Endocannabinoid System in Sepsis: A Scoping Review

Abstract

Sepsis is a life-threatening syndrome marked by a dysregulated host response to infection, resulting in systemic inflammation, organ dysfunction, and high mortality globally. Despite advancements in supportive care, effective immunomodulatory therapies remain elusive, necessitating exploration of novel biological pathways and subsequent therapeutic development. The endocannabinoid system (ECS), which regulates immune function and homeostasis, has emerged as a key modulator of immunological and metabolic pathways central to sepsis pathophysiology. The ECS mediates its effects through endogenous ligands, G-protein-coupled cannabinoid receptors (CB1 and CB1), and regulatory enzymes that control its synthesis and degradation. Following PRISMA-ScR guidelines, this scoping review synthesizes current evidence on the mechanistic roles of ECS components in experimental and clinical models of sepsis, identifies knowledge gaps, and delineates future areas of work. A comprehensive literature search across multiple databases without restrictions on date or publication type was executed to ensure broad coverage of original studies investigating ECS mechanisms and their intersection with sepsis and septic shock. Across 53 studies, CB2 receptor activation was consistently associated with anti-inflammatory process, organ-protective outcomes, and increased survival rates against septic challenges in preclinical rodent models. CB1 receptor activation trends, however, showed context dependent outcomes. Central antagonism improved hemodynamics and survival rate, but peripheral effects varied with cell type and timing. Non-canonical ECS components (TRPV1, GPR55, PPAR-α, FAAH, MAGL) also contributed to neuroimmune and metabolic regulation. Limited clinical data linked ECS lipid profiles and gene expression with sepsis severity and outcomes. Collectively, ECS modulation, particularly CB2 agonism, TRPV1 activation, and FAAH/MAGL inhibition, shows promise in mitigating sepsis-induced inflammation and organ dysfunction. However, complex, context-dependent effects, especially involving CB1, highlight the need for precision-targeted therapeutic approaches. Further preclinical research is needed to expand generalizable trends to allow translational research to refine ECS-based interventions for sepsis management.

1. Introduction

Sepsis continues to be recognized as one of the foremost contributors to global morbidity and mortality rates, characterized as a dysregulated response from the host to infectious agents that leads to an extensive and often deleterious cascade of inflammation, subsequent organ dysfunction, and outcomes that can frequently result in death. Current epidemiological estimates indicate that there are approximately 49 million cases of sepsis reported annually, with around 11 million fatalities attributed to this condition, which collectively accounts for nearly 20% of the total mortality observed worldwide [1]. In addition to the significant clinical implications of sepsis, it also imposes a considerable economic burden on healthcare systems. This condition is consistently identified as the most financially burdensome ailment encountered within hospital settings, a status that is largely driven by the necessity for prolonged stays in intensive care units (ICUs), the execution of invasive medical procedures, and the alarming rates of patient readmission that are often associated with sepsis. In the context of the United States, it has been reported that the annual financial expenditures related to the care and management of sepsis surpassed $38 billion as of the year 2017 [2].

Individuals who survive an episode of sepsis frequently confront a range of long-term health complications, which may include, but are not limited to, cognitive dysfunction, chronic kidney disease, and various cardiovascular issues [3]. Despite significant advancements in the provision of supportive medical care, there remains a notable absence of immunomodulatory therapies that have consistently proven to offer a tangible survival advantage to patients suffering from sepsis. The inherent clinical variability of sepsis—spanning from initial hyperinflammatory responses characterized by cytokine storms to the late-phase phenomenon of immunoparalysis—has introduced considerable challenges in the formulation of effective targeted therapeutic interventions. In light of its profound global implications, the World Health Organization has formally classified sepsis as a crucial public health concern and has identified it as the leading cause of mortality among children around the globe [4]. The repeated shortcomings observed in various immunomodulatory therapeutic strategies, which include corticosteroids, anti–TNF-α agents, and immune checkpoint inhibitors, highlight the pressing necessity for the development of innovative mechanistic understandings and novel therapeutic modalities [5,6,7,8].

The endocannabinoid system (ECS), which is a complex lipid-based signaling network that consists of endogenous ligands, G-protein-coupled receptors—primarily CB₁ and CB₂—and associated metabolic enzymes, has garnered attention as a significant regulator of immune homeostasis and inflammation within the body [9]. Initially identified for its involvement in neurobehavioral processes, the activity of the ECS is now recognized to exert influence over a broad spectrum of physiological functions, which encompass vascular tone regulation, cellular metabolic processes, and the differentiation of immune cells [10]. Considering the perturbations of these critical pathways that occur during sepsis, the modulation of the ECS represents a particularly compelling therapeutic opportunity for intervention.

Preclinical studies spanning murine and rat models of polymicrobial sepsis, endotoxemia, acute respiratory distress syndrome (ARDS), and septic encephalopathy have not revealed distinct ECS-mediated mechanisms [11,12,13,14,15]. CB2 receptor activation consistently exerts anti-inflammatory effects by tempering neutrophil influx, repressing pyroptosis-related pathways (e.g., NLRP3–caspase-1–GSDMD), and promoting immunoregulatory phenotypes such as M2 macrophages and FOXP3⁺ T cells [16,17,18,19]. Conversely, the signaling of the CB1 receptor demonstrates a dichotomous influence, whereby central antagonism may mitigate septic hypotension and cognitive deterioration, whereas peripheral modulation of CB1 may affect metabolic and immunological functions [20,21].

Emerging data also highlights the role of non-canonical ECS actors—such as TRPV1, GPR55, and PPAR-α—and enzymatic regulators like FAAH and MAGL, revealing intricate layers of neuroimmune crosstalk, metabolic reprogramming, and organ-specific protection [22,23,24,25,26]. Clinical observations, albeit limited, lend further credibility to ECS relevance, with circulating ligand profiles and transcriptomic signatures correlating with sepsis outcomes [27,28,29,30].

Given the breadth and heterogeneity of ECS research in sepsis, including variations in animal models, ligands, dosing, strategies, and outcomes, a scoping review is the most appropriate direction to map existing evidence. This scoping review aims to synthesize the mechanistic landscape of ECS in sepsis by cataloging experimental models, receptor-specific pathways, and immunological outcomes, thereby identifying knowledge gaps and guiding future translational efforts. This methodology allows for comprehensive data review while acknowledging the limitations of comparability across studies. Through critical evaluation of the literature, we explore whether ECS modulation can fulfill the long-standing therapeutic void in sepsis management and pave the way toward precision-targeted interventions.

2. Methods

This scoping review was guided by the framework developed by Arksey and O’Malley (2005) [31]. The purpose of this review is to examine the extent and breadth of available original literature for the ECS and its involvement with sepsis. As such, critical appraisal of individual sources is not required. To ensure adherence to structured, systematic reporting, this scoping review followed the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) [32] methodology and checklist.

2.1. Protocol and Registration

A review protocol was developed and registered on the Open Science Framework (OSF) [33]. The methods were established a priori and adhered to throughout the review process. While scoping review protocol registrations are not required, this registration ensures transparency and further aligns with PRISMA-ScR recommendations.

2.2. Information Sources

Following consultation with a qualified medical librarian, a comprehensive literature search was conducted across several medical and scientific databases, including Ovid (Medline), Central, Scopus, and CINAHL, all of which were executed on 13 June 2025. The objective was to identify studies investigating the mechanisms of the endocannabinoid system (ECS) during sepsis and septic shock. The search used controlled vocabulary to increase coverage such as: MeSH terms (e.g., “Sepsis”, “Endocannabinoids”), keywords (e.g., “bloodstream infection”) and Boolean operators (AND/OR). A key challenge in searching this field was a lack of standardized terminology and indexing. Relevant papers occasionally described individual components (e.g., CB1, CB2) without explicitly using “endocannabinoid system” as a keyword or may not have been indexed under our searched terms. Additionally, given the relatively niche research intersection between sepsis and ECS, there were no constraints regarding animal model, publication year, subject type, or publication type. These factors may have increased the initial pool of records, but ensure broad coverage of relevant terms and their subcategories. (for full search strategy see Appendix A

Table A1. Full Literature Database Search Strategy. * Identifies variations in the root word, “exp” and “mh” is a search field tag for “Exploded MeSH” which will include all related subcategories with functionality similar to “INDEXTERMS”, “NEXT” requires the search to have the two terms in immediate proximity, “.mp.” is a multi-purpose field search similar to “:ti,ab,kw” to search for text in the title, abstract, and keywords.

