Innate Immunity

The innate immune system provides immediate, non-specific defense against pathogens and plays a critical role in initiating and shaping adaptive immune responses. Multiple components of innate immunity show abnormalities in ME/CFS.

1 Natural Killer (NK) Cell Dysfunction

Natural killer cell abnormalities represent one of the most replicated findings in ME/CFS research, with impaired NK cell function reported across numerous independent studies spanning decades. A 2019 systematic review of 17 case-control studies (1994–2018) found that impaired NK cell cytotoxicity remained the most consistent immunological abnormality across all publications .

1.1 Reduced NK Cell Cytotoxicity

NK cells eliminate virus-infected and malignant cells through direct cytotoxic mechanisms. ME/CFS patients consistently demonstrate decreased cytotoxic activity, with reduced ability to kill target cells (typically K562 erythroleukemia cells in standard assays). The magnitude of this impairment is substantial, with studies reporting statistically significant reductions across multiple cohorts . Lower NK cell function correlates with greater symptom severity in some studies. These abnormalities remain stable over time, suggesting a chronic rather than transient dysfunction. A 2024 meta-analysis of 28 studies encompassing approximately 2,982 subjects quantified this deficit as a large effect (Hedges’ g = 0.96, 95% CI: 0.75–1.18), with ME/CFS NK cytotoxicity approximately half that of healthy controls .

However, the largest multi-site study to date produced a null result. The MCAM (Multi-site Clinical Assessment of ME/CFS) sub-study tested 174 ME/CFS patients and 86 healthy controls from five US specialty clinics using a flow cytometry-based killing assay on overnight-shipped PBMCs (Querec et al. 2023). Mean cytotoxicity was virtually identical: 34.1% in ME/CFS versus 33.6% in controls (\(p = 0.79\)). No associations with disease severity, illness duration, or onset type were found. Notably, Nancy Klimas—whose foundational 1990 study first documented large NK deficits—is a co-author on this null study, lending it credibility.

Limitation: NK Cytotoxicity: Unresolved Methodological Dispute

The MCAM null result was excluded from the Baraniuk 2024 meta-analysis on methodological grounds: overnight sample shipping may degrade cytotoxicity sufficiently to erase real group differences, and the intracellular DNA staining endpoint detects late-stage cell death rather than early apoptosis (unlike Annexin V used in positive studies). The MCAM team’s own validation study found good correlation between their shipped-sample fluorescence assay and same-day chromium-51 release (\(R^2 = 0.80\), \(p = 0.001\)), arguing that group-level differences should still be detectable (Querec et al. 2023). Whether this correlation is sufficient to claim equivalence remains contested. The dispute has practical implications: if NK cytotoxicity requires same-day local processing to detect, it cannot serve as a scalable clinical biomarker regardless of its biological reality.

1.2 Mechanisms of Impaired Cytotoxicity

Several mechanisms may underlie reduced NK cell function:

Perforin and Granzyme Deficiency NK cells kill targets by releasing cytotoxic granules containing perforin (which creates pores in target cell membranes) and granzymes (which trigger apoptosis). Maher et al. (2005) demonstrated a mechanistic basis for impaired cytotoxicity: ME/CFS patients show a 45% reduction in NK cell perforin content (3,320 vs 6,051 rMol/cell, \(p = 0.01\)), with significant correlation between perforin levels and cytotoxic function (Maher, Klimas, and Fletcher 2005). Additionally, Brenu et al. (2011) found a paradoxical pattern of elevated perforin but decreased granzyme A and K expression in a large cohort (n=95), suggesting dysfunction in granzyme production or granule composition despite adequate perforin (Brenu et al. 2011). These cells exhibit impaired degranulation despite successfully recognizing target cells, indicating dysfunction in granule trafficking and release mechanisms.

Receptor Abnormalities NK cell activation is regulated by a balance between activating and inhibitory receptors. ME/CFS patients show altered expression of activating receptors (NKG2D, NKp46, NKp30) along with changed inhibitory receptor profiles (Brenu et al. 2011) . Additionally, signaling downstream of activating receptors is impaired, and calcium flux following receptor engagement is disrupted .

Metabolic Dysfunction NK cells require substantial energy for cytotoxic function. ME/CFS NK cells exhibit impaired glycolytic metabolism and mitochondrial dysfunction affecting ATP production (Brenu et al. 2011). This reduced metabolic reserve limits their capacity for sustained activity. The concept of mitochondrial reserve as a functional bottleneck extends beyond immune cells: spare respiratory capacity may also limit thermoregulatory work (Spare Respiratory Capacity as Thermoregulatory Capacity Proxy in ME/CFS), illustrating how the same cellular energetic deficit manifests across multiple physiological systems. A 2026 study in SLE demonstrated that NK cell mitochondrial dysfunction — enlarged dysfunctional mitochondria with impaired mitophagy and lysosomal acidification — directly drives defective cytotoxicity (Fluder et al. 2026), raising the testable hypothesis that mitochondrial impairment may contribute to, rather than merely accompany, NK dysfunction in ME/CFS. Note that SLE NK cells operate in a distinct immunological environment (type I IFN, BAFF, immune complexes) absent in ME/CFS; whether mitochondrial structural abnormalities in ME/CFS NK cells are comparable remains uncharacterized.

1.3 NK Cell Subsets

Human NK cells are divided into functionally distinct subsets. CD56^bright NK cells primarily produce cytokines and are found mainly in lymphoid tissues, while CD56^dim NK cells are primarily cytotoxic and predominate in peripheral blood. ME/CFS studies have reported altered CD56^bright/CD56^dim ratios, with an increased proportion of CD56^bright cells in some studies . Reduced absolute numbers of CD56^dim cytotoxic cells and abnormal maturation patterns have also been observed .

1.4 Clinical Significance of NK Cell Dysfunction

Impaired NK cell function may contribute to ME/CFS through several mechanisms:

Viral reactivation: Inadequate control of latent herpesviruses (EBV, HHV-6, CMV)
Tumor surveillance: Theoretical increased cancer risk (though not clearly demonstrated)
Immune regulation: NK cells modulate other immune cells; dysfunction may permit chronic inflammation
Infection susceptibility: Reduced defense against new infections

These mechanisms may form a self-reinforcing cycle rather than a simple linear causal chain. In particular, the relationship between NK cell dysfunction and viral reactivation is bidirectional: impaired NK function permits viral reactivation, but chronic viral reactivation itself may further exhaust and dysregulate NK cells. The following section examines viral persistence and clearance failure mechanisms that may drive this cycle, with testable predictions for distinguishing recovery from chronicity. This bidirectional cycle represents one of several vicious cycles maintaining ME/CFS pathophysiology, discussed comprehensively in Section Unifying Mechanisms Across Systems of Chapter Integrative Models and Multi-System Pathophysiology.

2 Viral Persistence and Clearance Failure

ME/CFS is predominantly a post-infectious syndrome, with 60-80% of cases reporting onset following an acute infection. The critical question is not just which pathogens trigger ME/CFS, but why some patients successfully clear post-viral pathology while others develop chronic fatigue syndromes. The divergence between recovery and chronicity may depend on viral clearance dynamics during a critical window following acute infection.

2.1 The Critical Clearance Window Hypothesis

Hypothesis: Critical Clearance Window Determining Chronic Outcomes

Certainty: 0.40. There exists a time-limited window (approximately 2-8 weeks post-infection) during which effective viral clearance determines long-term outcome. Patients who fail to clear viral antigens and resolve immune activation within this window are at high risk of developing ME/CFS, while those who achieve complete clearance recover fully.

