by diverse stimuli, the responses of microglia and astrocytes are similar. The resulting modulation of neuroinflammation and decrease of the astrocyte marker GFAP bring the astrocyte to a central role in SAE together with the microglia.

Physiopathology of SAE and Neuroinflammation

An essential response that microglia may have to the environment is to recognize DAMPs and pathogen-associated molecular patterns through pattern recognition receptors and the release of cytokines [34, 35], similar to peripheral immune cells. Microglia respond to local and systemic modulators that enter through the BBB, acting as sensors and triggering the immune response in the CNS, the focus of numerous studies aiming to link inflammation to the brain.

When sepsis leads to SAE, microglial cells are at the center of the issue, responding to the insult with pro-inflammatory cytokines such as IL-1β, IL-18 [36], and TNF-α, which can exacerbate inflammation and cause neuronal death [30]. The upregulation of TNF-α secretion is linked to endothelial necroptosis, increased BBB permeability in a feedback loop response, and glutamatergic cytotoxicity. The IL-1β derived from microglia causes synaptic alterations associated with cognitive impairment as sequelae, resulting in synaptic elimination and inhibition. Microglial production of NO and ROS also has deleterious effects on neurons, primarily inducing apoptosis or exacerbating excitotoxicity, which results in neurological deficits [23]. Those findings reinforce the microglial role in inducing neuronal damage and further neurological consequences.

It has also been demonstrated that microglial synaptic pruning is altered in SAE models, leading to impaired synaptic plasticity and subsequent cognitive deficits, particularly within the hippocampus [37]. Microglial-disrupted function is closely associated with cognitive damage, in both acute phase of disease that causes sickness behavior, and in the resulting sequelae that occur later on, as cognitive damage.

The microglial cells can change their phenotype in response to the environment, so depending on the stimulus, these cells can reduce the number of their branches and alter the expression of proteins such as Iba1 and CD68. These events collectively characterize the so-called reactive to microglia, highlighting their ability to respond to stimuli [38].

To explain and categorize the distinct responses of microglia to different stimuli, as well as their consequent modulation and polarization, the terminology “M1” and “M2” microglia was widely adopted in microglial research. These cells were artificially classified into two opposing phenotypes based primarily on findings from in vitro models: the M1 pro-inflammatory microglia and the M2 anti-inflammatory and neuroprotective microglia. These phenotypes are associated with specific molecular markers and characteristic morphological changes [39].

However, over the years, it has increasingly been proposed that this classification system is overly simplistic and does not fully capture the complexity of microglial phenotypes. In particular, it lacks depth when examined through whole-genome transcriptomic, epigenomic, and proteomic analyses, and it overlooks the fact that multiple microglial subtypes may coexist within the CNS and that microglia exhibit varying characteristics in different environments. Although studies adopting the M1–M2 classification have significantly advanced our understanding of the central role of microglial responses in neuroinflammation, it has been debated the necessity of a broader spectrum, considering different combinations of markers, to fully understand these cells and consequently elucidating more therapeutic strategies targeting neuroinflammation [40].

Toll-like receptor 4 (TLR4) is a key transmembrane receptor involved in inflammatory processes. It is expressed both on the cell surface and within endosomes, primarily in immune and glial cells, where it plays a critical role in the induction of neuroinflammation. The myeloid differentiation primary response 88 (MyD88) is the most common adaptor protein interacting with the intracellular domain of TLR4. Upon activation, the TLR4-MyD88 signaling pathway triggers downstream cascades involving transcription factors such as NF-κB and mitogen-activated protein kinases (MAPKs), ultimately leading to the production of pro-inflammatory cytokines, such as TNF-α, IL-1β, and IL-6 [41, 42].