Ovid(Medline): Search ran on 6/13/25 and produced 134 results. 1. exp Sepsis/or (sepsis or “bloodstream infection *” or septicemia * or “blood poisoning *” or pyemia * or pyaemia * or pyohemia * or septic *).mp. 2. exp Endocannabinoids/or (endocannabinoid * or endo-cannabinoid *).mp. 3. exp Receptors, Cannabinoid/or “cannabinoid receptor *”.mp. 4. 2 or 3 5. 1 and 4

Cochrane Trials (CENTRAL): Search ran on 6/13/25 and produced 4 results. 1. [mh Sepsis] OR (sepsis:ti,ab,kw OR (“bloodstream” NEXT infection *):ti,ab,kw OR septicemia *:ti,ab,kw OR (“blood” NEXT poisoning *):ti,ab,kw OR pyemia *:ti,ab,kw OR pyaemia *:ti,ab,kw OR pyohemia *:ti,ab,kw OR septic *:ti,ab,kw) 2. [mh Endocannabinoids] OR (endocannabinoid*:ti,ab,kw OR endo-cannabinoid *:ti,ab,kw) 3. [mh “Receptors, Cannabinoid”] OR (“cannabinoid” NEXT receptor *):ti,ab,kw 4. #2 OR #3 5. #1 AND #4

Scopus: Search ran on 6/13/25 and produced 164 results. (INDEXTERMS (sepsis) OR TITLE-ABS-KEY (sepsis OR “bloodstream infection *” OR septicemia * OR “blood poisoning *” OR pyemia * OR pyaemia * OR pyohemia * OR septic *) AND (INDEXTERMS (endocannabinoids) OR TITLE-ABS-KEY (endocannabinoid * OR endo-cannabinoid *) OR INDEXTERMS (“Receptors, Cannabinoid”) OR TITLE-ABS-KEY (“cannabinoid receptor *”)))

CINAHL: Search ran on 6/13/25 and produced 11 results. (MH “Sepsis+” OR (sepsis OR “bloodstream infection *” OR septicemia * OR “blood poisoning *” OR pyemia * OR pyaemia * OR pyohemia * OR septic *)) AND ((endocannabinoid * OR endo-cannabinoid * OR “cannabinoid receptor *”))

)

2.3. Eligibility Criteria and Study Selection

This scoping review focuses on studies that investigate mechanisms of the ECS as it relates to sepsis and septic shock. Specifically, studies were included if they: (i) contain original experimental data with a mechanistic focus of the ECS; (ii) have direct experimental modulation of at least one of the ECS’s components (e.g., receptor expression, ligand modulation, enzymatic activity); (iii) utilize a recognized septic model (e.g., cecal ligation and puncture (CLP), lipopolysaccharide (LPS), bacterial infection) or experimental samples from septic patients; and (iv) were published in English. Studies were excluded if they: (i) did not contain original experimental data (reviews, editorials, etc.); (ii) did not elicit mechanistic understanding through direct modulation of ECS components (e.g., only clinical/observational associations); (iii) lacked an established sepsis model; or (iv) were non-English. Using these criteria, the original 253 results were screened independently by 2 authors (B.T. and H.Y.) for a final selection of 53 included studies. Any discrepancy was resolved through consensus among the authors. This process was facilitated using Rayyan (Rayyan Systems Inc.; https://www.rayyan.ai), an online systematic review management platform accessed in June 2025.

Although we did not perform a formal risk-of-bias assessment in line with scoping review methodology, it is notable that many included studies incorporated internal validity checks. For example, several studies confirmed receptor specificity by combining agonist administration with receptor antagonism or by comparing wild type to receptor knockout animals. These “opposite-control” designs strengthen the causality, ensuring that observed effects are attributable to modulation of the ECS and not due to unknown off-target interactions [34].

2.4. Data Charting

Data extraction was independently performed by the screeners using Microsoft Excel, with disputes resolved by author consensus. After several revisions, the charted data was finalized as: first author, year of publication, animal model system (e.g., C57BL/6 mice), sepsis model (e.g., CLP, LPS), experimental intervention, key mechanistic findings, outcomes, and summary. Study interventions not relevant to the scope of the review or our understanding of the study were excluded (see Appendix A for full detailed data chart). The extracted data was summarized by grouping individual studies based on their investigated ECS component(s) (e.g., CB1, CB2, enzyme, etc.) along with calculating percentages of data columns to show trends in study design.

As this was a scoping review, no formal statistical analyses or meta-analyses were performed. Instead, results were synthesized descriptively. Frequencies and proportions were calculated by dividing the number of studies reporting a given characteristic (e.g., model type) by the total number of studies included. Percentages are displayed to summarize trends across the various study categories (

Table 1. Study Characteristics. * Denotes that values are not mutually exclusive.

Characteristic	Category	n (%)
Animal Model *	Mice	31(58%)
	Rats	21 (40%)
	Other	5 (9%)
Sepsis Model *	LPS	37 (70%)
	CLP	17 (32%)
	Bacterial Exposure	2 (4%)
Primary ECS Target	CB1R	12 (23%)
	CB2R	17 (32%)
	CB1R + CB2R	7 (13%)
	Receptor Ligand	8 (15%)
	ECS Enzyme	5 (9%)
	Noncanonical Receptor	4 (8%)
Organ/Tissue Studied *	Lung	26 (49%)
	Heart	15 (28%)
	Brain	10 (19%)
	Kidney/Liver	5 (9%)
Publication Year	1998–2004	4 (8%)
	2005–2009	13 (25%)
	2010–2014	10 (19%)
	2015–2019	7 (13%)
	2020–2025	19 (36%)

and

Table 2. Themes of Septic Responses to ECS Modulation. This table lists the number of studies grouped by their ECS modulation and resulting septic response. Please note that the categories are not mutually exclusive. ↑ indicates upregulation, while ↓ denotes downregulation.

ECS Component	Modulation Type	Studies (n)	Direction of Effect	Summary of Outcome Trends
CB1R	Activation	4	↑ Sepsis Severity	↑ Mediates hypotension/hypothermia and inflammatory damage
	Activation	6	↓ Sepsis Severity	↑ Increases Il-10 production, reduced inflammatory markers (TNF-α and Il-12), increases survival
	Inhibition/Knockout	0	↑ Sepsis Severity	N/A
	Inhibition/Knockout	13	↓ Sepsis Severity	↓ Stabilize hemodynamics, through reduction in hypotension/dilation and increases response to vasopressors
CB2R	Activation	1	↑ Sepsis Severity	↑ Reduces IL-10 production in CD4-T cells, increasing lung injury
	Activation	17	↓ Sepsis Severity	↑ Reduces NLRP3 inflammasome related pyroptosis, leukocyte adhesion, pro-inflammatory cytokine (IL-6, TNF-α). Increases autophagy, especially in macrophages
	Inhibition/Knockout	2	↑ Sepsis Severity	↓ Results in increased leukocyte adhesion/influx
	Inhibition/Knockout	2	↓ Sepsis Severity	↓ Decreases mortality and organ damage, improves septic ileus
TRPV1	Activation	2	↓ Sepsis Severity	↑ Increases IL-10 production, reduces IL-6
FAAH	Modulation	2	Mixed Sepsis Severity	↑ Reduced septic pregnancy loss, ↓Protected gut microcirculation
MAGL	Inhibition	1	↓ Sepsis Severity	Reduced leukocyte adhesion and gut inflammation
AEA	Modulation	2	Mixed Sepsis Severity	↑ Reduced lung injury, ↓ Reduced shock survival
Δ9-THC	Administration	1	↓ Sepsis Severity	Reduced oxidative-nitrative cardiovascular stress

).