Evidence Base. Longitudinal studies of infectious mononucleosis show that approximately 10-12% of patients develop CFS-like symptoms at 6 months post-infection (the Dubbo cohort finding), with the failure rate being remarkably pathogen-agnostic (similar rates for EBV, Coxiella burnetii, and Ross River virus). This suggests that the pathogen matters less than the host response pattern. In Long COVID, similar proportions develop persistent symptoms, with immune signatures at 3 months predicting 12-month outcomes. The time course suggests a critical period where the immune system either successfully resolves the post-infectious state or becomes locked in a pathological configuration.

Mechanistic Rationale. During the clearance window, several interconnected processes must occur successfully:

Viral eradication or containment: Complete elimination of replicating virus or establishment of effective latency control
Immune resolution: Transition from acute inflammatory response to homeostatic baseline
Tissue repair: Clearance of damaged cells and restoration of normal tissue function
Immune reprogramming: Re-establishment of normal immune surveillance and tolerance mechanisms

Failure at any of these steps during the critical window may initiate self-sustaining pathological loops that persist indefinitely.

Clinical Implications. This hypothesis has profound implications for early intervention:

Treatment timing: Antiviral or immunomodulatory interventions may only be effective if administered during the clearance window
Biomarker development: Immune profiles during the acute phase could predict who will develop ME/CFS
Prevention strategies: Early aggressive treatment of high-risk patients might prevent chronicity

Testable Predictions.

Immune profiles at 4 weeks post-infection will predict 6-month ME/CFS status with >80% accuracy
Patients who receive antiviral treatment within 2 weeks of infection will have lower ME/CFS incidence than untreated controls
The duration of elevated inflammatory markers (IL-6, CRP) beyond 8 weeks will correlate with ME/CFS severity
Early intervention (within 4 weeks) with immunomodulators will prevent the establishment of chronic immune signatures

Limitations. The specific timing of the clearance window may vary by pathogen and individual factors. The hypothesis assumes that viral clearance is the primary determinant, whereas host factors (genetics, prior immune history) may be equally important. No prospective trial has tested early intervention strategies.

2.2 Viral Reservoir Mechanisms

Multiple mechanisms may enable viral persistence despite apparent clinical recovery:

Active Low-Level Replication. Some evidence suggests that ME/CFS patients may harbor ongoing low-level viral replication that evades standard detection. Sensitive PCR techniques have detected viral RNA in tissues (gut, nervous system) months after apparent recovery. However, the significance of these findings remains controversial, as low-level viral detection does not establish causality.

Viral Latency and Reactivation. Herpesviruses (EBV, HHV-6, CMV) establish lifelong latency following primary infection and can reactivate under conditions of immune stress. ME/CFS patients show elevated antibody titers to these viruses, suggesting frequent reactivation. However, antibody elevation alone cannot distinguish between active replication and immune memory. Consistent with this, a multicenter nucleic-acid surveillance study (PCR and high-throughput sequencing of blood, feces, and saliva) found no group-specific differences in viral nucleic acid between ME/CFS cases and controls, apart from lower anelloviruses in cases (Briese et al. 2023) — indicating that active productive viral replication is not readily detectable by gene-product surveillance, and that antibody/viral-load signals should be interpreted through the lens of immune dysregulation rather than ongoing replication.

Observation: Herpesvirus Triggers in ME/CFS

Certainty: 0.65. Multiple herpesviruses (EBV, CMV, HHV-6) are implicated as triggering factors for ME/CFS based on observational evidence and comprehensive immunopathobiology reviews. EBV-specific B- and T-cell responses are deficient in ME/CFS, while responses to other pathogens remain normal, suggesting specific viral memory dysfunction. CMV-seropositive individuals show expansion of CMV-specific innate-like CD4 T cells (CD161+ CD56+), providing a mechanistic link between persistent viral infection and tissue-resident T cell expansion. Review articles synthesize evidence from multiple cohorts, though primary data remain heterogeneous and not all findings are replicated.

Key findings:

EBV, HHV-6, and CMV infections are consistently reported as ME/CFS triggering factors across studies
Deficient EBV-specific B- and T-cell responses in ME/CFS patients, with normal responses to other viruses and bacteria
CMV-seropositive individuals exhibit expansion of CD161+ CD56+ CD4 T cells enriched in CMV-specific TCRs
This expansion provides a mechanistic link between persistent viral infection and tissue-resident innate-like T cell population changes

Implications. These findings support the viral persistence hypothesis and suggest that specific herpesvirus memory dysfunction may contribute to ME/CFS pathophysiology. The CMV-specific innate-like T cell expansion observed in healthy donors may be dysregulated in ME/CFS, contributing to both local (intestinal) and systemic immune dysfunction.

Limitations. The evidence is primarily from review articles and observational studies. No direct studies have examined CMV-specific tissue-resident T cells in ME/CFS patients. Heterogeneity across studies limits definitive conclusions about causal relationships.

Hypothesis: Compartmentalized Viral Reservoirs in ME/CFS

Certainty: 0.35. ME/CFS patients may harbor viral reservoirs in immune-privileged or poorly accessible compartments (gut-associated lymphoid tissue, central nervous system, bone marrow) that drive chronic immune activation through intermittent viral antigen release. These reservoirs evade standard blood-based detection but maintain continuous immune stimulation.

Evidence Base. Comparative virology provides a compelling model: Kol et al. (2026) demonstrated that feline infectious peritonitis virus (FIPV), a coronavirus, persists in B and T lymphocytes within mesenteric lymph nodes after antiviral treatment and clinical recovery. Because memory lymphocytes survive for years, even a small fraction of virus-harbouring cells constitutes a persistent antigenic reservoir that can explain relapse and chronic immune dysregulation. While FIPV is not SARS-CoV-2, the shared coronavirus biology—tropism for immune cells, persistence despite apparent clearance, and post-treatment relapse—strengthens the plausibility that analogous mechanisms operate in post-COVID ME/CFS.

Mechanistic Implications. If lymphocyte reservoirs exist in ME/CFS:

Viral antigen is continuously presented during any immune activation (infection, vaccination, exercise-induced immune mobilization)
This would explain why diverse immune stimuli can trigger symptom exacerbation
The reservoir maintains chronic antigenic stimulation without requiring active viral replication
Standard antiviral therapies may fail because they cannot access or eliminate the reservoir

Testable Predictions.

Single-cell RNA sequencing of ME/CFS patient lymphocytes will detect viral transcripts in a small percentage (less than 1%) of cells
The fraction of virus-positive lymphocytes will correlate with disease severity and duration
In vitro activation of ME/CFS patient lymphocytes will increase detectable viral antigen
Patients who recover from ME/CFS will show clearance of lymphocyte-associated viral RNA

Limitations. Cross-species extrapolation from felines to humans requires caution. No study has directly demonstrated SARS-CoV-2 or ME/CFS-triggering virus persistence in human lymphocytes. The “below PCR threshold” claim is currently untestable with standard methods.

Viral Protein Persistence. Even without intact virus, viral proteins (particularly SARS-CoV-2 spike protein) can persist in tissues for months. Spike protein has been detected in gut epithelium and immune cells of Long COVID patients up to 12 months post-infection. These persistent proteins may:

Serve as continuous antigenic stimuli driving immune activation
Directly activate immune cells through pattern recognition receptors
Induce autoimmunity through molecular mimicry
Cause endothelial dysfunction and microvascular damage

Impaired Viral Clearance Mechanisms.

Multiple mechanisms may contribute to failed viral clearance in ME/CFS:

NK Cell Dysfunction: Impaired cytotoxicity reduces elimination of virally-infected cells
T Cell Exhaustion: Exhausted CD8+ T cells cannot effectively clear virus-infected cells
Dendritic Cell Dysfunction: Impaired antigen presentation limits adaptive immune response initiation
Complement Consumption: Reduced opsonization impairs clearance of virus-antibody complexes
MDSC Expansion: Immunosuppressive myeloid cells inhibit effective antiviral immunity

These mechanisms create a self-reinforcing cycle: impaired viral clearance maintains antigenic stimulation, which drives immune exhaustion and further impairs clearance capacity.