Bacterial endotoxins, or LPSs, are directly attached to TLR4 surface receptors of microglia and astrocytes, also altering the cellular metabolism of immune cells. Activation of the TLR4 signaling pathway also plays a central role in oxidative stress. Experimental evidence indicates that prolonged neuroinflammation promotes the generation of ROS, which is accompanied by a reduction in the expression of nuclear factor erythroid-2-related factor 2 (Nrf2) and its downstream antioxidant enzyme heme oxygenase-1 (HO-1) in the brains of LPS-treated mouse [43]. The Nrf2/HO-1 pathway exerts significant anti-inflammatory effects, with studies demonstrating its critical role in modulating inflammatory responses in BV2 microglial cells. Additionally, activation of this pathway protects mice and rats against endotoxic shock by suppressing the induction of inducible nitric oxide synthase and the subsequent production of NO [44].

Diverse models can be used to study sepsis, even though LPS should not be referred to as a sepsis model, but rather as an endotoxemia model, which acts primarily through the primary receptor for LPS, TLR4. That bacterial component initiates the signaling that results in the production of inflammatory mediators [45]. Demonstrating the LPS effect on the CNS, evidence indicates that LPS triggers astrocytic and microglial reactivity, leading to increased expression of pro-inflammatory cytokines and activation of the NOD-, LRR-, and pyrin domain-containing protein 3 (NLRP3) inflammasome pathway [46], and elevated levels of pro-inflammatory cytokines such as TNF-α, IL-6, and IL-1β. Furthermore, disruption of the BBB may potentially occur, allowing bacteria from the lungs to translocate to the brain [47]. This disruption may lead to lymphocyte infiltration in the brain [48], thereby exacerbating microglial inflammation.

In a model of endotoxemia caused by LPS, an increase in ROS is also observed in the cerebral cortex of mice [49, 50], indicating that the pathway is affected and may contribute to neuroinflammation. Studies suggest that pro-inflammatory cytokines trigger a neurotransmitter imbalance, which in turn affects neuronal activity [51]. The alteration of neurotransmitters may affect the maintenance of SAE, and changes in amino acids in the cerebrospinal fluid may alter neurotransmitter production and patient behavior [52]. Animals that received LPS injections showed a decrease in dopamine, serotonin, and norepinephrine levels in the hippocampus 24 h after the induction [53].

LPS stimuli in mice may lead to neuronal cell death, which can also be observed in some brain regions in patients after sepsis [54]. Single-nucleus RNA sequencing reveals gene modulation toward an inflammatory profile in the brains of mice injected with LPS, characterized by increased IL-1β expression and decreased TMEM119 protein levels, indicative of a microglial phenotypic shift in this SAE model [55, 56]. Supporting the role of microglia in neuroinflammation, Mein et al. [57] using PLX5622, a pharmacological agent that depletes microglial cells, prevented both neurocognitive deficits and neuroinflammatory responses in an LPS-induced endotoxemia model. The microglial depletion diminished acute astrogliosis and long-term microgliosis and prevented long-term neurocognitive decline in sepsis.

Aquaporin 4 functions as a selective water-conducting channel, and its increased expression may activate astrocytes, inducing an inflammatory response in the CNS, affecting BBB integrity, and releasing pro-inflammatory cytokines [58, 59]. Astrocytes can also release IL-33, a cytokine that may enhance the microglial response to inflammation, thereby contributing to the inflammatory development of SAE [60]. Once BBB integrity is compromised, astrocytes can become more activated, amplifying BBB permeability and neuroinflammation [61].

Although the relationship between tripartite motif-containing protein 45 (TRIM45) and sepsis remains unclear, TRIM45 expression is upregulated in the murine microglial cell line BV-2 following LPS stimulation. TRIM45 expression has been linked to the activation of the NF-κB signaling pathway during ischemic conditions, contributing to increased neuronal injury and neuroinflammation. In a murine model of SAE, TRIM45 expression was found to colocalize with microglial cells, and its silencing appeared to mitigate SAE-related brain damage [62], presenting longer term cognitive deficits [54]. NF-κB is one of the most studied pro-inflammatory pathways in sepsis-induced ARDS [63], and its activation is increased in the brain cortex in a model of endotoxemia caused by LPS [49]. Furthermore, NF-kB participates in the activation of a diverse range of pathways, including the NLRP3 inflammasome, leading to increased levels of pro-inflammatory cytokines, e.g., IL-1β and IL-18 [46]. The NLRP3 inflammasome may induce pyroptosis, a type of cell death initiated by inflammation, and molecules such as TRIM45 participate in that process. Following the overexpression of TRIM45, the upregulation of pyroptosis proteins occurs, and studies have demonstrated that their levels are elevated in SAE model mice [64]. Taken together, these findings underscore the impact of SAE on brain cells and activity.