3. Results

3.1. Study Identification and Inclusion

In the initial search conducted by the librarian, we identified 253 studies across 4 databases. Using an automated Rayyan program, 73 duplicate studies were removed following manual approval from the authors. 180 records remaining for screening. Of the 180, 106 records were excluded based on title and abstract due to lack of relevance or adherence to the stated inclusion criteria. The remaining 74 records were sought for retrieval, with 5 being unavailable. Thus, 69 studies were available for full-text eligibility screening, of which 16 were excluded due to failure to meet criteria as listed in the PRIMSA flow diagram (Figure 1). A final total of 53 studies remained for data extraction and further analysis. This methodology ensures a more comprehensive review that should reflect the rigor as defined by PRISMA-ScR guidelines [32].

Each stage of the study selection is listed, with showcased reasons for narrowing of the article pool.

PRISMA, Preferred Reporting Items for Systematic Review and Meta-Analyses; ECS, endocannabinoid system

3.2. Characteristics of Sources

A comprehensive overview of the proportions pertaining to the animal models utilized in the included studies, the sepsis models employed, the ECS targets investigated, the organ tissues examined, and the respective publication years are presented in Table 1. An elaborated account of the extracted data is provided in Appendix A.

Among the 53 studies incorporated in this analysis, a notable augmentation in publication frequency was observed post-2020 (36%), potentially associated with advancements in technology and heightened focus on immune dysregulation in the context of the COVID-19 pandemic (2020+) (Figure 2). A majority of the investigations concentrated on murine models (58%) employing lipopolysaccharide (LPS) (70%) to simulate a septic condition. LPS constitutes the predominant percentage within its corresponding study characteristic across all classifications, yielding findings that support the involvement of Gram-negative bacteria. Additionally, lung tissue and the cannabinoid receptor type 2 (CB2R) were subjects of substantial inquiry, attributed to the extensive distribution of CB2R among immune cells and the prevalent pulmonary manifestations associated with sepsis.

This plots the number of publications for the included studies over time. It is noted that 3 of the 5 highest years for publications are after 2020, indicating growing interest in ECS immunoregulatory effects.

3.3. Study Results and Outcomes

A total of fifty-three scholarly investigations were incorporated into the conclusive analysis and categorized thematically according to their principal endocannabinoid system (ECS) target of action. The majority of these studies, to varying extents, examined the cannabinoid receptor type 2 (CB2R) (42%), revealing two predominant patterns: the activation of CB2R is associated with the promotion of anti-inflammatory responses and the enhancement of survival via organ-protective mechanisms (e.g., amelioration of lung injury [16,35,36,37,38,39,40,41] and attenuation of neuroinflammation [42,43]) (Table 2). In contrast, it was frequently demonstrated that the inhibition of cannabinoid receptor type 1 (CB1R) mitigates the severity of sepsis in 13 out of the 23 studies that predominantly concentrated on CB1R. A recurring theme was the stabilization of hemodynamics through either the reduction in hypotensive/dilatory effects [44,45,46] or the enhancement of activity/sensitivity to vasopressor modulation (including norepinephrine and arginine vasopressin) [47,48,49,50].

Across 53 preclinical and clinical studies, the endocannabinoid system (ECS) emerged as a multifaceted modulator of sepsis pathophysiology. Of the 53 studies, 18 rodent models sought to illicit the mechanistic insights regarding CB2 activation in sepsis. The majority (10/18) reported reductions in pro-inflammatory cytokines, most notably: IL-6, IL-1β, and TNF-α. In eight studies, CB2 agonists such as HU-308, JWH-133, and AM-1241, attenuated organ-damage during septic challenge. This organ protection was most frequently observed in lung (6 studies), followed by heart (1 study) and brain (1 study) tissue. One study using a CLP sepsis model, found that JWH-133 agonism of CB2 shielded lung, liver, brain, and heart tissue in rats. Histopathological scores were dose-dependent, dropping to below 50% at the max 5 mg/kg dosage [36]. Many of the studies presenting decreased organ injury also had corresponding attenuation of pyroptosis markers (4 studies) (e.g., NLRP3, GSDMD) or NF-κB (2 studies). Reduction in leukocyte adhesion was also noted (6 studies), mainly in lung or intestinal tissue. Survival improvement was reported in 4 studies, with the longest duration being a decrease in 7-day mortality rates [39]. Mechanistically, CB2 signaling promoted anti-inflammatory polarization of microglia and macrophages [35,42], repressed dendritic cell maturation [40], and supported autophagy flux [38,51,52]. Interestingly, 1 study found that CB2 activation of CD4-T-cells resulted in decreased IL-10 production, increased lung injury, and decreased survival [53]. CB2 inhibition/knockout presented with a comparably more varied septic outcome. Of the 4 studies with CB2 inhibition as the primary investigation, 2 displayed an increase in leukocyte adhesion, either through AM630 antagonism or genetic deletion [17,54]. Two additional studies where CB2 inhibition was a secondary experiment also showed an increase in or maintenance of pro-inflammatory leukocyte influx [55,56]. Other studies suggested that CB2 signaling impairs antibacterial defense, with its downregulation resulting in less lymphoid apoptosis and aversion of gut stasis (Appendix A 

Table A2. Full Detailed Data Collection Chart. This table lists the entirety of the data collected from the included studies. Arrows denote regulatory direction: ↑ indicates upregulation/increase, ↓ indicates downregulation/decrease, and → indicates a causal relationship.