Hypothesis: MDSC Expansion via Exosomal HSAT2 as the Upstream Cause of NK Cytotoxicity Loss

(Certainty: 0.50 — mechanistic fit to the most-replicated ME/CFS finding; all supporting evidence is from non-ME/CFS contexts.)

The NK cytotoxicity deficit (Hedges’ g = 0.96, 95% CI 0.75–1.18, 28 papers, 55 data points ) has one of the most extensive replication records of any ME/CFS immunological finding, though this assessment is contested — the MCAM null result (n = 174/86, p = 0.79; see above) was excluded from the meta-analysis on methodological grounds that remain disputed. The deficit involves reduced perforin content (Maher, Klimas, and Fletcher 2005), impaired degranulation despite intact target recognition, and dysregulated activating receptor expression (Brenu et al. 2011) . No single upstream driver has been established.

Myeloid-derived suppressor cells (MDSCs; CD33+HLA-DR−) are an immunosuppressive myeloid population expanded by viral infection (Agrati et al. 2020) and shown to persist at 5 months post-SARS-CoV-2 infection in a human cohort . In the context of EBV — the most common ME/CFS trigger following infectious mononucleosis — MDSC expansion during acute primary infection has also been documented (Mihatsch et al. 2026). MDSCs suppress NK cells through mechanisms that directly match the ME/CFS NK phenotype, with the arginase-1 pathway established in HCV-infected patients —

Arginase-1 depletes L-arginine required for NK perforin synthesis (explaining the Maher 2005 perforin deficit (Maher, Klimas, and Fletcher 2005)); Goh et al. demonstrated this suppression is cell-contact-independent and reversed by L-arginine supplementation
Membrane-bound TGFβ1 on M-MDSCs downregulates NKG2D on NK cells, abolishing activation receptor signaling (consistent with Brenu 2011 receptor abnormalities (Brenu et al. 2011))
IDO1-driven tryptophan catabolism impairs NK activation in a paracrine fashion
IL-10 and IL-35 further suppress NK cytotoxicity

Evdokimova et al. demonstrated that exosomal HSAT2/HERV-K transmission to CD33+ myeloid cells induces the full MDSC program — including arginase-1, IDO1, IL-10, IL-35, and TGFβ — proposing a candidate upstream driver for MDSC expansion in post-viral ME/CFS. This positions exosomal HSAT2-driven MDSC expansion as a candidate explanation for the most replicated ME/CFS finding — a single node reconciling perforin depletion, receptor dysregulation, and metabolic impairment. The chronicity follows from the self-perpetuating exosomal loop Exosomal HSAT2/HERV-K RNA as a Self-Perpetuating MDSC Expansion Mechanism: as long as HSAT2 EVs circulate, MDSC-mediated NK suppression is continuous.

Falsifiable prediction: ME/CFS patients with the lowest NK cytotoxicity will have the highest peripheral CD33+HLA-DR− MDSC frequency (Spearman ρ < −0.4). Selective CD33+ depletion from patient PBMCs ex vivo will partially restore NK killing (≥ 20% recovery, K562 assay). If MDSC frequency does not correlate with NK cytotoxicity, the MDSC-bridge model is not supported.

Limitations: No ME/CFS-specific data on MDSC frequency exist — the mechanistic chain is imported from HCV , COVID-19 (Agrati et al. 2020), and cancer contexts. The Beliakova-Bethell 2022 study measured arginase-dependent T-cell suppression; NK cell assays were not performed. EBV-IM MDSC data (Mihatsch 2026) show no ME/CFS follow-up. The upstream exosomal HSAT2 mechanism is speculative in ME/CFS. Not replicated.

Hypothesis: Arginine Depletion as the Nutritional Choke-Point Linking MDSC Expansion to NK Metabolic Failure

(Certainty: 0.45 — mechanism established by Goh 2016 in chronic viral disease; ME/CFS-specific arginine data absent.)

Monocytic MDSCs express high arginase-1, which depletes extracellular L-arginine in the tissue microenvironment . L-arginine is required for NK cell mTOR complex 1 activation; mTOR is required for perforin synthesis, granule polarization, and cytotoxic effector function. In chronic HCV infection, MDSC-driven arginase-1 activity depletes L-arginine and suppresses NK IFN-γ production by approximately 60% — fully reversible by L-arginine supplementation in vitro . ME/CFS NK cytotoxicity is reduced to approximately 50% of controls in meta-analysis .

The arginase-1 mechanism provides a biochemical bridge between MDSC expansion and NK metabolic failure: not direct contact-dependent killing, but enzymatic substrate deprivation. This is clinically significant because: (a) it is theoretically reversible with arginine precursors (L-citrulline, which bypasses intestinal arginine catabolism); (b) plasma arginase-1 activity is measurable as a pharmacodynamic biomarker; (c) the same mechanism operates in iNOS-dependent MDSC suppression documented in SLE (different enzyme, same substrate), suggesting the L-arginine axis is a shared pathway across multiple MDSC-driven immune dysfunctions.

Falsifiable prediction: Plasma L-arginine concentration in ME/CFS will be at least 15% lower than in matched controls and will inversely correlate (Spearman ρ < −0.3) with plasma arginase-1 activity. L-citrulline 6 g/day × 8 weeks will raise plasma L-arginine by ≥ 30% and increase NK IFN-γ ex vivo by ≥ 20% in a n = 20 crossover study. If plasma L-arginine is normal in ME/CFS, the arginase-depletion mechanism is not operative.

Limitations: Plasma L-arginine data in ME/CFS have not been systematically reported. Arginase-1 can be elevated by multiple causes beyond MDSC expansion. L-citrulline has a good safety profile but no ME/CFS trial data exist. The iNOS route (SLE context) and arginase-1 route (HCV/post-COVID context) are distinct — the dominant suppression mechanism in ME/CFS is unknown. Replication status: mechanism well-replicated in cancer/HCV; ME/CFS application not replicated.

### TRPM3 Ion Channel Dysfunction {#sec-trpm3-dysfunction}

A major breakthrough in understanding impaired calcium signaling in ME/CFS immune cells came from research on the TRPM3 ion channel . TRPM3 (Transient Receptor Potential Melastatin 3) is a calcium-permeable ion channel, and calcium signaling is essential for healthy immune cell activity—including the degranulation process disrupted in ME/CFS NK cells.

A study conducted by researchers at Griffith University’s National Centre for Neuroimmunology and Emerging Diseases (NCNED) found that TRPM3 functions abnormally in immune cells of ME/CFS patients compared to healthy controls. This finding was reproduced across two laboratories within the same NCNED network (Gold Coast and Perth, Australia). As noted in the limitation box below, independent replication by groups outside this network has not yet been published; the “4,000 km separation” refers to internal network sites, not independent research groups.

The researchers describe the faulty ion channels as acting like “stuck doors,” preventing cells from receiving the calcium they need for normal function. Calcium signaling is essential for immune cell activity, including NK cell cytotoxic function (degranulation requires calcium influx).

This discovery has several important implications:

Diagnostic potential: TRPM3 dysfunction could serve as an objective biomarker for ME/CFS
Therapeutic targets: Drugs that modulate TRPM3 function might restore normal immune cell activity
Disease legitimacy: Measurable cellular abnormalities provide concrete evidence of biological dysfunction
Mechanistic understanding: TRPM3 dysfunction may explain why NK cells fail to degranulate properly despite recognizing targets

The TRPM3 findings connect to broader ion channel research in ME/CFS and suggest that channelopathy—dysfunction of ion channels—may be a unifying mechanism underlying multiple immune abnormalities observed in the condition.

