18.10.02 · organismal-bio / immunology

Inflammation: innate recognition via TLRs and NLRs, cytokine storm, and resolution mechanisms

stub3 tiersLean: nonepending prereqs

Anchor (Master): Janeway's Immunobiology, 10th ed. (2022), Ch. 3-4

Intuition Beginner

Inflammation is the body's first response to injury or infection. When tissue is damaged or pathogens invade, the affected area becomes red (increased blood flow), hot (heat from increased blood flow), swollen (fluid leaking from blood vessels into tissue), and painful (chemical mediators stimulating pain receptors). These are the classical signs of inflammation, first described by Celsus in the 1st century AD: rubor, tumor, calor, and dolor. A fifth sign, functio laesa (loss of function), was added later.

White blood cells -- especially neutrophils and macrophages -- rush to the site, guided by chemical signals called cytokines. These cells engulf and destroy bacteria, dead cells, and debris. The process is remarkably effective: most minor infections and injuries are resolved within days.

Usually inflammation resolves once the threat is gone. Specialized chemical messengers actively turn the inflammatory response off and promote tissue repair. But sometimes inflammation spirals out of control -- a cytokine storm -- where the immune system releases so many inflammatory signals that healthy tissue is damaged, sometimes causing more harm than the original threat. Cytokine storms occur in severe infections like sepsis, in some viral infections (influenza, COVID-19), and as a complication of certain immunotherapies. Understanding how inflammation starts, how it is normally resolved, and what happens when resolution fails is one of the most important problems in medicine.

Visual Beginner

Stage	What happens	Key mediators
Recognition	Immune cells detect damage or pathogens	Pattern recognition receptors (TLRs, NLRs)
Vascular response	Blood vessels dilate and become leaky	Histamine, prostaglandins, bradykinin
Cell recruitment	White blood cells migrate to the site	Cytokines (IL-1, TNF-alpha, IL-6), chemokines
Destruction	Phagocytes engulf and kill pathogens	Reactive oxygen species, enzymes
Resolution	Inflammation is actively turned off	Lipoxins, resolvins, IL-10, TGF-beta

Tissue injury or infection
        |
   Pattern recognition (TLRs, NLRs detect PAMPs/DAMPs)
        |
   Cytokine release (IL-1, TNF-alpha, IL-6)
        |
   Vasodilation and increased permeability
        |
   Leukocyte recruitment (rolling, activation, adhesion, diapedesis)
        |
   Phagocytosis and pathogen destruction
        |
   Resolution (pro-resolving mediators, efferocytosis)
        |
   Tissue repair and return to homeostasis

   (If resolution fails --> chronic inflammation or cytokine storm)

Worked example Beginner

Consider what happens when you get a splinter in your finger contaminated with bacteria.

Step 1. Tissue damage and recognition (seconds to minutes). The splinter damages skin cells and introduces bacteria. Tissue-resident macrophages and mast cells immediately detect the damage. Macrophages recognize bacterial components (like lipopolysaccharide from Gram-negative bacteria or peptidoglycan from Gram-positive bacteria) through pattern recognition receptors on their surface. Mast cells in the tissue degranulate, releasing histamine stored in their granules.

Step 2. Vascular response (minutes). Histamine and other mediators (prostaglandins, bradykinin) cause local blood vessels to dilate (vasodilation), increasing blood flow to the area -- this causes redness and heat. Simultaneously, the vessel walls become more permeable, allowing fluid, proteins, and white blood cells to leak into the tissue -- this causes swelling. Prostaglandins and bradykinin also stimulate pain receptors, causing the tenderness you feel.

Step 3. Leukocyte recruitment (minutes to hours). The increased blood flow brings large numbers of neutrophils to the area. Cytokines from macrophages (especially IL-1 and TNF-alpha) cause endothelial cells lining the blood vessels to express adhesion molecules (selectins) on their surface. Neutrophils in the bloodstream first roll along the vessel wall (selectin-mediated weak adhesion), then activate (chemokine signaling), then adhere firmly (integrin-mediated strong adhesion), and finally squeeze between endothelial cells (diapedesis) to enter the tissue.

Step 4. Phagocytosis and destruction (hours). Once in the tissue, neutrophils and macrophages engulf bacteria into phagosomes, which fuse with lysosomes containing destructive enzymes. The respiratory burst -- a sudden increase in oxygen consumption -- generates reactive oxygen species (superoxide, hydrogen peroxide) that kill the engulfed bacteria. Neutrophils can also release neutrophil extracellular traps (NETs), web-like structures of DNA and antimicrobial proteins that ensnare bacteria.

Step 5. Resolution (hours to days). Once the bacteria are cleared, the inflammatory response must be actively turned off. Neutrophils undergo apoptosis (programmed cell death) and are cleared by macrophages (efferocytosis). Macrophages shift from a pro-inflammatory to a pro-repair phenotype, releasing anti-inflammatory cytokines (IL-10, TGF-beta) and growth factors that promote tissue healing. Specialized pro-resolving mediators (lipoxins, resolvins, protectins) actively suppress further inflammation and promote tissue repair. The area returns to normal.

Check your understanding Beginner

Exercise (easy, multiple choice).

What is a cytokine storm?

A. A type of autoimmune disease B. An excessive, uncontrolled release of pro-inflammatory cytokines that damages healthy tissue C. The normal process by which inflammation is resolved D. A type of allergic reaction

Hint

Cytokine storms occur when the inflammatory response spirals out of control. Think about what happens in severe infections like sepsis.

Answer

Option B. A cytokine storm is a life-threatening systemic inflammatory response in which massive, uncontrolled release of pro-inflammatory cytokines (especially TNF-alpha, IL-1, and IL-6) causes widespread tissue damage, organ failure, and potentially death. Cytokine storms occur in sepsis, severe viral infections (H5N1 influenza, COVID-19), and as a complication of CAR-T cell immunotherapy. The storm is not the normal inflammatory response but a pathological amplification of it.

Exercise (easy, true/false).

Inflammation resolves passively -- the inflammatory mediators simply run out and the response fades away on its own.

Hint

Consider whether there are active processes that turn inflammation off, including specialized molecules produced during the response.