Study (1st Author, Year)	Model System	ECS Component	Intervention/Comparison	Key Mechanistic Finding(s)	Outcome(s)	Quick Summary
Caraceni, 2009 [83]	Sprague-Dawley rats; partial hepatic ischemia–reperfusion (IR) + LPS; in vivo	CB1R	CB1 antagonist (Rimonabant) vs. vehicle at 3 and 10 mg/kg	CB1 antagonism reduced neutrophil infiltration, modulated cytokines (↓ TNF-α, ↑ IL-6, IFN-γ), and improved STAT3 signaling	Reduced liver injury (↓ ALT, necrosis), improved hemodynamics, reduced oxidative stress	Rimonabant mitigated LPS-enhanced hepatic IR injury via CB1 antagonism, modulating inflammation and oxidative stress.
Gardiner et al., 2005 [78]	Sprague-Dawley rats; 24 h LPS infusion (150 µg kg−1 h−1); conscious	CB1R	CB1 antagonist AM251 (3 mg kg−1) pretreatment	AM251 inhibited LPS-induced tachycardia & hind-quarter vasodilation, an effect mimicked by β-adrenoceptor blockade	Partial normalization of regional vascular conductance	Suggests ECS amplifies sympathetic beta-adrenergic vasodilator drive during sustained endotoxemia.
Joffre 2020 [64]	C57BL/6 mice, LPS endotoxemia, in vivo	CB1R	Δ9-THC (agonist) ± SR141716 (CB1 inverse agonist)	CB1 activation sharply increased IL-10 from Mo-MDSCs and suppressed IL-6/CCL-2 and NF-κB activation	Lower clinical sepsis score, reduced lung & spleen injury	Cannabinoid agonism can be protective when it drives early IL-10 via CB1R.
Kadoi et al., 2005 [46]	Male Wistar rats; bolus LPS (10 mg kg−1 i.v.); in vivo	CB1R	AM281 antagonist (1 mg kg−1 i.v.) given with LPS	Prevented systemic hypotension and carotid blood-flow fall; moderated TNF-α/IL-1β rise	Improved hemodynamics & halved 12 h mortality	CB1 blockade stabilized circulation and enhanced short-term survival during endotoxic shock.
Kadoi, 2005 [75]	Male Wistar rats; cecal ligation & puncture (CLP);	CB1R	CB1 antagonist AM281 (1 mg kg−1 i.p.) vs. vehicle	AM281 blunted caspase-3 activation in hippocampus and preserved brainstem reflexes	↓ Neurologic dysfunction & mortality after CLP	Blocking CB1 improved survival and neurological outcomes in septic rats.
Kadoi, 2008 [44]	Streptozotocin-induced diabetic and non-diabetic rats; LPS-induced endotoxemia; in vivo	CB1R	CB1 antagonist (AM281) vs. vehicle	AM281 reduced hypotension in diabetic rats more effectively; affected survival rates	Improved blood pressure and survival in diabetic rats	CB1 antagonism by AM281 improved hemodynamics and survival in diabetic endotoxemia model.
Kianian 2014 [93]	Male Lewis rats, LPS-induced endotoxemia—in vivo	CB1R	AM281 (CB1 antagonist) ± ACEA (CB1 agonist)	CB1 blockade lowered leukocyte adhesion in intestinal venules and restored functional capillary density (FCD)	Improved gut microcirculation; no hypotension reported	Blocking CB1 during early endotoxemia rapidly normalized intestinal microvascular perfusion, suggesting CB1 drives microvascular inflammation.
Leite-Avalca 2016 [47]	Male Wistar rats, CLP (severe, 3 punctures); central i.c.v. cannula (in vivo)	CB1R	CB1 antagonist Rimonabant (oral or i.c.v.) vs. vehicle	CB1 blockade boosted plasma vasopressin and dropped core temperature without altering peritoneal bacterial load	↑ 12-h and 5-day survival after CLP	Inhibiting CB1 increases AVP release and improves survival in late septic shock.
Leite-Avalca 2019 [48]	Wistar male rats, CLP (3 × 16 G punctures), in vivo	CB1R	Rimonabant 10 mg kg−1 p.o. at 4 h post-CLP (antagonist) vs. vehicle	Blockade did not change NOx but normalized aortic AVP reactivity and tempered AVP hyper-responsiveness	↓ plasma lactate, LDH, CK-MB; attenuated tail-artery vasoconstriction; improved organ-dysfunction panel	Late-phase CB1 antagonism limits metabolic/vascular dysfunction in sepsis.
Varga 1998 [57]	Sprague-Dawley rats, LPS 15 mg kg−1 i.v.; plus adoptive transfer of LPS-stimulated macrophages/platelets; in vivo	CB1R	LPS or transfer ± CB1 antagonist SR141716A 3 mg kg−1 i.v.	LPS ↑ 2-AG in platelets & anandamide in macrophages → CB1-mediated vasodilation; SR141716A blocks hypotension	SR141716A prevents hypotension & improves survival after LPS	Macrophage anandamide + platelet 2-AG paracrine drive septic hypotension via CB1; antagonist rescues BP and survival.
Vercelli 2009 [94]	Pregnant mouse uterine explants; LPS challenge; in vitro	CB1R	CB1 antagonist AM-251 vs. vehicle (±exogenous AEA)	LPS raises uterine AEA (↓ FAAH) and drives iNOS-derived NO; blockade of CB1 abolishes NO spike and tissue injury	AEA–CB1 promotes NO-mediated uterine damage in septic abortion model	Endotoxin uses local ECS tone (AEA/CB1) to amplify inflammatory NO production
Villanueva 2009 [45]	Sprague-Dawley rats; i.v. LPS endotoxemia; in vivo	CB1R	I.c.v. rimonabant (CB1 antagonist) vs. vehicle	CB1 blockade prevents early & late hypotension and blunts LPS-evoked NE overflow in pre-optic anterior hypothalamus, plus lowers plasma TNF-α	Maintained mean arterial pressure; ↓ systemic TNF-α	Central CB1 activation is required for endotoxic hypotension
Çakır 2020 [36]	Sprague-Dawley rats + CLP sepsis	CB2R	JWH-133 (agonist) 0.2–5 mg kg−1	CB2 activation lowers NF-κB, caspase-3 and pro-inflammatory cytokines in brain, lung, liver & heart	↓ Multi-organ histopathology; ↑ IL-10	Systemic CB2 agonism shields multiple organs from CLP-induced damage.
Chen 2022 [53]	C57BL/6 mice; CLP sepsis; in vivo lung (plus CD4-specific Cnr2 knock-out)	CB2R	HU-308 (CB2 agonist) vs. vehicle; CD4-Cre Cnr2^fl/fl vs. WT;	CD4-T-cell CB2R signaling suppresses IL-10; loss of CB2R switches the effect	↓ Survival, ↑ lung injury with HU-308; protection in CD4-CB2-KO	CB2 activation on CD4 T cells hampers IL-10 and worsens CLP sepsis.
Chen 2025 [42]	C57BL/6 mice, CLP sepsis; BV-2 microglia in vitro, brain	CB2R	HU-308 agonist ± AM-630 inverse agonist; Nogo-B over-expression	CB2R activation ↓ Nogo-B, shifts microglia M1→M2, ↓ IL-1β/IL-6/TNF-α; neuroprotection	↑ Cognitive scores; Nogo-B over-expression blunted benefits	CB2 activation represses Nogo-B, steering microglia to a pro-repair M2 state and easing septic encephalopathy.
Csóka 2009 [82]	CB2-KO vs. WT mice; CLP polymicrobial sepsis; in vivo	CB2R	Genetic knockout of CB2	Loss of CB2 → ↓ NF-κB activation, ↓ IL-10/IL-6/MIP-2, less lymphoid apoptosis, more immune cells	Lower mortality, ↓ bacteremia & organ (kidney, muscle) injury	CB2 signaling impairs antibacterial defense; antagonism could be beneficial
Gui 2013 [81]	C57BL/6 mice, high-dose LPS endotoxemia—in vivo & ex vivo splenocytes/macrophages	CB2R	GW405833 (CB2 agonist) and CB2-/- knockout	Agonist inhibited ERK1/2, STAT3 and NF-κB activation in macrophages; CB2-/- mice showed exaggerated cytokine storm	GW405833 improved 72 h survival and lowered serum TNF-α/IL-6/HMGB1	Genetic and pharmacologic evidence that CB2 signaling restrains pro-inflammatory pathways and improves survival in acute LPS sepsis.
Kapellos 2017 [17]	C57BL/6 WT vs. CB2−/− mice, LPS 10 mg kg−1 i.p., in vivo	CB2R	Genetic deletion	CB2 deficiency significantly increased rolling/adhesion and neutrophil influx into lung & liver (pro-recruitment)	Heightened neutrophil tissue accumulation; no mortality difference reported.	CB2 signaling restrains early neutrophil recruitment during endotoxemia.
Lehmann 2012 [34]	Lewis rats, LPS endotoxemia & CASP peritonitis models—in vivo	CB2R	HU308 (CB2 agonist) or AM630 (CB2 antagonist)	CB2 activation reduced intestinal leukocyte adhesion and lowered systemic cytokines in both sepsis models	Decreased inflammatory mediator release; microvascular protection	Pharmacologic CB2 stimulation dampens early hyper-inflammation in sepsis, pointing to CB2 as a therapeutic target.
Liu 2014 [37]	Wistar rats, CLP sepsis; lung tissue (in vivo)	CB2R	Oral Melilotus extract vs. vehicle (2 h pre-CLP)	Extract up-regulated CB2R and impeded NF-κB activation in PBMCs/lung	↓ TNF-α, ↓ IL-6, fewer BAL neutrophils, attenuated histologic lung injury	Herbal up-regulation of CB2 breaks NF-κB-driven lung injury in polymicrobial sepsis.
Liu 2020 [38]	C57BL/6 mice + CLP; RAW264.7 macrophages + LPS	CB2R	HU-308 (agonist) ± 3-MA (autophagy blocker)	HU-308 triggers protective autophagy; 3-MA abrogates the effect, linking CB2 → autophagy → anti-inflammation	↓ Lung pathology, ↓ TNF-α/IL-6, ↑ cell viability	CB2-driven autophagy is central to limiting septic lung injury.
Sardinha 2014 [56]	C57BL/6 mice, i.v. LPS endotoxemia; intestinal microcirculation (in vivo)	CB2R	CB2 agonist (HU-308), antagonist (AM630), FAAH inhibitor (URB597), MAGL inhibitor (JZL184)	HU-308, URB597 & JZL184 each lowered leukocyte rolling/adhesion; AM630 maintained adherent leukocyte levels	↓ Leukocyte endothelial interactions; preserved capillary density	Multiple pharmacologic routes that raise CB2 activation reduce gut microvascular inflammation in sepsis.
Souza 2023 [39]	Swiss mice (♂/♀), K. pneumoniae pneumonia-sepsis (in vivo, lung)	CB2R	AM-1241 agonist (0.3/3 mg kg−1 i.p.)	CB2R activation ↓ neutrophil influx, NOS-2, MPO, protein leak; plasma IL-1β ↓, IL-10 ↑	↓ Lung bacterial load, injury score & 7-day mortality	Peripheral CB2 agonism tempers lung neutrophilia and systemic cytokines, boosting survival in pneumonia-sepsis.
Tschöp 2009 [55]	CB2-KO & WT mice; CLP; plus WT + CB2 agonist (GP1a); in vivo	CB2R	KO (loss-of-function) & agonist (GP1a) (gain-of-function)	CB2-KO → ↑ IL-6, bacteremia, lung injury, neutrophil influx & ↓ activation; GP1a reverses in WT, boosts p38 in neutrophils	KO: ↓ survival (22%); GP1a: longer survival, ↓ IL-6, ↓ bacteremia	Here CB2 supports antibacterial immunity; precise role is host-context dependent
Yang 2022 [43]	Male C57BL/6 mice; CLP-induced sepsis-associated encephalopathy; in vivo	CB2R	HU-308 (agonist) vs. vehicle	CB2 activation curbs microglial over-activation and neuronal pyroptosis (↓ NLRP3, GSDMD-NT)	Improved cognition (OFT, NORT, MWM) & histology	CB2 agonism safeguards the brain from CLP-induced neuro-inflammation and pyroptotic damage.
Zhang 2021 [16]	C57BL/6 mice + CLP (in vivo); BMDM + LPS/ATP (in vitro)	CB2R	HU-308 (agonist) vs. AM-630 (antagonist) and CB2-knock-down	CB2 activation dampens NLRP3-caspase-1-GSDMD pyroptosis cascade	↓Pro-inflammatory cytokines, ↓ lung injury, ↑survival	CB2 agonism curbs pyroptosis-driven inflammation and protects septic mice.
Zhang 2023 [65]	C57BL/6 mice, CLP sepsis; heart tissue	CB2R	HU-308 agonist vs. CLP vehicle	CB2R activation ↓ NLRP3, caspase-1, GSDMD and myocardial pyroptosis; ↓ IL-1β, LDH, CK-MB	Improved myocardial histology & injury markers	HU-308-driven CB2 signaling weakens cardiac pyroptosis and biochemical injury after CLP.
Zhao 2023 [40]	C57BL/6 mice, LPS-induced SA-ALI; DCs in vitro/in vivo	CB2R	HU-308 agonist, SR144528 antagonist, DC-specific CNR2 KO	CB2R signaling limits dendritic-cell maturation & pro-cytokines, mitigating lung pathology	↓ Histologic lung injury & cytokines; antagonist/KO reversed protection	CB2 signaling in DCs dampens cytokine storm and rescues LPS-driven acute lung injury.
Zhou 2020 [51]	Mouse peritoneal macrophages + LPS (in vitro)	CB2R	GW-405833 (agonist); 3-MA vs. MG-132	CB2 activation enhances autophagy-lysosome flux → Cathepsin-B-dependent degradation of HMGB1	↓ Extracellular HMGB1 and downstream cytokines	CB2 signaling clears danger-signal HMGB1 via autophagy, tempering inflammation.