Pneumosepsis, SAE, and Neuroinflammation

A model to study sepsis of lung injury as pneumonia, leading to pneumosepsis and neuroinflammation, uses intratracheal instillation. The presence of P. aeruginosa in the lungs leads to increased plasma levels of pro-inflammatory cytokines. There is also an increase in BBB permeability and pro-inflammatory cytokines in the cortex and hippocampus, demonstrating that peripheral inflammation accesses the brain, causing anxiety-like behavior [65]. These psychiatric symptoms are accompanied by elevated NLRP3 levels and its components, including caspase-1, ASC, GSDMD, and pro-inflammatory cytokines such as IL-1β [66], along with microglial alterations. Exosomes derived from patients with depression may further exacerbate microglia-mediated inflammation [67]. A similar response occurs in a coinfection with S. pneumoniae and influenza virus, with enhanced microglial activation and increased transcript levels of IL-6, TNF-α, and IL-1β in the paraventricular nucleus and hypothalamus [68]. K. pneumoniae also accesses the brain after entering the circulation and alters mouse behavior, increasing pro-inflammatory cytokine levels in the brain [46]. Cytokine release from microglia, such as IL-6, may elevate caspase-3 levels and disrupt astrocyte homeostasis, thereby influencing the apoptotic state of these cells [69]. Those findings bring once again microglia and astrocytes as essential players causing and exacerbating neuroinflammation in a pneumonia context besides endotoxemia and peritoneal sepsis. Similar to intratracheal instillation, another model uses noninvasive oropharyngeal administration of LPS and bacteria. The bacterial infection had a more pronounced effect on elevating pro-inflammatory cytokines than LPS, but both increased IL-1β, TNF, and IL-6 [70].

Evidence from intratracheal injection of K. pneumoniae demonstrated anxiety- and depression-like behaviors in mice, and increased TNF-α and IL-1β in the hippocampus and prefrontal cortex 24 h and 7 days after infection. At 14 days after infection, mice still exhibited elevated levels of TNF-α and IL-1β in the hippocampus, which may translate into increased Iba1 expression in the CA1 region of the hippocampus 30 days after infection [71]. Using a similar protocol, emerging evidence demonstrates elevated colony-forming unit levels in the brain 24 h after infection [72], corroborating the possibility of BBB disruption due to inflammation.

Furthermore, an intranasal model using K. pneumoniae showed decreased locomotor activity 14 days after infection and elevated levels of cytokines and DAMPs, such as S100A8, 72 h after infection. A significant event occurs after infection, involving the infiltration of myeloid cells, such as neutrophils and monocytes [13], which may exacerbate the inflammation. That evidence shows the role of circulating cells in exacerbating in brain during pneumonia.

Both astrocytes and microglia undergo alterations in meningitis models as well [73], and their pathological mechanisms largely overlap with those of sepsis. In a S. pneumoniae model, increased hippocampal cell death was observed, accompanied by elevated levels of the p75 form of brain-derived neurotrophic factor. It increased the activity of the p65 subunit of NF-κB in both the cortex and hippocampus 24 h after meningitis induction, accompanied by higher concentrations of IL-1β, TNF, IL-6, and inducible nitric oxide synthase [73]. Additional studies have demonstrated the presence of diverse immune cell populations in the brains of mice with meningitis, including neutrophils, microglia, dendritic cells, macrophages, monocytes, B cells, and CD4⁺ and CD8⁺ T cells, all of which progressively increased in number over the course of disease progression [74]. Those data do not prove glia as the main players, but we may suggest, based on kinetics, that microglia play an important role in the initial event of SAE. In vitro, microglia and astrocytes showed increased NF-κB activity and upregulated NLRP3 expression when primed with IL-6 and IFN-γ, respectively [75], supporting the notion that these cells produce and respond to these inflammatory mediators, contribute to BBB disruption, and exacerbate neuroinflammation.