Answer

False. Inflammation resolution is an active, highly regulated process, not merely the passive decay of pro-inflammatory signals. Specialized pro-resolving mediators (SPMs) -- including lipoxins, resolvins, protectins, and maresins -- are actively synthesized from essential fatty acids during the inflammatory response. These mediators actively suppress neutrophil infiltration, promote macrophage efferocytosis (clearing dead neutrophils), stimulate tissue repair, and counteract pro-inflammatory signaling. The discovery that resolution is active, championed by Charles Serhan and colleagues, has transformed the understanding of inflammation from a two-phase model (on/off) to a three-phase model (on/resolution/repair).

Formal definition Intermediate+

Inflammation is the coordinated physiological response to tissue injury and infection, involving innate immune recognition, vascular changes, leukocyte recruitment, pathogen clearance, and active resolution. It is mediated by a complex interplay of cellular receptors, soluble mediators, and cell-cell interactions.

Innate pattern recognition

The inflammatory response is initiated when innate immune receptors detect either pathogen-associated molecular patterns (PAMPs) -- conserved molecular structures shared by classes of pathogens -- or damage-associated molecular patterns (DAMPs) -- molecules released by damaged or stressed host cells.

Toll-like receptors (TLRs) are transmembrane pattern recognition receptors that detect extracellular and endosomal PAMPs. Ten functional TLRs are expressed in humans. Key examples: TLR4 recognizes lipopolysaccharide (LPS) from Gram-negative bacteria; TLR2 recognizes peptidoglycan and lipoproteins from Gram-positive bacteria; TLR3 recognizes double-stranded RNA (viral); TLR7 and TLR8 recognize single-stranded RNA (viral); TLR9 recognizes unmethylated CpG DNA (bacterial and viral). TLR signaling activates the transcription factor NF-kappaB and IRF (interferon regulatory factor) pathways, leading to production of pro-inflammatory cytokines (TNF-alpha, IL-1beta, IL-6), chemokines, and type I interferons.

NOD-like receptors (NLRs) are cytoplasmic receptors that detect intracellular PAMPs and DAMPs. NOD1 detects gamma-D-glutamyl-meso-diaminopimelic acid (a peptidoglycan fragment from Gram-negative bacteria); NOD2 detects muramyl dipeptide (a peptidoglycan fragment found in both Gram-positive and Gram-negative bacteria). Several NLRs (NLRP3, NLRC4, AIM2) form inflammasomes -- multiprotein complexes that activate caspase-1, which cleaves pro-IL-1beta and pro-IL-18 into their active, secreted forms and triggers pyroptosis (inflammatory cell death).

RIG-I-like receptors (RLRs) -- RIG-I and MDA5 -- detect cytoplasmic viral RNA and activate the MAVS signaling pathway, leading to type I interferon production.

C-type lectin receptors (CLRs) -- including Dectin-1 (recognizes beta-glucan from fungi) and mannose receptor -- detect carbohydrate structures on pathogens.

Inflammatory mediators

Upon activation, innate immune cells release a cascade of inflammatory mediators:

Vasoactive amines: Histamine (from mast cells and basophils) causes vasodilation and increased vascular permeability.

Lipid mediators: Prostaglandins (especially PGE2, produced by cyclooxygenase enzymes COX-1 and COX-2) cause vasodilation, potentiate edema, and sensitize pain receptors. Leukotrienes (LTB4, LTC4, LTD4, LTE4, produced by 5-lipoxygenase) cause bronchoconstriction, increased vascular permeability, and neutrophil chemotaxis. Platelet-activating factor (PAF) promotes platelet activation and inflammatory cell recruitment.

Peptide mediators: Bradykinin (generated by the kinin-kallikrein system) causes vasodilation, increased vascular permeability, and pain. Complement fragments C3a and C5a (anaphylatoxins) cause mast cell degranulation, vasodilation, and neutrophil chemotaxis.

Pro-inflammatory cytokines: TNF-alpha (from macrophages, T cells, NK cells) promotes inflammation, fever, and acute-phase protein synthesis; at high concentrations it causes septic shock (hypotension, disseminated intravascular coagulation). IL-1beta (from macrophages, processed by the inflammasome) promotes fever, T cell activation, and endothelial adhesion molecule expression. IL-6 (from macrophages, endothelial cells, fibroblasts) induces acute-phase protein synthesis by the liver and B cell differentiation.

Leukocyte recruitment

Leukocyte migration from the blood into inflamed tissue follows a four-step cascade:

Rolling: Selectins on activated endothelial cells (E-selectin, P-selectin) bind to glycoprotein ligands on leukocytes (e.g., PSGL-1 on neutrophils), producing a weak, transient adhesion that slows the leukocyte and causes it to roll along the vessel wall.
Activation: Chemokines displayed on the endothelial surface (e.g., IL-8/CXCL8) bind to G-protein-coupled receptors on the rolling leukocyte, triggering intracellular signaling that activates the leukocyte's integrins.
Firm adhesion: Activated integrins on the leukocyte (e.g., LFA-1, Mac-1) undergo conformational change and bind tightly to immunoglobulin superfamily molecules on the endothelium (ICAM-1, VCAM-1), arresting the leukocyte on the vessel wall.
Transmigration (diapedesis): The adherent leukocyte squeezes between endothelial cells (paracellular route) or through endothelial cells (transcellular route) via interactions with PECAM-1 (CD31) and other junctional molecules, entering the tissue. The leukocyte then follows chemokine gradients to the site of injury or infection.

Phagocytosis and destruction

Neutrophils and macrophages engulf pathogens into phagosomes, which fuse with lysosomes to form phagolysosomes. Killing mechanisms include: the respiratory burst (NADPH oxidase generates superoxide O2-, which is converted to hydrogen peroxide and hypochlorous acid by myeloperoxidase); enzymatic degradation by lysosomal hydrolases; and antimicrobial peptides (defensins, cathelicidins). Neutrophils can also release neutrophil extracellular traps (NETs) -- extruded chromatin decorated with histones and granule proteins -- that ensnare and kill bacteria. NET formation can be suicidal (NETosis, cell death) or vital (DNA release without cell death).

Acute-phase response

Systemic inflammation triggers the acute-phase response: IL-6 (and to a lesser extent IL-1beta and TNF-alpha) acts on the liver to increase synthesis of acute-phase proteins, including C-reactive protein (CRP, an opsonin and clinical marker of inflammation), serum amyloid A, fibrinogen, and complement components. IL-1beta and TNF-alpha act on the hypothalamus to induce fever: they stimulate COX-2 expression in hypothalamic endothelial cells, producing PGE2, which raises the thermoregulatory set point. Fever enhances immune function by increasing phagocytosis, T cell proliferation, and interferon production.