Braile 2021 [67]	Human neutrophils (PMNs) + LPS (in vitro)	CB1R/CB2R	ACEA (CB1 agonist)/JWH-133 (CB2 agonist) ± AM-251(CB1 antagonists)/AM630(CB2 antagonist)	Low-dose CB agonists selectively inhibit LPS-induced VEGF-A transcription & release without affecting CXCL8/HGF	↓ Endothelial permeability & tube formation (VEGF-A-driven)	Cannabinoid signaling tempers neutrophil-driven angiogenic leakage relevant to septic vasculopathy.
Godlewski et al., 2004 [49]	Pithed, vagotomised Wistar rats; LPS (0.4–4 mg kg−1); in vivo	CB1R/CB2R	CB1 antagonist SR141716A; CB2 antagonist; VR1 & H3 antagonists	LPS suppressed neurogenic vasopressor response via presynaptic CB1—not CB2/VR1/H3—receptors	CB1 blockade restored sympathetic vasopressor tone in early septic shock	Endocannabinoids acting on presynaptic CB1 dampen sympathetic vasoconstriction during sepsis.
Li 2010 [77]	Rat jejunal myoelectrical activity & mouse charcoal transit; LPS septic ileus; in vivo	CB1R/CB2R	Agonists HU210 (CB1) & JWH133 (CB2); antagonists AM251 (CB1) & AM630 (CB2)	LPS lowers spike amplitude/frequency & GI transit; CB1/CB2 agonists mimic this, while antagonists prevent motility loss and cytokine rise	Antagonists restore GI transit and jejunal activity	Blocking either receptor averts LPS-induced gut stasis—CB1/CB2 antagonists may treat septic ileus
Matias 2023 [66]	Wistar rats, CLP (1- or 3-puncture) sepsis; CNS focus	CB1R/CB2R	CB1 antagonist (AM-251), CB2 antagonist (AM-630) 4 h post-CLP	Early CB1 blockade ↑ survival & prevented long-term fear-memory generalization; CB2 blockade or minocycline prevented fear generalization without ↑ survival, linking early ECS & neuro-inflam. to PTSD-like phenotype	Survival, hyperalgesia, contextual-fear tests; hippocampal TNF-α levels ↓ with minocycline	Early central CB1 (±CB2) antagonism raises survival and blocks long-term fear generalization via dampened neuro-inflammation.
Smith 2000 [76]	BDF1 mice (±C. parvum priming) → LPS endotoxemia (E. coli O55:B5); in vivo	CB1R/CB2R	i.p. agonists (0.05–0.4 mg HU-210; 3.1–50 mg WIN 55 212-2) ± SR141716A/SR144528	CB1 activation ↓ TNF-α & IL-12, ↑ IL-10; effects abolished by CB1—but not CB2—antagonism	Protected primed mice from lethal LPS challenge (↑ survival)	Synthetic CB1 agonists blunt the cytokine storm and rescue survival in endotoxemic, C. parvum–primed mice.
Smith 2001 [68]	BDF1 mice (±C. parvum priming) → LPS endotoxemia (E. coli O55:B5); in vivo	CB1R/CB2R	Low-dose i.c.v. agonists (≈¼ i.p. dose) ± central SR141716A	Central CB1 activation suppresses TNF-α & IL-12 and elevates IL-10 at much lower doses; blockade by SR141716A, not SR144528	Demonstrated potent cytokine modulation; survival not assessed	Brain CB1 receptors are a sensitive switch that controls systemic cytokine output during endotoxemia.
Steiner 2011 [58]	Conscious rats, systemic LPS (25–100 µg kg−1)—in vivo	CB1R/CB2R	Rimonabant (CB1 antagonist), SLV319, CB2/TRPV1 antagonists, ICV AEA	Central CB1 blockade abolished LPS-induced hypothermia; CB2/TRPV1 antagonism had no effect; ICV AEA enhanced hypothermia	Body-temperature drop during severe sepsis is CB1-dependent	Endocannabinoids acting on brain CB1 receptors drive the hypothermic phase of systemic inflammation, distinguishing CB1 from CB2/TRPV1 roles.
Bányai 2023 [71]	Wistar rats, LPS (5 mg kg−1 i.v.) endotoxemia; aorta & heart ex vivo	Mixed CB1/CB2 (THC)	Δ9-THC 10 mg kg−1 i.p.	THC ↓ 4-HNE, 3-nitrotyrosine & COX-2; preserved endothelial-dependent relaxation & ventricular volumes	Maintained cardiac output; mitigated vascular dysfunction	THC safeguards cardiovascular function in endotoxemic rats by cutting oxidative-nitrative stress and COX-2.
Angelina 2022 [52]	BALB/c female mice + LPS sepsis; human monocyte-DC/T-cell co-culture	WIN55,212-2 (mixed CB1/CB2 agonist)	WIN55-212-2 (mixed CB1/CB2 agonist) ± autophagy, CB1 or PPAR-α blockers	CB1/PPAR-α-driven autophagy reprograms DC metabolism → tolerogenic DCs → FOXP3+ Tregs; blocking autophagy or receptors reverses benefit	↑ 78 h survival from 0% → 90% in lethal LPS; ↑ splenic Tregs; ↓ IL-6	Cannabinoids re-educate immunity through autophagy-dependent DC-Treg axis, rescuing mice from endotoxic shock.
He 2022 [35]	MH-S alveolar-macrophage line (LPS) & C57BL/6 LPS-sepsis mice; in vitro + in vivo lung	WIN55,212-2 (mixed CB1/CB2 agonist)	WIN (0.625–40 µM in vitro; dosing in mice) vs. LPS; miR-29b-3p/FOXO3 gain–loss	WIN ↑ miR-29b-3p → ↓ FOXO3 → ↓ PFKFB3 glycolysis → shifts M1 → M2 phenotype	↓ BALF protein, cells, lactate; better lung histology	WIN blocks macrophage glycolysis via a miR-29b-3p axis, easing LPS-induced acute lung injury.
Toguri 2015 [80]	Lewis rats, LPS endotoxemia; (in vivo)	WIN55,212-2 (mixed CB1/CB2 agonist)	WIN55,212-2 ± CB1 antagonist (AM281) or CB2 antagonist (AM630)	CB2 (but not CB1) activation curtailed leukocyte-endothelium adhesion and normalized micro-vascular flow	↓ Adherent leukocytes; ↑ capillary perfusion	Systemic CB2 stimulation rapidly rescues microcirculatory dysfunction in endotoxemia.
Wilhelmsen 2014 [79]	Primary human lung microvascular ECs (HMVEC-lung) stimulated with LPS/FSL-1/TNF-α (in vitro)	WIN55,212-2 (mixed CB1/CB2 agonist)	NADA or WIN vs. vehicle ± CB1/CB2 antagonists	WIN and NADA dose-dependently cut IL-6/IL-8 release & neutrophil adhesion; NADA effect is CB1/CB2-dependent;	↓ Endothelial cytokines & leukocyte binding; no effect on permeability	Endothelial ECS (via NADA or WIN) is a brake on cytokine-driven vascular activation relevant to sepsis.
Singh 2018 [50]	Swiss-albino mice, CLP (6 h & 20 h), ex vivo aortic rings	2-AG/CB1R	DAGL inhibitor (KT-109), MAGL inhibitor (JZL-184), CB1 antagonist (AM-251)	Sepsis up-regulated CB1R & reduced MAGL mRNA; blocking DAGL or CB1 restored NA-induced vasoconstriction	KT-109 or AM-251 normalized vascular reactivity; improved mean arterial tone (survival not assessed).	Endocannabinoid 2-AG acting on CB1 underlies septic vasoplegia.
Sultan 2021 [41]	Female C57BL/6 Mice; SEB-induced ARDS (systemic super-antigen sepsis); in vivo lung	Anandamide (AEA) ligand	Exogenous AEA 40 mg kg−1 vs. vehicle	AEA down-regulates miR-23a-3p/miR-34a-5p → ↑ ARG1, TGF-β2, FOXP3 → expansion of MDSCs & Tregs; ↓ IL-2, IFN-γ, TNF	↑ Lung function, ↓ cell infiltrates, cytokines	AEA re-programs miRNA networks to tilt immunity toward MDSC/Treg-driven resolution of SEB-ARDS.
Murakami 2009 [69]	C57BL/6 mice sensitized with D-galactosamine + LPS (endotoxin shock, in vivo) ± RAW264.7 cells (in vitro)	Anandamide (ligand)	Cationic antimicrobial peptide CAP11 vs. scrambled control	CAP11 prevents LPS-induced rise in AEA, HMGB1, TNF-α & IL-6 in macrophages and in serum	↓ systemic mediators, improved survival in shock model; suppresses ECS ligand generation	CAP11’s protection is partly via curbing endocannabinoid (AEA) over-production
Liu, 2006 [73]	RAW264.7 macrophages with/without LPS; mouse brain extracts; in vitro	AEA biosynthetic enzymes (NAPE-PLD, PLC, PTPN22)	LPS stimulation; siRNA knock-down; overexpression; enzyme inhibitors	LPS up-regulates a PLC–phosphatase pathway (involving PTPN22) that converts NAPE→phospho-AEA→AEA, bypassing NAPE-PLD	↑ AEA synthesis under endotoxin challenge; identifies alternate biosynthetic route	Defined a novel PLC/PTPN22 route for LPS-induced anandamide production in macrophages.
De Filippis, 2008 [74]	Swiss OF1 mice; LPS-induced sepsis; in vivo + intestinal muscle strips	CB1 receptor, FAAH (indirectly through cannabidiol)	Cannabidiol alone and with CB1 antagonist (AM251)	Sepsis upregulated CB1 and FAAH in intestine; cannabidiol reversed FAAH increase but worsened motility via CB1	Cannabidiol reduced motility in septic mice; CB1 antagonist reversed this effect	Cannabidiol worsened sepsis-induced ileus through CB1 signaling; reversed by CB1 antagonist.
Szafran 2015 [70]	Female C57BL/6 mice, single i.p. LPS (6–24 h); spleen & splenocytes (in vivo & ex vivo)	Ces2g (2-AG-hydrolyzing carboxylesterase)	Endotoxin vs. saline	LPS selectively suppressed Ces2g activity → ↓ 2-AG hydrolysis in spleen & splenocytes	Potential ↑ 2-AG levels (not directly measured)—suggests feedback dampening of inflammation	Endotoxin shuts down a 2-AG catabolic enzyme, potentially prolonging anti-inflammatory eCB signaling during systemic inflammation.
Wolfson 2013 [59]	Murine peripheral blood mononuclear cells (PBMCs) from non-pregnant vs. pregnant mice—in vitro/in vivo LPS	FAAH	Progesterone (PR agonist) ± PR antagonist	LPS lowered FAAH activity in PBMCs; progesterone restored FAAH via progesterone-receptor signaling	Normalized FAAH prevents excess anandamide buildup linked to pregnancy failure	Hormonal regulation can counter LPS-driven suppression of FAAH, highlighting cross-talk between endocrine and ECS during systemic inflammation.
Kianian 2013 [54]	Lewis rats, LPS endotoxemia—in vivo	FAAH & CB2	URB597 (FAAH inhibitor) ± AM630 (CB2 antagonist)	FAAH inhibition raised endocannabinoids, cutting leukocyte adhesion and increasing FCD; CB2 blockade reversed adhesion benefit	Enhanced intestinal perfusion; reduced microvascular inflammation	Elevating endogenous cannabinoids via FAAH inhibition protects gut microcirculation during endotoxemia, largely through CB2.
Chen 2022 [60]	C57BL/6 mice + CLP sepsis; serum from septic patients (observational)	GPR55 (non-classical ECS GPCR)	CID16020046 (antagonist) 20 mg kg−1	GPR55 blockade suppresses ROCK1/2 signaling, inflammation and apoptosis in kidney	↓ BUN/Creatinine, ↓ KIM-1/NGAL, ↓ TNF-α/IL-6, no off-target toxicity	Targeting GPR55 mitigates sepsis-induced AKI via anti-inflammatory/anti-apoptotic routes.
Joffre 2022 [61]	C57BL/6 mice, LPS i.v./i.t. endotoxemia; S. aureus i.t. pneumonia/sepsis; in vivo; plus BMDM, HMVEC, astrocytes in vitro	TRPV1; OLDA agonist	OLDA 5–10 mg/kg i.v. given at 0–2 h post-challenge vs. vehicle; WT vs. Trpv1−/−; cell-specific TRPV1 knockdowns; monocyte/macrophage depletion	Pan-neuronal (CNS) TRPV1 triggers an early surge in IL-10, suppresses IL-6/TNF/chemokines; reversed with knockdown, not with PNS or myeloid TRPV1 knockdown. Monocyte depletion blocks OLDA-induced IL-10.	↓ Lung injury, ↑ Lung IL-10; improved mouse sepsis scores; bacterial load unchanged in S. aureus model.	CNS Neuronal TRPV1 activation by ODA results in a neuro-immune pathway that drives peripheral monocyte IL-10; attenuates systemic inflammation; reduces clinical severity
Lawton 2017 [95]	Mice, LPS i.v., CLP; ex vivo hematopoietic	TRPV1; NADA agonist	NADA 1–10 mg/kg i.v. (often 5–10 mg/kg) ± LPS/CLP; WT vs. Trpv1−/−; bone-marrow chimeras	Non-hematopoietic cell TRPV1 (pointing to neuronal) reduces systemic pro-inflammatory cytokines and PAI-1; raises IL-10. NADA alters neuropeptides (↓ CGRP, ↑ substance P) during LPS.	↑ Survival in endotoxemic mice; ↓ inflammatory mediator levels across models; ↑ IL-10, ↓ IL-6, CCL2, PAI-1	Nonhematopoietic TRPV1 activation attenuates systemic inflammation and improves survival in endotoxemic mice
Orliac 2007 [96]	Rat, i.p. endotoxemia (6 h); isolated mesenteric vascular bed; in vivo + ex vivo	TRPV1; AEA agonist	LPS vs. saline; ex vivo anandamide (0.01–10 μM) and capsaicin; TRPV1 antagonist capsazepine; PKC activator PMA	LPS increases TRPV1 and CGRP-positive nerve density in mesenteric arteries; ↑ AEA mediated CGRP and vasorelaxation (TRPV1 dependent) post-LPS; PKC activation ↑ AEA efficacy	Upregulation of TRVP1 during endotoxemia implicated in septic hypotension	Early endotoxemia upregulates TRPV1/CGRP signaling in perivascular sensory nerves, amplifying AEA-driven CGRP release and vasodilation; indicates an ECS–TRPV1 contribution to vascular changes in sepsis.