Limitation: TRPM3 Channelopathy: Single Research Group

TRPM3 ion channel dysfunction in ME/CFS has been characterized primarily by a single research group (Griffith University NCNED). While the multi-site validation (Gold Coast and Perth) strengthens internal reproducibility, independent replication by groups outside this collaboration has not yet been published. The diagnostic biomarker potential, therapeutic implications, and status as a “unifying mechanism” for immune abnormalities remain speculative until independently replicated and mechanistically linked to clinical outcomes in prospective studies. Clinical availability: TRPM3 testing is not commercially available outside the NCNED research context and cannot be ordered by clinicians for individual patient decision-making. See Appendix Ongoing and Planned ME/CFS Research Studies, Section Neuroimaging and Ion Channel Research for current study status and planned LDN trials.

3 Neutrophil and Monocyte Function

3.1 Neutrophil Abnormalities

Neutrophils are the most abundant circulating white blood cells and serve as first responders to infection. Kennedy et al. (2004) demonstrated that ME/CFS patients exhibit increased neutrophil apoptosis (37.4% vs 22.8% annexin V binding, \(p = 0.001\)) with elevated death receptor TNFRI expression (\(p = 0.004\)) and raised active TGF-\(\beta\) 1 concentrations (\(p < 0.005\)), consistent with an activated inflammatory process (Kennedy et al. 2004). Additional ME/CFS-associated abnormalities include:

Phagocytosis Impairment Neutrophils from ME/CFS patients show reduced uptake of bacteria and particles, with impaired phagosome formation and decreased acidification of phagolysosomes .

Respiratory Burst Defects The respiratory burst produces reactive oxygen species to kill ingested pathogens. Some studies have found reduced superoxide production in ME/CFS neutrophils, along with impaired NADPH oxidase function and altered baseline oxidative status .

Chemotaxis Impairment Neutrophils in ME/CFS demonstrate reduced migration toward chemoattractants, with impaired directional sensing and decreased expression of chemokine receptors .

Neutrophil Extracellular Traps (NETs) and NETosis Dysregulation

NETs are web-like chromatin structures extruded by neutrophils to trap and kill pathogens; their formation is termed NETosis. While NETs serve an antimicrobial function, excessive or dysregulated NETosis drives thrombo-inflammation, coagulopathy, and autoimmunity. ME/CFS patients show altered NET formation, which may contribute to the chronic inflammatory state and autoimmune features of the condition .

Recent evidence from COVID-19 and long COVID research reveals a specific dysregulation: impaired NET degradation rather than (or in addition to) excessive NET formation. Garcia et al. (Garcia et al. 2024) demonstrated that in severe and critical COVID-19, functional DNase levels—the enzymatic activity that degrades NETs—were significantly diminished compared to ambulatory patients (\(n=145\), multicentre French cohort), while NET markers (MPO-DNA, H3cit, H3cit-DNA complexes) rose proportionally with disease severity. The NET/DNase ratio was markedly elevated in the most severe patients and correlated with CRP and neutrophil/lymphocyte ratio—indicating that the imbalance itself, rather than NET production alone, drives clinical deterioration. This imbalance has two upstream mechanisms: (1) reduced plasmacytoid dendritic cell (pDC) counts in critical patients, which impair production of DNase1L3 (the primary serum NET-degrading enzyme expressed predominantly by pDCs), with single-cell RNAseq confirming reduced DNASE1L3 expression per pDC with increasing severity; and (2) genetic polymorphisms—nine linked DNASE1 variants associated with ~75% reduction in DNase1 antigen in carriers, all three of whom were critical patients—that constitutionally limit DNase production, perpetuating NET accumulation (Garcia et al. 2024). The consequence is persistent NET-mediated thrombo-inflammation: in vitro, sera from severe COVID-19 patients induced elevated NET formation in control neutrophils, positively associated with IL-1\(\beta\), IFN-\(\gamma\), and IL-6 (Romano et al. 2022). Remarkably, breakthrough infections (in vaccinated individuals) also showed elevated NETosis despite mild clinical disease, indicating “dysregulated immune response decoupled from pathogen load” (Romano et al. 2022)—a pattern that parallels the immune activation seen in ME/CFS despite minimal ongoing infection. Certainty: 0.70 (multicentre cohort, robust mechanistic design; not yet ME/CFS-specific).

The thrombo-inflammatory consequences are substantial: NET-elevated patients showed elevated D-dimer (coagulation activation marker; mean 503 ng/mL in severe cases) and lactate dehydrogenase, indicating tissue damage and microthrombi formation (Romano et al. 2022). This NET-driven coagulation dysregulation may explain ME/CFS features such as orthostatic intolerance (if microthrombi reduce cerebral perfusion), post-exertional malaise (reduced microvascular oxygen delivery), and cognitive dysfunction. Furthermore, NETosis-mediated mechanisms persist in long COVID, where substantial organ damage occurs in over 70% of infected individuals, with lasting structural remodeling (e.g., pulmonary scarring, fibrosis) rather than reversible functional impairment (Ewing et al. 2024).

Speculation: Defective NET Clearance Despite Diminished NET Production

Certainty: 0.25. Given the suppressed neutrophil activation signature observed in ME/CFS (discussed above), the dysregulation may be more nuanced: ME/CFS neutrophils may have impaired NETosis formation at baseline (contributing to immune suppression), yet still experience dysregulated NET degradation, allowing accumulated NETs to drive chronic thrombo-inflammation. This speculation—defective NET clearance despite diminished NET production—remains to be directly tested in ME/CFS but is biologically plausible given the innate immune suppression and persistent inflammatory markers characteristic of the disease.

Testable predictions: (a) ME/CFS patients show elevated circulating NET remnants (MPO-DNA complexes, citrullinated histone H3) despite reduced ex vivo NETosis induction; (b) serum DNase1L3 activity is reduced in ME/CFS patients relative to healthy controls; (c) NET remnant levels correlate with thrombo-inflammatory markers (D-dimer, fibrinogen) and orthostatic intolerance severity.

Achievement: Persistent NETosis Induction in Long COVID as Potential Pathogenic Driver

Krinsky et al. (Krinsky et al. 2023) measured NETosis induction capacity across 177 COVID-19 patients at two Israeli centres spanning acute and convalescent timepoints. NETosis induction correlated strongly with disease severity, platelet activation, and coagulation factor elevation, and remained significantly elevated in long COVID patients months after acute resolution. Critically, NETosis induction was more sensitive than MPO-DNA for stratifying disease severity, and dexamethasone treatment reduced induction. Persistent NETosis capacity in long COVID suggests ongoing neutrophil priming as a mechanistic contributor to post-acute pathology rather than a transient inflammatory response. Study: (\(n=177\) + 54 controls, two-centre cohort, Journal of Thrombosis and Haemostasis, certainty: 0.75, partially replicated).

Neutrophil extracellular trap formation is directly triggered by viral infections through both pattern-recognition receptor activation and cytokine-mediated pathways (Schönrich and Raftery 2016), providing a mechanistic bridge between acute viral illness and the sustained neutrophil dysfunction observed in ME/CFS (Romano et al. 2022) (Krinsky et al. 2023) (Monsalve et al. 2025). Notably, herpesviruses implicated in ME/CFS triggering (EBV, CMV) are among the established viral NETosis inducers (Schönrich and Raftery 2016). A recent comprehensive review confirms that NET dysregulation is not COVID-19-specific but a general post-viral mechanism, with conserved viral strategies to evade NET degradation across multiple viral families (Asaba et al. 2026), supporting extrapolation to ME/CFS regardless of triggering pathogen.

Speculation: NET-Microclot Structural Association as Persistence Mechanism

NETs are not merely soluble thrombo-inflammatory mediators—they are physically incorporated into circulating microclots. Thierry et al. (Thierry et al. 2025) demonstrated that NET markers (myeloperoxidase, neutrophil elastase, circulating DNA) are quantitatively and structurally associated with microclots in long COVID patients, with strong diagnostic performance for discriminating patients from controls. NETs appear to be a structural component promoting microclot stabilization, providing a mechanistic bridge between thrombo-inflammation and persistent microvascular obstruction.