Cytokine storm

A cytokine storm (cytokine release syndrome, CRS) is a pathological, self-amplifying positive feedback loop of cytokine production. The key drivers are TNF-alpha, IL-1beta, and IL-6, which at high systemic concentrations cause: widespread endothelial activation and vascular leak (leading to hypotension and edema); disseminated intravascular coagulation (DIC); metabolic disturbances (lactic acidosis); and multi-organ failure. Cytokine storms occur in sepsis (systemic bacterial infection), severe viral infections (H5N1 and H1N9 influenza, SARS-CoV-2), and as a complication of CAR-T cell therapy (where massive T cell activation upon engaging tumor cells triggers systemic cytokine release).

Resolution of inflammation

Resolution is an active, receptor-mediated process governed by specialized pro-resolving mediators (SPMs):

Lipoxins (LXA4, LXB4) are arachidonic acid-derived mediators produced by the sequential actions of 5-lipoxygenase (in leukocytes) and 12/15-lipoxygenase (in platelets and epithelial cells). Lipoxins inhibit neutrophil chemotaxis and adhesion, stimulate non-phlogistic (non-inflammatory) macrophage phagocytosis of apoptotic neutrophils, and promote tissue repair.

Resolvins (from EPA and DHA, omega-3 fatty acids) are produced by the E-series (RvE1, from EPA via COX-2 and 5-LOX pathways) and D-series (RvD1-RvD6, from DHA via 15-LOX and 5-LOX). Resolvins potently block neutrophil infiltration, reduce pro-inflammatory cytokine production, and enhance macrophage efferocytosis.

Protectins (PD1/neuroprotectin D1, from DHA) and maresins (MaR1, from DHA, produced by macrophages) also promote resolution and tissue repair.

Anti-inflammatory cytokines: IL-10 (produced by Tregs, macrophages, and other cells) suppresses pro-inflammatory cytokine production, downregulates MHC class II and costimulatory molecules on antigen-presenting cells, and inhibits T cell proliferation. TGF-beta suppresses T cell and macrophage activation, promotes tissue repair (fibrosis), and is involved in maintaining peripheral tolerance.

Efferocytosis: Apoptotic neutrophils undergo programmed cell death and display "eat me" signals (especially phosphatidylserine on the outer leaflet of the plasma membrane). Macrophages recognize these signals and engulf the apoptotic cells without releasing pro-inflammatory mediators (non-phlogistic phagocytosis), completing the resolution cycle.

Key results Intermediate+

Result 1 (TLR4 as the LPS receptor). The identification of TLR4 as the signaling receptor for lipopolysaccharide (LPS) by Poltorak et al. (1998) and the positional cloning work of Beutler and colleagues established the molecular basis for innate immune recognition. Mice with mutations in TLR4 (C3H/HeJ strain) are hyporesponsive to LPS, confirming that TLR4 is essential for Gram-negative bacterial detection. LPS first binds to LBP (LPS-binding protein), which transfers it to CD14 (a GPI-anchored receptor on macrophages), which then presents it to the TLR4/MD-2 complex. TLR4 dimerization triggers intracellular signaling via the adaptor MyD88 (MyD88-dependent pathway, leading to NF-kappaB activation and pro-inflammatory cytokine production) and TRIF (MyD88-independent pathway, leading to IRF3 activation and type I interferon production). Bruce Beutler received the Nobel Prize in 2011 for this work.

Result 2 (The inflammasome). The NLRP3 inflammasome, characterized by Tschopp and colleagues (2002), is a cytoplasmic multiprotein complex consisting of NLRP3, the adaptor ASC (apoptosis-associated speck-like protein containing a CARD), and procaspase-1. When activated by diverse stimuli -- extracellular ATP (via P2X7 receptor and K+ efflux), crystalline substances (monosodium urate crystals in gout, asbestos, silica), reactive oxygen species, and lysosomal damage -- NLRP3 oligomerizes, recruits ASC, and activates caspase-1. Active caspase-1 cleaves pro-IL-1beta and pro-IL-18 into their active forms and cleaves gasdermin D, whose N-terminal fragment forms membrane pores, triggering pyroptosis. Gain-of-function mutations in NLRP3 cause cryopyrin-associated periodic syndromes (CAPS): familial cold autoinflammatory syndrome (FCAS), Muckle-Wells syndrome, and chronic infantile neurologic cutaneous articular (CINCA) syndrome.

Result 3 (Resolution is active, not passive). The paradigm shift from viewing resolution as passive dissipation of mediators to an active, receptor-mediated process was established by Serhan and colleagues beginning in the 1990s. They discovered that during the inflammatory response, there is a lipid mediator class switch: pro-inflammatory prostaglandins and leukotrienes are progressively replaced by anti-inflammatory and pro-resolving lipoxins, resolvins, protectins, and maresins. These specialized pro-resolving mediators (SPMs) act at nanomolar concentrations through specific G-protein-coupled receptors (ALX/FPR2 for lipoxins, ChemR23 for resolvin E1, GPR32 for resolvin D1) to block neutrophil infiltration, enhance macrophage efferocytosis, promote tissue repair, and stimulate mucosal host defense. This discovery has therapeutic implications: rather than blocking inflammation (as NSAIDs and steroids do), enhancing resolution may offer a more physiologic approach to treating inflammatory diseases.

Exercise 1

Exercise 1 (medium, short answer).

Describe the four-step cascade of leukocyte recruitment from the bloodstream into inflamed tissue, naming the key adhesion molecules at each step.

Hint

The steps are: rolling, activation, firm adhesion, and transmigration. Selectins mediate rolling, chemokines trigger activation, integrins mediate firm adhesion, and PECAM-1 facilitates transmigration.