).

Following CB2 modulation, the next most frequently investigated ECS component was CB1 receptor inhibition. Thirteen of the 23 studies assessing CB1 inhibition reported attenuation of sepsis severity. Eight of the 13 studies demonstrated that application of CB1 antagonist (e.g., Rimonabant, AM281) to rodent models reduced sepsis associated hypotension by stabilizing vasoplegia or increasing vascular response to vasopressors (e.g., vasopressin). Three studies employing central blockage of CB1 signaling reported attenuation of late septic hypotension and increasing survival; in 1 study, 7-day survival rate significantly improved by 39% in a CLP rat model [47]. This phenomenon of increased survival benefit was elicited in 4 other CB1 inhibitor studies. Improved neurological outcomes were seen in 2 studies by attenuation of pro-inflammatory mechanisms, such as caspase-3 activation [46]. Collectively, 4 studies showcased a reduction in pro-inflammatory cytokines, of which all were documented with a decrease in TNF-α. In contrast, CB1 activation produced mixed results. Among the 10 studies investigating CB1 activation, 6 reported improved sepsis outcomes through a reduction in pro-inflammatory cytokine levels, most notably IL-6 and TNF-α, with 3 studies concurrently displaying an upregulation of IL-10. Conversely, other studies showed that CB1 activation translated to worsened septic outcomes with hypotension, hypothermia, nitric oxide mediated organ damage, and reduced intestinal motility [57,58] (Appendix A Table A2).