Those phenomena are also evident in postmortem tissue, where septic patients present increased expression of microglial Iba1 marker [76], potentially impacting neuronal connectivity and survival. The interaction between microglia and neurons involves multiple proteins, including phosphatidylserine and calreticulin [77, 78], as well as complement system proteins such as C1q. Studies have demonstrated that a reduction in C1q protein expression leads to decreased synaptic pruning following sepsis [76], indicating that microglial action is mediated by a diverse array of molecules, and treatment strategies for SAE should consider this complexity.

Another potential consequence of SAE is ROS production. In a S. pneumoniae model using the BV-2 cell line, a direct increase in ROS-induced damage has been observed [79]. These alterations in BV-2 cells are also evident in RNA sequencing, which reveals upregulation of immune response-related genes, including those in the NF-κB pathway, indicating that this pathway may regulate cellular surveillance metabolism [80]. This microglial response generates distinct populations – or clusters – that adapt to the microenvironment, exhibiting differential protein expression across diverse functions, such as upregulation of TNF, C1qa, and CCL5, whereas proteins like TREM2 and P2RY12 are downregulated [81]. Taking all the data together, apparently several models of sepsis and disease, triggering different pathways, induce a similar cellular and molecular response, which may affect the BBB, and activate glial cells that decrease neuronal survival (shown in Fig. 1 and better described in Table 1), altering the behavior, and decreasing cognitive impairment.

Fig. 1.

Experimental models using pneumonia-inducing bacteria or their components cause a similar response, triggering BBB disruption and microglia reactivity. The cell activation occurs bidirectionally between astrocytes and the BBB. Inflammation also modifies microglia and astrocytes into a reactive form, increasing pro-inflammatory cytokine levels and altering neurotransmitter balance, thereby affecting neuronal survival.

Table 1.

Experimental models of neuroinflammation using pneumonia-inducing bacteria or their components and the main effects in mice

Study	Stimulus	Inflammatory response
Villalba et al. [65]	Intratracheal instillation of P. aeruginosa in C57BL/6 mice between 8 and 16 weeks of age	Presence of bacteria into lungs and an increase of pro-inflammatory cytokines in plasma after 24 h. In the brain, it was observed an increase of BBB permeability and pro-inflammatory cytokines on cortex and hippocampus
Wang et al. [68]	Intranasally coinfection with S. pneumoniae and influenza virus in male C57BL/6J mice (8–10 weeks old)	Enhanced microglia activation and increased transcript levels of IL-6, TNF-α, and IL-1β in the paraventricular nucleus and hypothalamus
Ma et al. [47]	LPS-induced intratracheal inflammation of female KM mice (4 weeks)	Increased microglial and astrocyte reactivity and elevated levels of TNF-α, IL-6, and IL-1β
Scheffzük et al. [70]	Oropharyngeal application of LPS and P. aeruginosa in adult male wild-type and SP-A-deficient mice	Increased levels of IL-1β, TNF, and IL-6
Bonorino et al. [71]	Intratracheal injection of K. pneumoniae in male Swiss mice (6–8 weeks old)	Anxiety- and depression-like behaviors in mice, and increased TNF-α and IL-1β in the hippocampus and prefrontal cortex
Denstaedt et al. [13]	Intranasally infection with K. pneumoniae in male C57BL/6J mice (8–12 weeks old)	Decreased locomotor activity 14 days after infection and elevated levels of IL-1β, TNFα, and IL-17 in response to secondary LPS stimulation in pneumosepsis survivors