Certainty: 0.30 (ME/CFS extrapolation). Microclots have been reported in ME/CFS (Chapter Cardiovascular Dysfunction) but their structural composition (including NET content) has not been directly examined. If NETs stabilize microclots, DNase I therapy could degrade the NET scaffold and destabilize microclots—a mechanism distinct from anticoagulation that would not increase bleeding risk.

Testable prediction: ME/CFS microclots should contain MPO and NE by immunofluorescence; DNase I treatment should reduce microclot stability in vitro.

Hypothesis: NETosis as Bridge from Acute Viral Infection to Chronic Autoimmunity

Monsalve et al. (Monsalve et al. 2025) propose that excessive NETosis following SARS-CoV-2 infection generates persistent autoantigenic stimuli—citrullinated histones, extracellular DNA, and NET-associated proteins—that activate NLRP3 inflammasome pathways and sustain IL-1\(\beta\)/IL-18 release characteristic of long COVID. Under this model, impaired NET clearance (reduced DNase activity or persistent neutrophil priming) creates a self-amplifying loop linking the acute viral trigger to chronic autoimmune-like pathology indistinguishable from primary autoimmune disease.

Testable prediction: Long COVID patients with persistent NETosis should show higher NLRP3 activation markers (IL-1\(\beta\), IL-18, caspase-1) and higher citrullinated autoantibody titres than those who resolved NET formation normally.

Treatment implication: PAD4 inhibitors (reducing NET citrullination) and recombinant DNase I (enhancing NET clearance) represent mechanistically grounded therapeutic targets that could be tested in ME/CFS NETosis-positive subgroups. Study: (systematic review, certainty: 0.70, synthesizes findings across multiple groups).

Speculation: Insulin Resistance as a Metabolic Driver of Persistent NETosis

Sanhueza et al. (Sanhueza et al. 2026) demonstrated that 24 of 36 (67%) patients without pre-existing glucose disorders developed de novo insulin resistance at 4 months post-COVID. Insulin-resistant patients exhibited increased basal NETosis, and their plasma significantly enhanced NETosis in control neutrophils in vitro. Insulin enhanced NETosis independently of glucose concentrations, establishing a direct metabolic-immune mechanism: post-infectious insulin resistance drives neutrophil metabolic reprogramming toward sustained NET production through glycolysis-dependent pathways.

Certainty: 0.35 (ME/CFS extrapolation). ME/CFS patients have documented metabolic disturbances including impaired glucose metabolism and mitochondrial dysfunction (Chapter Energy Metabolism and Mitochondrial Function). Post-infectious insulin resistance could create a self-reinforcing loop: metabolic dysfunction → enhanced NETosis → thrombo-inflammation → microvascular impairment → worsened tissue hypoxia → deeper metabolic dysfunction. This connects the energy metabolism and immune dysregulation domains via a testable pathway that has not been examined in ME/CFS.

Testable prediction: ME/CFS patients should show insulin resistance correlating with NET remnant levels; metformin or insulin-sensitising interventions should reduce NET markers. Study: (\(n=60\), prospective cohort, Frontiers in Immunology, certainty: 0.60, not yet independently replicated).

Speculation: pDC Migration Hijack: Blood-to-Tissue Sequestration Model

A critical unresolved tension exists in the pDC/NET literature: Garcia et al. (Garcia et al. 2024) showed pDC depletion in acute severe COVID-19 blood associated with reduced DNase1L3, while Augustin et al. (Augustin et al. 2026) demonstrated pDC expansion in terminal ileum tissue of patients with post-COVID syndrome (PCS), alongside elevated SARS-CoV-2 nucleocapsid protein and gut barrier dysfunction.

A comprehensive review of 23 COVID-19 studies found that 19 documented significant blood pDC depletion, with 10 of 13 reporting depletion proportional to disease severity—a pattern termed the “pDC desert” (Van der Sluis, Holm, and Jakobsen 2022). This blood depletion pattern is not COVID-19-specific: it is documented across HIV, HCV, HBV, and autoimmune diseases (SLE, RA, psoriasis), each showing the same blood↓/tissue↑ pattern driven by chemokine gradients (CXCR3/CXCL9-11, CXCR4/CXCL12) (Li et al. 2017). Critically, Pérez-Gómez et al. (Pérez-Gómez et al. 2021) demonstrated that blood pDC deficiency persists at least 7 months post-SARS-CoV-2 in both hospitalised and non-hospitalised patients—indicating this is not a transient acute-phase phenomenon.

The blood→tissue migration hijack is one resolution model: pDCs are recruited from blood to tissue sites of viral persistence/inflammation via CXCR3/CXCL9-11, producing tissue-level IFN-α and inflammation while depleting circulating pDC pools. Alternative explanations — true pDC apoptosis/necrosis, bone marrow precursor suppression, passive redistribution without chemokine mediation — have not been excluded by direct evidence. If the migration model is correct, this creates a dual deficit: (1) systemic DNase1L3 deficiency (reduced circulating pDCs → impaired NET degradation), and (2) tissue-level IFN-α-driven inflammation (gut, potentially CNS). SARS-CoV-2 may accelerate this process via NRP1/CD304-mediated suppression of pDC IFN production (Van der Sluis, Holm, and Jakobsen 2022), impairing antiviral defense while pDCs are still numerically present.

Certainty: 0.35. The migration model is plausible and mechanistically grounded in autoimmune disease literature (SLE, RA), but chemokine profiling and tissue pDC functional assays have not been performed in ME/CFS or Long COVID. The model is consistent with COVID-19 blood depletion data but has not been experimentally validated in post-viral syndromes.

Testable prediction: ME/CFS and Long COVID patients should show elevated serum CXCL9/10/11, decreased blood pDCs (flow cytometry), and if tissue-accessible, increased gut-mucosal pDCs. pDC chemokine receptor CXCR3 expression on circulating pDCs should correlate inversely with blood pDC count.

Open Question: Can pDC Functional Assays Distinguish Migration from Depletion in Post-Viral ME/CFS?

The blood pDC depletion documented across viral infections and autoimmune diseases could reflect true cell death (apoptosis/necrosis) or chemokine-driven migration to tissues. Distinguishing these requires functional assays beyond enumeration: (1) pDC chemokine receptor profiling (CXCR3, CXCR4, CCR7) to test whether residual blood pDCs have downregulated migration receptors; (2) serum CXCL9/10/11 measurement to assess chemokine gradients driving migration; (3) paired blood and tissue pDC enumeration where accessible (gut biopsies); and (4) functional pDC output assays — IFN-α production capacity (TLR7 agonist stimulation) and DNase1L3 protein release per pDC. If residual blood pDCs retain normal CXCR3 and normal IFN-α/DNase1L3 output per cell, the deficit is migration-driven (cells are present but sequestered). If they show high CXCR3, low IFN-α/DNase1L3 output, the deficit involves functional exhaustion on top of migration. Either scenario has different therapeutic implications: migration blockade (CXCR3 antagonists) vs. pDC restimulation (TLR7 agonists, Flt3L).

No pDC enumeration, chemokine profiling, or functional assay has been performed in ME/CFS. The migration hijack model and persistent depletion timeline are established in COVID-19 and autoimmune disease literature but remain extrapolation to ME/CFS (Van der Sluis, Holm, and Jakobsen 2022) (Li et al. 2017) (Pérez-Gómez et al. 2021).