Beyond the canonical ECS, several ECS-adjacent targets we included in the 53 selected studies. TRPV1 activation triggered IL-10–centered neuroimmune reflexes that alleviated lung and brain injury. This was achieved through central activation but required peripheral macrophages independent of TRPV1 to convey their protective effect. Inhibition of GPR55 reduced renal apoptosis and inflammation in acute kidney injury models. Additionally, pharmacologic modulation of ECS-metabolizing enzymes like FAAH and MAGL enhanced endogenous cannabinoid tone, supporting vascular and immunologic recovery [54,56,59,60,61].

Human observational studies demonstrated correlations between ECS lipid levels (AEA, 2-AG) and clinical outcomes such as invasive mechanical ventilation and ICU stay length [62]. Transcriptomic profiling further identified ECS-linked gene signatures and hub targets (e.g., PTEN, HIST2H2BE) with diagnostic potential [63].

Collectively, these results indicate that ECS modulation—particularly via CB2 agonism and TRPV1 signaling—confers organ-specific and systemic benefits in sepsis. However, context-dependent findings, particularly regarding CB1 and IL-10 dynamics, reveal a complex landscape requiring precise therapeutic targeting (Figure 3).

This figure displays the general mechanistic trends of the three most studied endocannabinoid receptors: CB1, CB2, and TRPV1. Activation of CB1 shows the most nuanced and context dependent effects, often resulting in hypotension, loss of vasopressin tone, and anti-inflammatory release. CB2 and TRPV1 were more consistent, leading to an increase in systemic anti-inflammatory cytokines like IL-10. CB2 activation additionally demonstrated attenuation of leukocyte adhesion, pyroptosis, and tissue injury.

4. Discussion

This scoping review elucidates the intricate and contextually dependent function of the endocannabinoid system (ECS) within the pathophysiology of sepsis, synthesizing molecular, cellular, and systemic perspectives derived from both preclinical and clinical investigations. The role of the CB₂ receptor has repeatedly been interpreted as a safeguard marker within the ECS components in numerous testing environments. Selective agonists of the CB₂ receptor—including JWH-133, HU-308, and AM-1241—exhibited significant mitigation of inflammation, immune dysregulation, and pyroptosis, thereby establishing CB₂ signaling as a pivotal regulator of immune homeostasis [16,40,52,64,65,66,67,68,69].

In models of endotoxemia and cecal ligation and puncture (CLP), the activation of CB₂ receptors has been observed to mitigate neutrophil infiltration, oxidative stress, and the expression of pro-inflammatory cytokines (such as TNF-α, IL-1β, IL-6), whilst concurrently augmenting IL-10 production and promoting autophagic clearance [51,53,55,69,70]. Additionally, CB₂ signaling has been implicated in the promotion of M2 macrophage polarization, the inhibition of dendritic cell maturation [16,53,55], and the facilitation of microglial reprogramming from the pro-inflammatory M1 phenotype to the reparative M2 phenotype—thereby ameliorating cognitive impairments associated with septic encephalopathy [43,54,59,71]. Nevertheless, the ramifications of CB₂ activation were not entirely advantageous. Within CD4⁺ T cells, the activation of CB₂ paradoxically resulted in the inhibition of IL-10 production and exacerbated lung injury and mortality, thereby highlighting the necessity for therapeutic strategies that are tailored to specific cell types and tissue contexts [16]. These findings underscore the risk that indiscriminate CB2 agonism could potentially worsen immunosuppression in patients already progressing toward sepsis-associated immune paralysis. Translational strategies will therefore require precise approaches guided by biomarkers to identify patient subsets that would most likely benefit. Circulating cytokine ratios such as IL-6/IL-10, or immune cell phenotyping may help stratify patients in either hyperinflammatory or immunosuppressed states, guiding clinical evaluation and guarding against indiscriminate use [72].

In a different light, the interaction with CB₁ receptors illustrated an increased complexity and range of variability. Central antagonism of CB₁ was correlated with enhanced survival rates and hemodynamic stability in multiple rodent models [46,56,73,74], with one investigation documenting a 39% increase in survival among CLP rats. Also, blocking CB₁ led to reduced caspase-3 activation and neuroinflammation, which points towards possible neuroprotective roles within the scenario of septic encephalopathy [75]. Conversely, the triggering of CB₁ generated effects that changed substantially in relation to the anatomical and cellular frameworks. Specifically, central CB₁ activation intensified the severity of sepsis, whereas peripheral CB₁ signaling—especially in monocytic myeloid-derived suppressor cells—facilitated the release of IL-10 and provided immunosuppressive advantages [50,76].

Conversely, the overactivation of CB₁ receptors has been associated with gastrointestinal dysmotility [77], diminished vascular responsiveness [49,75,78], and an exacerbation of inflammatory processes [35,58]. The application of enzymatic modulation through the use of FAAH and MAGL inhibitors has proven effective in restoring the equilibrium of the endocannabinoid system (ECS), thereby alleviating these detrimental consequences [51,69,70]. Collectively, these findings indicate that CB₁ signaling operates with dual, context-dependent roles in the pathophysiology of sepsis, with central blockade and selective peripheral activation presenting unique therapeutic opportunities. Despite this, the unregulated activation of CB₁ may cause hypotensive incidents, collapse of the circulatory system, and organ malfunction—underscoring the need for accurate approaches that explicitly separate the central and peripheral CB₁ signaling routes.