Answer

Rolling. Activated endothelial cells express E-selectin (induced by IL-1 and TNF-alpha, appears 2-4 hours after activation) and P-selectin (translocated from Weibel-Palade bodies to the surface within minutes of histamine stimulation). These selectins bind to carbohydrate ligands on leukocytes, especially PSGL-1 (P-selectin glycoprotein ligand-1) on neutrophils. The interaction is weak and rapidly forming/breaking, causing the leukocyte to roll along the vessel wall.
Activation. Chemokines displayed on the endothelial surface (bound to glycosaminoglycans to maintain high local concentration), particularly IL-8/CXCL8 and other CXC chemokines, bind to CXC chemokine receptors (CXCR1, CXCR2) on the rolling leukocyte. This G-protein-coupled receptor engagement triggers intracellular signaling that activates the leukocyte's beta-2 integrins (LFA-1/CD11aCD18 and Mac-1/CD11bCD18).
Firm adhesion. The activated integrins undergo a conformational change from a low-affinity to a high-affinity state and bind tightly to immunoglobulin superfamily adhesion molecules on the endothelium: ICAM-1 (intercellular adhesion molecule-1, the ligand for LFA-1 and Mac-1) and VCAM-1 (vascular cell adhesion molecule-1, the ligand for VLA-4/alpha4beta1 on eosinophils, lymphocytes, and monocytes). This arrests the leukocyte on the vessel wall.
Transmigration (diapedesis). The arrested leukocyte crawls along the endothelial surface to find an intercellular junction, then squeezes between endothelial cells (paracellular transmigration) or through endothelial cells (transcellular transmigration). PECAM-1 (CD31), CD99, and JAM (junctional adhesion molecules) at endothelial cell borders facilitate this process. Once through the endothelium, the leukocyte degrades the basement membrane and follows chemokine gradients through the tissue to reach the site of inflammation.

Exercise 2

Exercise 2 (medium, short answer).

Explain how the NLRP3 inflammasome is activated and how its activation leads to IL-1beta secretion. Why can NLRP3 be activated by such diverse stimuli (ATP, uric acid crystals, silica)?

Hint

NLRP3 activation requires two signals. The first primes the cell (upregulates NLRP3 and pro-IL-1beta); the second activates the inflammasome complex. Think about what common intracellular event (like potassium efflux or ROS production) these diverse triggers share.

Answer

NLRP3 inflammasome activation requires two signals:

Signal 1 (Priming): A TLR ligand (e.g., LPS binding TLR4) or cytokine receptor engagement activates NF-kappaB, which upregulates transcription of NLRP3 and pro-IL-1beta (which are not constitutively expressed at high levels). This "primes" the cell but does not activate the inflammasome.

Signal 2 (Activation): A second stimulus triggers NLRP3 oligomerization. Diverse activators converge on common intracellular events:

K+ efflux: Extracellular ATP binds the P2X7 receptor, opening a cation channel and activating pannexin-1 pores, causing potassium efflux from the cell. The drop in intracellular K+ concentration is a key trigger for NLRP3 activation.
Lysosomal damage: Crystalline substances (monosodium urate in gout, calcium pyrophosphate in pseudogout, asbestos, silica, cholesterol crystals) are phagocytosed but damage the phagolysosomal membrane. Lysosomal rupture releases cathepsin B and other lysosomal enzymes into the cytoplasm, which activate NLRP3.
Reactive oxygen species (ROS): Mitochondrial dysfunction and ROS production, triggered by various cellular stresses, can activate NLRP3.
Translocation to the trans-Golgi network: Recent evidence suggests that certain signals cause NLRP3 to traffic to dispersed trans-Golgi network structures, where it senses changes in lipid composition.

The reason diverse stimuli activate NLRP3 is that it functions as a sensor of cellular homeostasis disruption rather than a specific molecular pattern. It detects the common downstream consequences of cellular stress (K+ efflux, lysosomal damage, mitochondrial dysfunction) rather than the specific triggering molecules themselves.

Once activated, NLRP3 oligomerizes and recruits the adaptor ASC (apoptosis-associated speck-like protein containing a CARD) through homotypic PYRIN domain interactions. ASC then recruits procaspase-1 through CARD-CARD interactions. The proximity of multiple procaspase-1 molecules in the inflammasome complex allows auto-cleavage and activation. Active caspase-1 cleaves pro-IL-1beta (31 kDa) into mature IL-1beta (17 kDa) and pro-IL-18 into mature IL-18. Caspase-1 also cleaves gasdermin D; the N-terminal fragment of gasdermin D forms pores in the plasma membrane, allowing IL-1beta and IL-18 secretion and triggering pyroptosis.

Advanced treatment Master

Sepsis pathophysiology

Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection. The current definition (Sepsis-3, 2016) defines sepsis as infection plus organ dysfunction assessed by the SOFA (Sequential Organ Failure Assessment) score. Septic shock is sepsis with persistent hypotension requiring vasopressors to maintain MAP >= 65 mmHg and serum lactate > 2 mmol/L despite adequate fluid resuscitation.

The pathophysiology involves an initial hyperinflammatory phase (SIRS: systemic inflammatory response syndrome) driven by massive cytokine release, followed by a prolonged immunosuppressive phase characterized by T cell exhaustion, apoptotic depletion of lymphocytes, and increased susceptibility to secondary infections. The hyperinflammatory phase features widespread endothelial activation, glycocalyx degradation, increased vascular permeability ("leaky capillaries"), hypovolemia, hypotension, disseminated intravascular coagulation (DIC), and multi-organ failure. The metabolic consequences include hyperglycemia (stress-induced), lactic acidosis (from tissue hypoperfusion and mitochondrial dysfunction), and catabolic muscle wasting.

Clinical assessment: The qSOFA (quick SOFA) score is a bedside screening tool: altered mental status (GCS < 15), systolic blood pressure <= 100 mmHg, and respiratory rate >= 22/min. A score of 2 or more in a patient with suspected infection predicts in-hospital mortality and should prompt further assessment.

Management: The Surviving Sepsis Campaign guidelines emphasize early intervention: (1) Obtain blood cultures and other relevant cultures before initiating antibiotics; (2) Administer broad-spectrum antibiotics within 1 hour of recognition (each hour of delay increases mortality by approximately 8%); (3) Administer 30 mL/kg crystalloid fluid for hypotension or lactate >= 4 mmol/L; (4) Apply vasopressors (norepinephrine first-line) for persistent hypotension despite fluid resuscitation, targeting MAP >= 65 mmHg; (5) Measure serum lactate and remeasure if elevated; (6) De-escalate antibiotics based on culture results and clinical response.

CAR-T cytokine release syndrome

CAR-T cell therapy (chimeric antigen receptor T cells) can cause cytokine release syndrome (CRS) as a result of massive T cell activation upon engaging tumor cells bearing the target antigen (CD19 in B cell malignancies). CRS typically begins 1-14 days after CAR-T infusion. The mechanism involves: CAR-T cells engage tumor cells --> T cell activation and massive proliferation --> release of IFN-gamma, TNF-alpha, and IL-2 --> activation of bystander monocytes and macrophages --> secondary release of IL-6, IL-1, and IL-10 --> systemic inflammatory response. IL-6, produced primarily by activated monocytes and endothelial cells (not the CAR-T cells themselves), is the central mediator of CRS.