Hypothesis: Early Innate Immune Dysregulation Predicts Post-Viral Outcomes

Perez Mazzali et al. (Perez Mazzali et al. 2026) used mass cytometry (40-marker panel) to longitudinally profile immune cells from acute COVID-19 to 3-month follow-up, demonstrating that early abnormalities in innate immunity—including dendritic cell subset alterations—predicted which patients developed long COVID. If dendritic cell dysfunction (including pDC deficits) is an early predictor of post-acute outcomes, then DNase1L3 deficiency may be an early feature rather than a late consequence of severe disease—a hypothesis that changes the temporal framing of the NET/DNase imbalance from “damage accumulates over time” to “clearance failure is established early and perpetuates damage.” However, pDC-specific findings were not detailed in the paper, direct DNase measurements were not performed, and assessment is based on abstract-level information — the early-immune-prediction claim should be treated as provisional pending full-text verification.

Certainty: 0.35. pDC-specific findings not detailed in this study; direct DNase measurements were not performed; abstract-level assessment.

Testable prediction: Early pDC count and serum DNase activity at acute COVID-19 presentation should predict long COVID development at 6-month follow-up. Study: (prospective longitudinal, Journal of Infection, certainty: 0.55).

Aging-associated contraction of the naive lymphocyte pool specifically impairs SARS-CoV-2-specific CD8⁺ T cell induction in severe COVID-19, while antibody responses remain intact (Autaa et al. 2025). Critically, IL-18—elevated in severe disease and implicated in the NETosis–NLRP3 loop identified by Monsalve et al. above—was identified as a direct suppressor of antigen-specific CD8⁺ T cell induction and memory recall. This IL-18-mediated CD8 suppression could impair viral reservoir clearance in older post-COVID ME/CFS patients, linking immunosenescence to the failure to eliminate SARS-CoV-2 tissue persistence.

Preclinical evidence supports NET disruption as a viable therapeutic strategy: DNase I treatment reduced multi-organ injury in SARS-CoV-2-infected mice (Veras et al. 2023), with NETs shown to directly damage lung epithelium independently of other viral effects, suggesting that pharmacological NET clearance could be protective across multiple organ systems in post-viral inflammatory states.

NETosis-mediated thromboinflammation has been documented across diverse immune stimuli: in vaccine-associated thrombotic complications (VITT), H3-NET levels and calprotectin correlated strongly with clinical severity (\(r \geq 0.745\)) (Hetland et al. 2022), establishing calprotectin (S100A8/S100A9) as a clinically accessible proxy for NET burden that could be applied in ME/CFS monitoring protocols. Nunes et al. (2024) subsequently confirmed S100-A9 upregulation directly in ME/CFS plasma using data-independent LC-MS/MS proteomics, providing the first direct proteomic evidence of S100A8/A9 dysregulation in the condition (Nunes et al. 2024).

Speculation: HMGB1-S100A8/A9 Double-DAMP Synergy at TLR4/RAGE

Certainty: 0.35. HMGB1 and S100A8/A9 (calprotectin), both elevated in ME/CFS, may form heterocomplexes in the extracellular space that produce synergistic TLR4/RAGE activation, amplifying inflammatory signaling beyond what either DAMP alone would achieve (Kim et al. 2018) (Yang et al. 2013). HMGB1 exists in three redox forms — fully reduced (FR-HMGB1, chemokine activity via CXCR4), disulfide (DS-HMGB1, cytokine activity via TLR4), and sulfonyl (oxidized, inactive). S100A8/A9 preferentially binds TLR4 and RAGE, but its affinity and downstream signaling may be modified by HMGB1 complex formation.

Mechanism. FR-HMGB1 binds CXCR4 for chemotactic activity; upon oxidation to DS-HMGB1 it switches to TLR4 binding. S100A8/A9 signals through TLR4-MyD88 and RAGE, activating NF-\(\kappa\)B and MAPK pathways. If HMGB1 and S100A8/A9 form mixed heterocomplexes, the complex may engage both TLR4 and RAGE simultaneously, producing signal integration at the receptor level. This could: (1) lower the activation threshold for each DAMP individually; (2) prolong signaling duration through delayed complex dissociation; (3) recruit distinct adaptor pools (MyD88 vs TRIF) that normally are not co-engaged by either DAMP alone. The combined engagement of TLR4+RAGE may drive a qualitatively different cytokine output (e.g., IL-1\(\beta\) + IL-18 dominance) than either DAMP alone (IL-6, TNF-\(\alpha\)).

ME/CFS implication. If double-DAMP synergy operates in ME/CFS, the inflammatory threshold would be lowered even when both DAMPs are individually below their nominal activation thresholds. This would explain why ME/CFS patients mount disproportionate inflammatory responses to minor triggers — the DAMP synergy effectively amplifies weak signals. The concept also predicts a non-linear dose-response relationship between DAMP concentrations and symptom severity: small DAMP increments could produce disproportionately large flare-ups once the synergy threshold is crossed.

Falsifiable predictions. (1) Recombinant HMGB1 + S100A8/A9 co-incubation will produce synergistic IL-1\(\beta\) and IL-18 release from THP-1 monocytes (Synergy Score >1; Bliss independence test). (2) HMGB1-S100A8/A9 heterocomplexes will be detectable in ME/CFS plasma by co-immunoprecipitation (anti-HMGB1 pull-down, anti-S100A9 immunoblot). (3) ME/CFS plasma will show higher synergy-level cytokine induction on reporter cells than additive predictions from individual DAMP concentrations. (4) Blocking either HMGB1 (glycyrrhizin, mAb) or S100A9 (tasquinimod) will reduce the synergy more than blocking the receptor (TLR4 + RAGE antagonists), because the synergy depends on complex integrity rather than receptor-level convergence.

Limitations. HMGB1-S100A8/A9 heterocomplexes have not been demonstrated in any human disease — their existence in ME/CFS plasma is inferred from in vitro binding assays. The null hypothesis — that HMGB1 and S100A8/A9 act additively rather than synergistically, each activating TLR4/RAGE independently — is equally consistent with simple co-elevation and cannot be ruled out without direct detection of heterocomplexes. HMGB1 redox status in ME/CFS plasma is uncharacterized; the three redox forms have dramatically different activities. S100A8/A9 heterocomplexes with other proteins (S100A12, RAGE itself) may produce equally complex interactions not captured by the simple binary model. The synergy prediction assumes DAMP concentrations reach the tissue compartion required for complex formation, which may not occur in ME/CFS.

Proteomic Evidence: Suppressed Neutrophil Signature A large-scale serum proteomics study by Hoel et al. (2026) (Hoel et al. 2026) revealed a broad and pronounced reduction in circulating neutrophil-derived proteins in ME/CFS patients. By cross-referencing their dataset with a published list of proteins released by PMA-activated neutrophils, the authors found that 146 of 254 known neutrophil activation products were detectable, and over 85% of those showing significant differences were reduced in ME/CFS. Hallmark neutrophil granule components—bactericidal/permeability-increasing protein (BPI), peptidylarginine deiminase 4 (PADI4), matrix metalloproteinase 9 (MMP9), neutrophil elastase (ELANE), azurocidin 1 (AZU1), and lactoferrin (LTF)—were all significantly lower. This suppressed neutrophil protein signature occurred despite normal peripheral neutrophil counts (3.38 \(\pm\) 1.03 \(\\times\) 10⁹/L), suggesting a qualitative deficiency in neutrophil activation or degranulation rather than a quantitative cell deficit. This proteomic finding complements the functional neutrophil abnormalities described above (impaired phagocytosis, respiratory burst defects, chemotaxis impairment) and suggests that the circulating neutrophil pool in ME/CFS is hypoactivated at baseline—potentially contributing to the impaired innate immune defense and altered inflammatory milieu characteristic of the disease.

More broadly, the Hoel et al. study found that the innate immune arm—granulocyte- and monocyte-associated proteins—showed a primarily downward shift (70.3% and 64.6% of altered proteins reduced, respectively), while NK cell and T cell-associated proteins were predominantly elevated (72.0% and 62.5% upward). This innate-suppressed / adaptive-elevated pattern suggests a fundamental immune reprogramming in ME/CFS, where innate first-responder functions are dampened while adaptive immune cells show signs of chronic activation or compensatory expansion.