It is noteworthy that divergent thermoregulatory outcomes were documented in investigations utilizing Rimonabant for the antagonism of CB₁ receptors. Leite-Avalca et al. (2016) observed hypothermia subsequent to CB₁ blockade, whereas Steiner et al. (2011) ascribed hypothermia to the activation of CB₁ receptors [47,58]. These inconsistencies are likely attributable to variations in the route of administration (intracerebroventricular versus oral) and the timing of intervention (pretreatment versus post-sepsis induction), thereby underscoring the necessity for mechanistic studies to clarify the temporal and spatial dynamics associated with the modulation of the ECS.

Beyond cannabinoid receptors, the activation of TRPV₁ by both endogenous and synthetic ligands (for instance, OLDA and NADA) initiated IL-10-mediated neuroimmune reflexes, which contributed to the attenuation of inflammation within the pulmonary and central nervous systems [61,67,79]. The inhibition of GPR55 resulted in a reduction in renal injury through the downregulation of the RhoA/ROCK signaling pathway and the induction of apoptotic processes [60], whereas the modulation of GPR18 and PPARγ introduced novel pathways of endocannabinoid system-driven resolution signaling [38,41,42].

The enzymatic modulation of the endocannabinoid system (ECS) tone through the inhibition of fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MAGL) markedly enhanced the protective mechanisms of autophagy, augmented vascular perfusion, and preserved intestinal integrity [37,51,70,80]. These observed phenomena were particularly pronounced within the endothelial and gastrointestinal systems, wherein lipids derived from the ECS influenced permeability and alleviated oxidative stress [34,49,57]. Moreover, ECS-mediated augmentation of autophagy significantly mitigated systemic cytokine storms through the reduction in high mobility group box 1 (HMGB1) and damage-associated molecular patterns (DAMPs), thus promoting the resolution process [69,70].

Clinically, the levels of circulating endocannabinoid system (ECS) lipids, such as anandamide (AEA) and 2-arachidonoylglycerol (2-AG), exhibited an inverse correlation with the duration of mechanical ventilation, the length of stay in the intensive care unit (ICU), and the severity of sepsis [17,36,45,60]. Transcriptomic and metabolomic investigations have elucidated ECS-associated enzymes and pivotal genes, including phosphatase and tensin homolog (PTEN) and HIST2H2BE, which may possess significant diagnostic and therapeutic implications [35,36,81].

Notwithstanding these encouraging observations, several vulnerabilities associated with the ECS necessitate prudent consideration. Reports have indicated the occurrence of CB₁ receptor-induced vasoplegia, maladaptive responses of transient receptor potential vanilloid 1 (TRPV₁), and timing-dependent immunosuppressive effects stemming from monoacylglycerol lipase (MAGL) inhibition [47,82,83]. This highlights the necessity for a more profound mechanistic comprehension of the interactions involving the endocannabinoid system across the immune, vascular, and neural systems.

In synthesis, the modulation of the endocannabinoid system—particularly through the mechanisms of CB₂ receptor agonism, TRPV₁ receptor activation, and the inhibition of fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MAGL)—provides a complex therapeutic framework in the management of sepsis. Nonetheless, the variable outcomes associated with CB₁ receptor activation and the interactions between various receptors necessitate the implementation of precision medicine strategies that consider receptor subtype specificity, cellular lineage, temporal dynamics, and organ-specific environments. Lastly, examining interventions associated with the endocannabinoid system should be done concurrently with existing immunomodulatory frameworks. Corticosteroids, though widely adopted, are tied to the possibility of indiscriminate immunosuppression along with extended recovery intervals. Innovative treatment strategies like cytokine adsorption and immune checkpoint inhibitors are still in early development phases and show variable effectiveness [84].

In contrast to these considerations, endocannabinoid system (ECS) modulation—specifically through the agonism of CB₂ receptors—may distinctly mitigate hyperinflammation whilst fostering reparative immune phenotypes [85]. Nevertheless, challenges associated with translation into clinical practice persist. The cardiovascular safety profile and the immunosuppressive implications, particularly in relation to CB₂ receptor activation, are yet to be thoroughly delineated. The specificity inherent in ECS modulation, although therapeutically beneficial, introduces complexities regarding predictability. As an illustration, equivalent pharmacological substances can have divergent impacts due to differing model types, administration routes, and intervention timings. Furthermore, existing sepsis models (such as lipopolysaccharide (LPS), cecal ligation and puncture (CLP), and caspase-1 activation (CASP)) inadequately represent the heterogeneity observed in human sepsis, which frequently results from concurrent infections in older populations with multiple comorbidities. Additionally, the reliance on the LPS sepsis model, which was present in 70% of selected studies, introduces a Gram-negative bacterial bias to mechanistic insights, overshadowing non-Gram-negative infections that make up approximately 42% of sepsis cases [86].

The unification of preliminary research models along with the segmentation of patient groups is key to increasing the broader applicability and clinical importance of our findings. Additionally, neurobehavioral side effects warrant careful consideration; however, the modulation of the endocannabinoid system (ECS) through the inhibition of fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MAGL) may potentially reduce the risks associated with cannabinoid-related adverse events [87,88]. The limited availability of human ECS data further constrains the comprehension of interindividual variability in ECS functionality throughout various stages of sepsis and the presence of comorbid conditions. Future translational investigations should emphasize the importance of stratified patient populations in order to exploit ECS signaling mechanisms for the maintenance of immunologic homeostasis and recovery in the context of human sepsis.

From a translational perspective, several limited clinical studies investigating ECS-targeting compounds that have moved beyond preclinical models. Proof-of-pharmacology clinical trials of FAAH inhibitor PF-04457845 sought to establish bioavailability and dosing parameters, while CB1 antagonist ANEB-001 has demonstrated penetration of CNS blood–brain barrier in early human studies [89,90]. Off-label cannabinoid drugs such as Sativex and Nabilone, investigated in related neuroinflammatory conditions such as Huntington’s or Parkinson’s disease, could shine light on potential applications regarding neuroinflammation found in sepsis [91,92]. Although selective CB2 agonists (e.g., HU-308, JWH-133) remain in preclinical development, their consistent anti-inflammatory effects highlight promising pharmacological direction. While to date there are no clinical studies directly investigating ECS compounds in the context of human sepsis, the growing clinical experience with ECS modulation in other indications suggests the feasibility for future translational studies.

This scoping review is subject to several constraints. First, the exclusion of articles published in languages other than English may have resulted in the inadvertent omission of pertinent studies, particularly from regions such as China, where research pertaining to ECS-related sepsis is notably vigorous. This judgment reflects practical concerns, consequently narrowing the scope and impact of our perspectives. Subsequent reviews may derive advantages from implementing multilingual search methodologies or fostering international cooperative efforts. Aligned with the standards described in scoping review methodology, we resolved not to investigate the methodological robustness of the studies highlighted in our evaluation. As a result, it is recommended that individuals carefully assess the conclusions, recognizing that the strength of the data might vary among the different research works.

5. Conclusions

In conclusion, the endocannabinoid system constitutes a sophisticated yet promising focal point for sepsis intervention, with empirical evidence corroborating the beneficial effects of CB₂ agonism, TRPV₁ activation, and FAAH/MAGL inhibition on immune modulation, organ safeguarding, and the resolution of inflammatory responses. Although no ECS-directed therapies have been tested clinically, the preclinical outcomes provide hypothetical extrapolations to frame clinical translation. CB2 agonism associated with improved survival in rodent sepsis, while TRPV1 activation attenuated lung and brain injury, suggest potential to shorten duration of mechanical ventilation and reduce encephalopathy. Similarly, FAAH and MAGL modulation showed improvement to vascular and intestinal microcirculation, showcasing potential relevance for attenuation of multi-organ dysfunction and mortality. Nevertheless, the ambivalent roles of CB₁ signaling and context-dependent receptor interactions underscore the imperative for precise, cell- and organ-specific therapeutic modalities. While preclinical and preliminary clinical investigations emphasize ECS modulation as a multifaceted approach for reinstating immune equilibrium and enhancing clinical outcomes, additional translational research is critically required to refine treatment timing, selectivity, and safety in the context of human sepsis. Customized ECS-targeted interventions may improve the management of this complex syndrome, provided that these strategies are guided by a thorough comprehension of molecular, cellular, and systemic principles.