Grading: ASTCT consensus grading uses fever, hypotension, and hypoxia: Grade 1 (fever >= 38.0 degrees C), Grade 2 (fever plus hypotension responsive to fluids or hypoxia responsive to < 40% O2), Grade 3 (fever plus hypotension requiring vasopressors or hypoxia requiring >= 40% O2), Grade 4 (life-threatening).

Management: Tocilizumab (anti-IL-6 receptor monoclonal antibody) rapidly reverses CRS symptoms by blocking IL-6 signaling. It was FDA-approved for CAR-T-associated CRS in 2017 and typically works within hours. For severe or refractory CRS, corticosteroids are added (though they may impair CAR-T cell persistence and efficacy). Siltuximab (anti-IL-6 monoclonal antibody, neutralizes IL-6 directly) is an alternative. Anakinra (IL-1 receptor antagonist) is used for neurotoxicity (immune effector cell-associated neurotoxicity syndrome, ICANS), which often accompanies CRS.

Chronic inflammation and pharmacologic intervention

When acute inflammation fails to resolve, or when the inflammatory stimulus persists, chronic inflammation develops. Chronic inflammation is characterized by infiltration of monocytes/macrophages, lymphocytes, and plasma cells; tissue destruction; and attempts at healing (angiogenesis, fibrosis). Chronic inflammation underlies many diseases: rheumatoid arthritis, inflammatory bowel disease, psoriasis, atherosclerosis, type 2 diabetes, Alzheimer disease, and cancer.

TNF inhibitors: Infliximab, adalimumab, etanercept (monoclonal antibodies or receptor fusion proteins that neutralize TNF-alpha) are used in rheumatoid arthritis, Crohn disease, psoriasis, and ankylosing spondylitis. They reduce inflammation but increase infection risk (especially reactivation of latent tuberculosis) and may increase lymphoma risk.

Anti-IL-17: Secukinumab, ixekizumab (anti-IL-17A monoclonal antibodies) are used in psoriasis, psoriatic arthritis, and ankylosing spondylitis. IL-17, produced by Th17 cells, is a key driver of neutrophil recruitment and tissue inflammation in these conditions.

JAK inhibitors: Tofacitinib, baricitinib, upadacitinib inhibit Janus kinases (JAK1, JAK2, JAK3, TYK2), blocking signaling through multiple cytokine receptors (IL-6, IFN-gamma, IL-12, IL-23, and others) that use JAK-STAT pathways. Used in rheumatoid arthritis, psoriatic arthritis, ulcerative colitis, and atopic dermatitis. They carry risks of infection, thromboembolic events, and malignancy.

Inflammasome disorders

Cryopyrin-associated periodic syndromes (CAPS): Autosomal dominant gain-of-function mutations in NLRP3 cause constitutive inflammasome activation, with excessive IL-1beta release. The clinical spectrum ranges from FCAS (familial cold autoinflammatory syndrome: episodes of fever, urticaria-like rash, and arthralgia triggered by cold exposure) to Muckle-Wells syndrome (progressive sensorineural hearing loss, amyloidosis) to CINCA/NOMID (chronic infantile neurologic cutaneous articular syndrome: chronic aseptic meningitis, intellectual disability, joint deformities). Treatment with anakinra (IL-1 receptor antagonist) or canakinumab (anti-IL-1beta monoclonal antibody) is highly effective.

Gout: Monosodium urate crystals deposited in joints activate the NLRP3 inflammasome, causing acute attacks of intense inflammatory arthritis (classically the first metatarsophalangeal joint -- podagra). Colchicine inhibits microtubule polymerization, blocking neutrophil activation and chemotaxis. NSAIDs and corticosteroids are also used acutely. Long-term urate-lowering therapy (allopurinol, febuxostat) prevents crystal formation.

Autoinflammatory vs. autoimmune diseases

Autoinflammatory diseases are caused by dysregulation of the innate immune system, with episodes of seemingly unprovoked inflammation without high-titer autoantibodies or antigen-specific T cells. Examples: CAPS, familial Mediterranean fever (MEFV/pyrin mutations), TRAPS (TNF receptor-associated periodic syndrome, TNFRSF1A mutations), and gout. Treatment targets innate cytokines (especially IL-1beta).

Autoimmune diseases involve dysregulation of the adaptive immune system, with loss of self-tolerance, autoantibodies, and autoreactive T cells targeting specific self-antigens. Examples: systemic lupus erythematosus, rheumatoid arthritis, type 1 diabetes, multiple sclerosis. Treatment targets adaptive immune mechanisms (B cells, T cells, costimulation, cytokines driving adaptive responses).

The distinction is not absolute -- many diseases have elements of both (e.g., rheumatoid arthritis involves both innate and adaptive immune dysregulation) -- but it guides therapeutic strategy.

NSAID and steroid anti-inflammatory mechanisms

NSAIDs (non-steroidal anti-inflammatory drugs) inhibit cyclooxygenase enzymes, which convert arachidonic acid to prostaglandin H2 (the precursor of PGE2, PGI2, and thromboxane A2). COX-1 is constitutively expressed in most tissues and produces prostaglandins that maintain normal physiological functions (gastric mucosal protection via PGE2, platelet aggregation via thromboxane A2, renal blood flow). COX-2 is induced at sites of inflammation and produces pro-inflammatory prostaglandins. Traditional NSAIDs (ibuprofen, naproxen, indomethacin) inhibit both COX-1 and COX-2, which accounts for both their anti-inflammatory effect and their side effects (gastric ulceration from loss of protective prostaglandins, bleeding from platelet inhibition). COX-2 selective inhibitors (celecoxib) were developed to spare gastric mucosal protection but carry increased cardiovascular risk (rofecoxib/Vioxx was withdrawn for this reason).

Glucocorticoids (cortisol, prednisone, dexamethasone) are the most potent anti-inflammatory agents available. They bind the cytoplasmic glucocorticoid receptor, which translocates to the nucleus and: (1) directly binds glucocorticoid response elements (GREs) in target gene promoters to upregulate anti-inflammatory genes (annexin-1/lipocortin-1, IkappaB-alpha, IL-10); (2) transrepresses pro-inflammatory transcription factors, particularly NF-kappaB and AP-1, by physical interaction that prevents their binding to DNA or recruitment of coactivators. This transrepression downregulates virtually all pro-inflammatory cytokines, chemokines, adhesion molecules, COX-2, and iNOS. Glucocorticoids also induce apoptosis of lymphocytes and eosinophils and inhibit leukocyte migration to inflamed sites. Their side effects (osteoporosis, hyperglycemia, immunosuppression, adrenal suppression, cataracts, skin atrophy) result from these same potent anti-inflammatory and metabolic effects on non-target tissues.