3.2 Monocyte and Macrophage Dysfunction

Monocytes and their tissue-resident derivatives (macrophages) bridge innate and adaptive immunity:

Monocyte Subset Alterations Human monocytes are classified into three functionally distinct subsets: classical (CD14⁺⁺CD16^-) monocytes perform phagocytic and antimicrobial functions; intermediate (CD14⁺⁺CD16⁺) monocytes handle antigen presentation and cytokine production; and non-classical (CD14⁺CD16⁺⁺) monocytes conduct patrolling and vascular surveillance. ME/CFS studies have found increased intermediate monocytes (associated with inflammation), altered cytokine production profiles, abnormal responses to stimulation, and changed expression of activation markers.

The largest comparative immunophenotyping study to date (n=207 total; 103 ME/CFS, 63 long COVID, 41 healthy controls — though ‘’largest’’ is relative to a field with few comparative studies) provides a refined picture of monocyte dysfunction in ME/CFS (Petrov et al. 2026). Using 12-parameter flow cytometry, Petrov et al. demonstrated that ME/CFS monocytes exhibit reduced expression of the costimulatory molecule CD80 on M1-like (CD206^low) monocytes — a finding that is paradoxical given M1-polarized cells typically upregulate CD80, suggesting a functional defect within the M1-like subset beyond polarization state. ME/CFS monocytes also showed impaired C-C chemokine receptor type 7 (CCR7) expression, a receptor essential for lymph node homing and immune cell trafficking (though reduced CCR7 could also reflect tissue redistribution after homing, rather than a primary trafficking defect). At the same time, Petrov et al. found T cell activation features in ME/CFS — increased CD8+ effector memory/terminal differentiation (CD45RA+CD62L^-) and elevated CD4+CD95⁺ — which complicate a uniform ‘’immune suppression’’ label. The monocyte/DC compartment therefore appears predominantly suppressively skewed, with concurrent T cell activation, in contrast to the persistent immune activation with exhaustion observed in long COVID (M2-like monocyte polarization, elevated CD80, dendritic cell expansion). The study achieved moderate discrimination between ME/CFS and long COVID using composite immune markers (VIP scores: CD80 on monocytes, CCR7 on dendritic cells, M2-like polarization); AUC was not reported, and the magnitude of ‘’moderate’’ discrimination is therefore unquantified. Certainty: 0.50. Limitations: single-center, cross-sectional, unreplicated; healthy controls significantly younger than patient groups (age-adjusted ANCOVA applied, but residual confounding possible given age-dependent CCR7 expression); excluded common ME/CFS medications (LDN, antihistamines, beta-blockers) — which may bias the cohort toward milder or less-treated patients; vaccination status not analyzed; significant within-group heterogeneity with partial inter-group overlap; post-hoc power (≥ 0.80) is reported in the source paper but conveys no information beyond the observed p-values (Hoenig & Heisey 2001). Certainty reduced from initial 0.70 because the study is unreplicated, medication exclusions limit generalizability, and functional costimulatory capacity was not measured (surface expression ≠ function).

Macrophage Polarization Tissue macrophages can adopt pro-inflammatory (M1) or anti-inflammatory (M2) phenotypes. Evidence suggests M1 polarization in ME/CFS, with impaired transition to the resolving M2 phenotype, resulting in chronic inflammatory macrophage activation (Heng et al. 2025). Note: the Petrov finding of reduced CD80 on M1-like monocytes (above) creates an unresolved tension with M1-polarization claims — M1-polarized cells normally upregulate CD80, so reduced CD80 on M1-like cells may indicate a dysfunctional M1 state (polarized but functionally impaired), an M1 surface marker profile that does not reflect true M1 functional polarization, or divergence between tissue macrophages and circulating monocytes.

4 Complement System

The complement system consists of plasma proteins that enhance (“complement”) antibody and phagocyte function. Abnormalities in ME/CFS include:

4.1 Complement Activation Patterns

ME/CFS patients show elevated activation products, with increased C3a, C4a, and C5a fragments indicating ongoing complement activation . Reduced levels of C3 and C4 suggest consumption of these complement components. Additionally, abnormal levels of complement regulatory proteins point to altered regulation of the system.

Serum proteomics data from Hoel et al. (2026) (Hoel et al. 2026) confirmed and extended these findings: complement factor D (CFD) and complement component C6 were among the most significantly elevated secreted proteins in ME/CFS. Elevated CFD indicates activation of the alternative complement pathway, while elevated C6 indicates increased assembly of the membrane attack complex (MAC). The authors noted that elevated CFD and C6 are implicated in vasculitis, lupus, dermatomyositis, and autoimmune nephritis, further supporting a possible autoimmune-like complement activation pattern in ME/CFS.

4.2 Exercise-Induced Complement Activation and PEM

A series of studies has documented an association between complement activation and post-exertional malaise. Sorensen et al. demonstrated significant C4a elevation at 6 hours post-exercise exclusively in CFS patients (\(P < 0.01\)), regardless of allergy status—the first demonstration of exercise-induced complement activation in CFS (Sorensen et al. 2003). Nijs et al. subsequently identified post-exercise C4a as a clinically important biomarker for PEM, linking complement activation to the molecular mechanism of post-exertional malaise alongside elastase and IL-1\(\beta\) (Nijs et al. 2010). Polli et al. extended this by demonstrating moderate associations between exercise-induced C4a changes and pain thresholds in ME/CFS (\(r=0.669\), \(p=0.001\)), directly connecting complement activation to exercise-induced hyperalgesia (Polli et al. 2019).

Most recently, Glass et al. (2025) found upregulation of complement system proteins in extracellular vesicles post-exercise in males with ME/CFS, with ER stress response changes correlating with PEM severity (Glass et al. 2025). This finding is consistent with the earlier soluble C4a observations and extends them to the vesicular compartment, though the male-only sample limits generalizability.

C4a generation is consistent with activation of the lectin and/or classical complement pathways rather than the alternative pathway. This is mechanistically distinct from the CFD (factor D) elevation reported in serum proteomics (Hoel et al. 2026), which indicates alternative pathway activation. The co-existence of both pathway activations suggests broad complement dysregulation rather than a single pathway defect.

4.3 Clinical Implications

Complement abnormalities may contribute to inflammation through anaphylatoxin (C3a, C5a) production and impair pathogen clearance . They may also promote autoimmune manifestations and trigger mast cell activation through complement fragment-induced degranulation .

Speculation: Complement Dynamics During PEM: Reconciling Sorensen 2003 vs Nunes 2024

Certainty: 0.45. Sorensen et al. measured C4a elevation at 6 h post-exercise in CFS patients (Sorensen et al. 2003), establishing complement activation as an acute PEM-phase event. Nunes et al. found reduced complement component C4a in ME/CFS plasma at rest (not post-exercise) (Nunes et al. 2024). These findings can be reconciled by a biphasic complement model: acute activation (Sorensen: elevated 6 h post-exercise) followed by chronic consumption with supranormal fragments but reduced intact component levels in the resting state (Nunes: low C4a at rest). This pattern — activation peaks during exertion, consumption depletes the reservoir between bouts — is characteristic of a system under chronic intermittent demand.

Time-resolved model. Phase 1 (0–6 h post-exertion): Complement cascade activated via classical (immune complex-driven) and lectin (endothelial glycocalyx damage) pathways → C4a, C3a, C5a spike. Phase 2 (6–48 h): Anaphylatoxins drive mast cell degranulation and vascular permeability; C5b-9 (MAC) deposits on endothelium → microvascular injury. Phase 3 (48 h+ sustained): Complement component consumption (reduced C4, C3, factor B) persists if activation continues, as seen in post-COVID ME/CFS (Klein et al. 2024). If the patient rests fully, complement returns to baseline by approximately 72 h. If repeat exertion occurs before full recovery, activation sums — explaining why repeated activity without recovery produces escalating PEM severity.