Conclusion and Final Remarks

Herein, we described and focused on aspects of neuroinflammation underlying pneumonia and SAE. The primary targets after triggering BBB dysfunction are, most of all, microglia and astrocytes, acting in collaboration, resulting in neuronal death and neurological and cognitive sequelae. Glial cell hyperactivation causes disturbances in neuronal plasticity, leading to the acute neurological symptom of sepsis, the sickness behavior. The interplay between microglia and astrocytosis is also essential to recover homeostasis after the insult and excessive co-stimulation lead to the further sequelae as cognitive impairment, learning, and memory loss. Some intracellular mechanisms and death pathways are described. Many gaps remain, as there are not enough reports showing cause-and-consequence relations. It also remains a matter of discussion and evidence whether bacteria opportunistically reach the SNC following the alteration of BBH permeability, or their released components activate the glial receptors. Cytokines also play a crucial role in the SAE, and they can themselves induce the following phenomena caused by bacteria, which are essential for exacerbation and disease progression during the sterile phase of the disease, when no bacteria are further detected. Understanding the basic mechanisms of cellular communication and interaction will help manage diseases caused by AMR bacteria worldwide.

Future therapeutic strategies will likely require precision approaches that selectively suppress detrimental glial responses while preserving protective functions. Integrating antimicrobial therapy with targeted immunomodulation, supported by advances in single-cell profiling, neuroimaging, and biomarker discovery, may enable more precise patient stratification and improve translational success. Ultimately, a deeper understanding of glial biology in systemic infection will be essential for developing interventions to prevent or reverse the neurological burden of pneumonia-driven sepsis and SAE.

Preclinical evidence increasingly supports targeting glial cells to attenuate neuroinflammation in SAE. In experimental models, pharmacological modulation of microglial activity has produced promising results. As discussed earlier, partial inhibition of microglial survival via CSF1R blockade with agents such as PLX5622 reduces synaptic engulfment, astrogliosis, and long-term post-sepsis neurocognitive decline, indicating that precise modulation – rather than complete ablation – of microglia is key. Neuropeptides such as orexin-A are also attractive because they reduce oxidative stress, suppress ERK/NF-κB signaling, and limit pro-inflammatory microglial activation and A1-type astrocyte polarization, thereby preserving neuronal integrity and improving cognition. Additionally, targeting intracellular inflammatory pathways, particularly the NLRP3 inflammasome, is promising given its central role in glial pyroptosis, cytokine amplification, and chronic neuroinflammation in SAE models.

Despite these advances, several obstacles still limit the clinical translation of glia-focused therapies. The pronounced functional and phenotypic heterogeneity of glial cells across brain regions, disease stages, and systemic inflammatory states makes it difficult to define optimal therapeutic targets and treatment windows. Broad suppression of microglial or astrocytic activity can also disrupt essential immune surveillance and repair, increasing infection risk or delaying recovery in septic patients. Translation is further hindered by interspecies differences between animal models and humans, the lack of reliable in vivo biomarkers of glial activation, and the complex interplay between central and peripheral immune responses, particularly amid rising AMR.

Future therapies will likely require precision approaches that selectively attenuate detrimental glial responses while preserving protective functions. Combining antimicrobial therapy with targeted immunomodulation, supported by advances in single-cell profiling, neuroimaging, and biomarker discovery, may enable better patient stratification and improve the chances of successful clinical translation. Ultimately, deeper insight into glial biology during systemic infection will be essential for developing interventions that prevent or reverse the neurological burden of pneumonia-driven sepsis and SAE.

Conflict of Interest Statement

The authors declare no conflicts of interest.

Funding Sources

This research was funded by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Instituto Oswaldo Cruz, FIOCRUZ, and Universidade Federal do Estado do Rio de Janeiro (UNIRIO).

Author Contributions

A.R.S. and C.F.G.A.: conceptualization, funding acquisition, and writing – review and editing. A.M.S.F., L.V.G.C., and M.F.C.: writing – original draft.

Funding Statement

This research was funded by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Instituto Oswaldo Cruz, FIOCRUZ, and Universidade Federal do Estado do Rio de Janeiro (UNIRIO).