Exercise 3

Exercise 3 (hard, analysis).

A patient with relapsed B cell acute lymphoblastic leukemia receives CD19-directed CAR-T cell therapy. Six hours after infusion, the patient develops fever (39.5 degrees C), hypotension (BP 85/50), and elevated serum IL-6 (5,000 pg/mL). Explain the pathophysiology of this cytokine release syndrome, including why IL-6 (not a T cell cytokine) is the dominant mediator. How would you manage this patient?

Hint

CAR-T cells are the initial trigger but not the primary source of IL-6. Think about the bystander cell cascade: CAR-T activation --> IFN-gamma/TNF-alpha release --> monocyte/macrophage activation --> massive IL-6 production. Tocilizumab targets the IL-6 receptor.

Answer

The pathophysiology of CAR-T-associated CRS follows a two-cell model:

Phase 1: CAR-T cell activation. The infused CAR-T cells encounter CD19-positive B cells/tumor cells in the patient's body. Engagement of the chimeric antigen receptor triggers massive T cell activation, proliferation, and release of T cell effector cytokines: IFN-gamma, TNF-alpha, and IL-2. This occurs primarily in the tumor microenvironment and secondary lymphoid organs.

Phase 2: Bystander cell activation. The T cell-derived cytokines (especially IFN-gamma and TNF-alpha) activate bystander innate immune cells -- particularly monocytes and tissue macrophages. These activated monocytes/macrophages produce large quantities of IL-6, IL-1, IL-10, and additional TNF-alpha. Endothelial cells activated by TNF-alpha and IL-1 also produce IL-6 and express adhesion molecules.

Why IL-6 dominates: IL-6 is the central effector cytokine of CRS, but it is produced primarily by activated monocytes and endothelial cells, not by the CAR-T cells themselves. IL-6 levels correlate with CRS severity and drive the clinical features: fever (via PGE2 synthesis in the hypothalamus), hypotension (via increased vascular permeability and NO-mediated vasodilation), coagulopathy, and endothelial activation. The "cytokine storm" is thus not a direct product of CAR-T cells but an amplification cascade through the innate immune system.

Management: This patient has Grade 2 CRS (fever plus hypotension responsive to fluids). The recommended treatment is:

Tocilizumab (8 mg/kg IV, maximum 800 mg, may repeat after 12 hours if needed): a humanized monoclonal antibody against the IL-6 receptor (binds both membrane-bound and soluble IL-6 receptors). It blocks IL-6 signaling through both classical (membrane receptor) and trans-signaling (soluble receptor) pathways. Tocilizumab typically reduces fever and improves hemodynamics within hours.
Supportive care: IV fluid resuscitation for hypotension, antipyretics for fever, and close monitoring (ICU if progressing to Grade 3-4 with vasopressor-requiring hypotension or hypoxia).
Corticosteroids (e.g., dexamethasone 10 mg IV q6h) are added if CRS is severe (Grade 3-4) or refractory to tocilizumab. However, corticosteroids may impair CAR-T cell persistence and anti-tumor efficacy, so they are used judiciously.
Anakinra (IL-1 receptor antagonist) may be considered for neurotoxicity (ICANS), which often accompanies CRS and may be driven by IL-1 rather than IL-6.

Exercise 4

Exercise 4 (hard, analysis).

Explain why NSAIDs and glucocorticoids, despite being potent anti-inflammatory drugs, do not promote resolution of inflammation. Contrast their mechanism with that of specialized pro-resolving mediators (SPMs). What are the therapeutic implications of this distinction?

Hint

NSAIDs block the production of prostaglandins, but prostaglandins are eventually converted into lipoxins (pro-resolving mediators). Steroids broadly suppress all inflammatory gene expression. Neither actively promotes clearance of inflammatory cells or tissue repair. SPMs are agonists that actively drive resolution.

Answer

NSAIDs and glucocorticoids are anti-inflammatory but not pro-resolving -- they block inflammation without promoting the active processes needed for tissue repair and return to homeostasis.

NSAIDs inhibit COX enzymes, blocking prostaglandin and thromboxane production. However, during normal resolution, prostaglandin E2 (PGE2) and prostaglandin D2 (PGD2) actually signal through their respective receptors to upregulate 15-lipoxygenase (15-LOX) expression in neutrophils and macrophages. 15-LOX is the enzyme that initiates the synthesis of lipoxins (from arachidonic acid) and resolvins (from omega-3 fatty acids). Thus, prostaglandins at the inflammatory site serve a dual role: early in inflammation they are pro-inflammatory, but later they trigger the biosynthesis of pro-resolving mediators. By blocking COX, NSAIDs may inadvertently impair the endogenous production of SPMs. Animal studies have shown that COX inhibition can delay resolution of acute inflammation, and low-dose aspirin uniquely promotes 15-epi-lipoxin A4 production (through acetylation of COX-2) while blocking pro-inflammatory prostaglandins, which may explain aspirin's unique cardiovascular benefits.

Glucocorticoids broadly suppress gene expression through NF-kappaB transrepression, inhibiting the production of virtually all pro-inflammatory cytokines, chemokines, adhesion molecules, COX-2, and iNOS. While this effectively stops the inflammatory response, it does not actively clear apoptotic neutrophils, stimulate macrophage phenotype switching, or promote tissue repair. Glucocorticoids may actually impair wound healing and tissue regeneration by suppressing fibroblast proliferation, collagen synthesis, and angiogenesis.

Specialized pro-resolving mediators (SPMs) -- lipoxins, resolvins, protectins, maresins -- are not anti-inflammatory in the conventional sense (they do not block mediator production). Instead, they are pro-resolving agonists that actively stimulate: (1) cessation of neutrophil infiltration; (2) macrophage efferocytosis of apoptotic neutrophils (clearing the inflammatory site without releasing pro-inflammatory contents); (3) macrophage phenotype switching from pro-inflammatory (M1) to pro-repair (M2); (4) tissue regeneration and return to homeostasis; and (5) mucosal host defense. SPMs act through specific G-protein-coupled receptors at nanomolar concentrations.

The therapeutic implication is that "anti-inflammatory" and "pro-resolving" are not synonymous. Current drugs (NSAIDs, steroids) block inflammation but may delay resolution and tissue repair. A new therapeutic approach -- resolution pharmacology -- aims to develop drugs that enhance SPM production or mimic SPM action (synthetic resolvins, lipoxin analogs) to promote active resolution without immunosuppression. This could be particularly valuable in conditions where impaired resolution contributes to chronic inflammation, such as atherosclerosis, periodontitis, and inflammatory bowel disease. Clinical trials of synthetic SPM analogs are ongoing.

Exercise 5

Exercise 5 (medium, short answer).

Compare autoinflammatory and autoimmune diseases. Give one example of each and explain how the distinction guides treatment.

Hint

Autoinflammatory: innate immune system, no autoantibodies, episodic inflammation. Autoimmune: adaptive immune system, autoantibodies and autoreactive T cells, specific self-antigens. Think about CAPS vs. lupus.

Answer

Autoinflammatory diseases result from dysregulation of the innate immune system, with episodes of seemingly unprovoked systemic inflammation in the absence of high-titer autoantibodies or antigen-specific T cells. The inflammation is driven by innate immune cells (neutrophils, macrophages) and innate cytokines (especially IL-1beta). Example: Cryopyrin-associated periodic syndromes (CAPS) caused by gain-of-function mutations in NLRP3 leading to constitutive inflammasome activation and excessive IL-1beta secretion. Treatment targets IL-1: anakinra (IL-1 receptor antagonist) or canakinumab (anti-IL-1beta monoclonal antibody) are highly effective, rapidly controlling symptoms.

Autoimmune diseases involve loss of self-tolerance by the adaptive immune system, with production of autoantibodies and/or activation of autoreactive T cells directed against specific self-antigens. Example: Systemic lupus erythematosus (SLE), in which autoantibodies against nuclear antigens (anti-dsDNA, anti-Sm) form immune complexes that deposit in kidneys, skin, joints, and other organs, causing inflammation through complement activation and Fc receptor engagement. Treatment targets adaptive immune mechanisms: hydroxychloroquine (modulates TLR signaling), mycophenolate mofetil (inhibits lymphocyte proliferation), belimumab (anti-BAFF, blocks B cell survival), and rituximab (anti-CD20, depletes B cells).

The distinction guides treatment: autoinflammatory diseases respond to innate cytokine blockade (especially IL-1 and IL-6), while autoimmune diseases require targeting adaptive immune components (B cells, T cells, costimulatory molecules, or cytokines driving adaptive responses like IL-12/23). However, the boundary is blurred -- rheumatoid arthritis, for example, has features of both (innate immune activation in the synovium and adaptive immune-driven autoantibodies like rheumatoid factor and anti-CCP).

Connections Master

Immunology overview 18.10.01. This unit extends the innate immunity section of 18.10.01, providing detailed mechanistic coverage of pattern recognition (TLRs, NLRs, RLRs, CLRs) and the inflammatory cascade that were introduced as part of the innate immune system. The cytokine storm and its management in CAR-T therapy connect to the CAR-T discussion in the immunology overview. The inflammasome details here expand on the NLRP3 inflammasome introduced in 18.10.01.
Cell signaling 17.07.01. TLR signaling uses the same adaptor protein and kinase cascade principles described in 17.07.01. TLR4 signals through MyD88 --> IRAK4 --> TRAF6 --> TAK1 --> IKK --> NF-kappaB, a pathway that parallels receptor tyrosine kinase signaling in its use of adaptor proteins and kinase cascades. The JAK-STAT pathway used by cytokine receptors (IL-6/JAK/STAT3, IFN-gamma/JAK1-2/STAT1) is a direct application of the tyrosine kinase-linked receptor signaling described in 17.07.01. Glucocorticoid receptor signaling (nuclear receptor --> transcription factor) connects to the steroid hormone signaling pathway.
Cardiovascular system 18.02.01. The vascular changes in inflammation (vasodilation, increased permeability, endothelial activation) are modifications of normal cardiovascular physiology. The selectin-integrin adhesion cascade occurs at the endothelial surface and depends on hemodynamic shear forces that are part of cardiovascular function. Septic shock involves cardiovascular collapse: distributive shock from profound vasodilation (mediated by iNOS-derived NO), hypovolemic component from capillary leak, and myocardial depression (from TNF-alpha and other myocardial depressant factors). Fluid resuscitation and vasopressor therapy are applications of cardiovascular pharmacology.
Renal physiology 18.08.01. NSAIDs inhibit renal prostaglandin synthesis (PGE2, PGI2), which normally maintains renal blood flow through vasodilation of the afferent arteriole. In patients dependent on prostaglandin-mediated renal vasodilation (heart failure, cirrhosis, volume depletion), NSAIDs can cause acute kidney injury by unopposed afferent arteriole constriction. This connects to the renin-angiotensin-aldosterone system and renal autoregulation discussed in 18.08.01.
Endocrine system 18.07.01. The hypothalamic-pituitary-adrenal (HPA) axis is the endogenous anti-inflammatory system: stress/inflammation triggers CRH --> ACTH --> cortisol release, and cortisol suppresses inflammation via the glucocorticoid receptor. Chronic inflammation or chronic steroid therapy suppresses the HPA axis, leading to adrenal insufficiency when steroids are withdrawn abruptly. The stress-induced hyperglycemia of sepsis involves cortisol, catecholamines, and glucagon, connecting to glucose homeostasis.
Molecular biology 17.06.01. NF-kappaB regulation involves IkappaB (inhibitor of kappaB) proteins that sequester NF-kappaB in the cytoplasm. TLR signaling leads to IKK-mediated phosphorylation of IkappaB, triggering its ubiquitination and proteasomal degradation, freeing NF-kappaB to translocate to the nucleus. This is a textbook example of regulated proteolysis controlling transcription factor activity. Glucocorticoid receptor transrepression involves protein-protein interactions that inhibit transcription factor binding without direct DNA binding.
Digestive physiology 18.06.01. NSAID gastropathy is a direct consequence of COX-1 inhibition: loss of PGE2-mediated gastric mucosal protection (mucus secretion, bicarbonate secretion, mucosal blood flow, epithelial cell turnover). This is the most common side effect of traditional NSAIDs and illustrates the physiological role of constitutive prostaglandin production. Inflammatory bowel disease (Crohn disease, ulcerative colitis) represents chronic inflammation of the gastrointestinal tract with both innate and adaptive immune dysregulation.

Historical & philosophical context Master

The concept of inflammation is one of the oldest in medicine. The Egyptian Edwin Smith papyrus (circa 1600 BCE) describes inflammatory wound healing. Celsus (1st century CE) defined the four cardinal signs: rubor, tumor, calor, dolor. John Hunter (18th century) recognized inflammation as a beneficial, adaptive response: "it is not to be looked upon as a disease, but as a salutary operation consequent on some irritation or violence." Rudolf Virchow (19th century) added functio laesa (loss of function) as a fifth sign and provided the cellular theory of inflammation, demonstrating that inflammation involves cellular proliferation and migration, not just humoral changes.

The molecular basis of innate immune recognition was established in the late 20th century. The discovery of the Drosophila Toll pathway by Jules Hoffmann and colleagues (1996) revealed that the fruit fly immune response to fungal infection depends on a transmembrane receptor (Toll) that activates an intracellular signaling cascade homologous to the mammalian IL-1 receptor pathway. Bruce Beutler subsequently identified mammalian Toll-like receptors as the sensors for bacterial products (TLR4 for LPS, 1998). Hoffmann and Beutler shared the Nobel Prize in 2011. Charles Janeway had predicted the existence of pattern recognition receptors in 1989, proposing that the immune system distinguishes infectious non-self from non-infectious self through germline-encoded receptors that detect conserved microbial patterns -- a theoretical framework that preceded and guided the experimental discoveries.

The discovery that inflammation resolution is an active process, not passive dissipation, represents a paradigm shift. Charles Serhan and colleagues at Brigham and Women's Hospital, beginning in the 1990s, used lipid mediator metabololipidomics to identify lipoxins, resolvins, protectins, and maresins as endogenous mediators that actively promote resolution. This work overturned the prevailing view (dating to John Hunter) that inflammation simply burns itself out when the stimulus is removed, replacing it with a model in which resolution is a programmed, receptor-driven phase of the inflammatory response with its own specialized biochemistry. The clinical implications are significant: if chronic inflammatory diseases result not only from excessive initiation but also from failed resolution, then therapies that enhance resolution (rather than merely blocking initiation) could be more effective and less immunosuppressive.

The cytokine storm concept emerged from two distinct clinical contexts. In infectious disease, the observation that the 1918 H1N1 influenza pandemic killed young adults (with robust immune systems) more frequently than the elderly suggested that an overactive immune response, not the virus itself, was the cause of death. This was confirmed in H5N1 avian influenza (1997-present), where autopsies showed diffuse alveolar damage and hemophagocytosis with high cytokine levels, and in the SARS-CoV-2 pandemic (2020), where elevated IL-6, TNF-alpha, and IL-1beta correlated with severe disease and mortality. In immunotherapy, the recognition that CAR-T cell therapy causes cytokine release syndrome led to the rapid development and FDA approval of tocilizumab (anti-IL-6 receptor) for CRS in 2017, one of the fastest drug approvals for an immune-related complication.

The pharmacology of anti-inflammatory drugs has its own rich history. Aspirin (acetylsalicylic acid), derived from salicylate-containing willow bark used since ancient times, was synthesized by Bayer in 1897. Its mechanism -- inhibition of prostaglandin synthesis by acetylating COX enzymes -- was elucidated by John Vane in 1971 (Nobel Prize 1982). Cortisone was discovered by Philip Hench and Edward Kendall in 1949 (Nobel Prize 1950) as a treatment for rheumatoid arthritis, though its mechanism (glucocorticoid receptor-mediated transrepression of NF-kappaB and AP-1) was not understood until decades later. The COX-2 selective inhibitors (celecoxib, rofecoxib) were developed in the 1990s based on the hypothesis that selective COX-2 inhibition would spare gastric mucosal protection while maintaining anti-inflammatory efficacy. The withdrawal of rofecoxib (Vioxx) in 2004 due to increased cardiovascular risk (myocardial infarction and stroke) demonstrated that prostaglandin biology is more complex than the simple COX-1/COX-2 dichotomy suggested: prostacyclin (PGI2, COX-2-derived) is cardioprotective (inhibits platelet aggregation, causes vasodilation), and its selective inhibition tips the balance toward thrombosis.

The philosophical dimension of inflammation concerns the balance between defense and damage. Inflammation is fundamentally a protective response -- without it, even minor infections would be lethal (as in neutropenic patients). But the same mechanisms that destroy pathogens also destroy host tissue, and when the response is excessive or inappropriately triggered, inflammation becomes pathological. This duality is not a design flaw but an inherent feature of a system that must act rapidly and destructively against threats that are themselves rapidly evolving. The evolutionary trade-off between speed and specificity in innate immunity, between effective defense and collateral damage, is a recurring theme in immunology. Polly Matzinger's "danger model" (1994) proposed that the immune system responds to danger signals (DAMPs from damaged cells) rather than to non-self per se, reframing inflammation as a damage-sensing system rather than a pathogen-sensing system -- a conceptual shift that unifies responses to infection, trauma, ischemia, and cancer under a single framework.

Bibliography Master

Sherwood, L. Human Physiology: From Cells to Systems, 9th ed. (Cengage Learning, 2016). Ch. 18.
Silverthorn, D. U. Human Physiology: An Integrated Approach, 8th ed. (Pearson, 2019). Ch. 24.
Murphy, K. & Weaver, C. Janeway's Immunobiology, 10th ed. (Garland Science, 2022). Ch. 3-4.
Medzhitov, R. "Origin and physiological roles of inflammation." Nature 454 (2008) 428-435.
Poltorak, A. et al. "Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene." Science 282 (1998) 2085-2088.
Martinon, F., Burns, K. & Tschopp, J. "The inflammasome: a molecular platform triggering activation of inflammatory caspases and processing of proIL-beta." Mol. Cell 10 (2002) 417-426.
Serhan, C. N. "Pro-resolving lipid mediators are leads for resolution physiology." Nature 510 (2014) 92-101.
Lee, D. W. et al. "Current concepts in the diagnosis and management of cytokine release syndrome." Blood 124 (2014) 188-195.
Singer, M. et al. "The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3)." JAMA 315 (2016) 801-810.
Vane, J. R. "Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs." Nat. New Biol. 231 (1971) 232-235.
Netea, M. G. et al. "A guiding map for inflammation." Nat. Immunol. 18 (2017) 826-831.