Reconciliation mechanism. Sorensen’s C4a spike at 6 h captures Phase 1 (acute activation). Nunes’ low resting C4a captures the inter-exertion trough (depleted reservoir between PEM episodes). No contradiction: they measure different temporal phases. The biphasic model predicts that C4a levels vary dramatically depending on time-since-last-exertion, explaining why single-timepoint resting measures produce inconsistent complement results across ME/CFS studies.

Falsifiable predictions. (1) Serial complement measurements (C4a, C3a, C5b-9) before and after standardized exercise will show a triphasic pattern: baseline → spike (4–6 h) → gradual decline (24–48 h) → possible sustained elevation if recovery insufficient. (2) The magnitude of the 6 h C4a spike will predict next-day symptom severity (PEM intensity) with \(r > 0.5\). (3) Patients with low resting C4a (depleted reservoir) will show the greatest post-exertion spikes, consistent with a system operating near consumption threshold. (4) Complement consumption markers (C3/C4 ratio, C4a/C4 ratio) will correlate with time-since-last-exertion better than absolute C4a levels alone.

Limitations. Direct time-series data in ME/CFS are absent — the model synthesizes Sorensen 2003 (6 h post-exercise) and Nunes 2024 (resting proteomics) with no study bridging the full temporal window. The 72 h recovery estimate is extrapolated from general complement biology (C4 half-life ~1–3 days), not ME/CFS-specific data. The biphasic model predicts high inter-study variability depending on sampling protocols, which would need to be confirmed by prospective time-series studies.

4.4 Complement Dysregulation and Infection Susceptibility

Complement deficiency—particularly of the terminal pathway (C5–C9) and properdin—is a well-established risk factor for invasive meningococcal disease, with a substantial proportion of recurrent meningococcal cases occurring in complement-deficient individuals (reviewed in standard infectious disease references). ME/CFS complement findings show dysregulation (activation with consumption) rather than genetic deficiency per se: reduced C3 and C4 levels suggest consumption of these components, while elevated activation products (C3a, C4a, C5a) indicate ongoing activation . Persistent complement consumption in Long COVID has been independently confirmed (Klein et al. 2024).

Limitation: Complement Dysregulation \(\neq\) Complement Deficiency

The ME/CFS complement phenotype—chronic activation with component consumption—is mechanistically distinct from the genetic complement deficiencies that confer high meningococcal susceptibility. Reduced C3/C4 from consumption may transiently impair opsonization and pathogen clearance, but the terminal pathway (C5–C9 membrane attack complex) critical for killing Neisseria meningitidis has not been specifically assessed in ME/CFS. Whether the exercise-induced C4a elevation (Sorensen et al. 2003) (Glass et al. 2025) reflects a complement system that is functionally compromised—rather than merely activated—for pathogen defense remains an open question. Extrapolating from complement dysregulation to increased meningococcal susceptibility is biologically plausible but empirically unsupported.

Speculation: Complement-Driven Perivascular Mast Cell Amplification Loop

Certainty: 0.25. Based on complement activation evidence (Sorensen, Maya, Bragee) and mast cell biology (established C3a/C5a mechanisms). Downgraded from initial 0.45 because: the loop has never been demonstrated in ME/CFS tissue or any disease; complement and mast cell activation could equally be parallel downstream effects of a shared upstream trigger (e.g., endothelial dysfunction); the Bragee CSF evidence cited has no control group and cannot support between-group claims; multiple complement regulatory proteins (Factor H, Factor I, CD55, CD59) would need to fail simultaneously for sustained amplification, which is not addressed. -genetics Bragée et al. (2026; Sorensen et al. 2003)

The Maya et al. complement pQTL subgroup (high C3/low Bb) may represent individuals predisposed to a perivascular amplification loop: complement activation produces C3a/C5a anaphylatoxins → mast cell degranulation at perivascular sites → histamine, tryptase, and heparin increase vascular permeability → further complement protein leakage into perivascular tissue → alternative pathway amplification on host surfaces → more C3a/C5a → more mast cell activation. However, this loop requires: (1) sustained mast cell responsiveness (tachyphylaxis after repeated anaphylatoxin exposure is not addressed), (2) spatial confinement (no mechanism provided for why amplification would be perivascular rather than systemic), and (3) failure of complement regulatory proteins (Factor H, CD55, CD59—none of which are discussed). The simpler alternative—complement activation and mast cell activation as parallel downstream effects of systemic inflammation or endothelial dysfunction—explains the co-occurrence without requiring a novel positive feedback mechanism.

The Bragée et al. CSF finding of complement cascade enrichment in severe ME/CFS cases Bragée et al. (2026) has been cited as supporting evidence for this loop, but that study had no healthy control group (certainty 0.35) and cannot establish differential expression.

Testable predictions: + Skin blister fluid or interstitial fluid from ME/CFS patients will show C3a:C3 ratio ≥2× in tissue vs plasma, indicating local complement amplification (research technique requiring specialized collection) + Complement-pQTL-positive subgroup will have serum tryptase ≥20% higher and PGD2 ≥30% higher than complement-pQTL-negative patients + C5a receptor antagonists (e.g., avacopan) would reduce mast cell mediators by ≥50% if the loop operates; ≤20% reduction supports the parallel-activation alternative

Open Question: Does NCS-1/InsP3R1 Amplify Mast Cell Degranulation in ME/CFS?

Mast cells express NCS-1 (neuronal calcium sensor protein 1), which amplifies InsP3R1 channel open probability approximately five-fold. In rat basophilic leukemia mast cells, NCS-1 regulates IgE-triggered exocytosis via PI4Kβ, a kinase that generates PIP₂ substrate from PI (Kappel et al. 2003). NCS-1 also stimulates ERK signaling through endocytic recycling in mast cells (Kappel et al. 2006) and co-regulates mast cell function alongside synaptotagmins (Kappel et al. 2007). Lithium disrupts the NCS-1/InsP3R1 interaction at IC₅₀ ~350 µM (Schlecker et al. 2006), raising the possibility that even ultralow concentrations of Li⁺ could dampen mast cell reactivity by reducing IP3-mediated Ca²⁺ release. A critical gap must be acknowledged: the estimated tissue Li⁺ concentration after 2 mg/day oral dosing is likely ~0.2–2 µM, which falls 100–1,000× below the in vitro IC₅₀ for NCS-1/InsP3R1 disruption. Lithium is not known to concentrate preferentially in mast cells. NCS-1/InsP3R1 disruption at ultralow dose is biochemically implausible unless one of the following applies: (a) Li⁺ has a higher-affinity binding mode at a different site on NCS-1 or InsP3R1 that is undetected in the existing binding assay; (b) chronic low-dose exposure produces cumulative effects (e.g., altered NCS-1 expression or post-translational modification) not captured by acute IC₅₀ measurements; or (c) the relevant target is not NCS-1/InsP3R1 but another Li⁺-sensitive enzyme (e.g., GSK-3β, inositol monophosphatase) with different dose-response characteristics. Without evidence for any of these conditions, the concentration gap undermines the hypothesis.

The mechanistic chain linking lithium to mast cell stabilization is assembled from separate literatures and has no direct experimental support: (a) Li⁺ disrupts NCS-1/InsP3R1 — proven in neurons (Schlecker et al. 2006); (b) NCS-1 is expressed in mast cells and regulates degranulation — proven by Kappel 2003; (c) MRGPRX2, C3aR, and C5aR signal through Gαq → PLC → IP3 → Ca²⁺ → degranulation — well-established (reviewed in Roy 2021 (Roy et al. 2021)); (d) Orai/STIM1-dependent SOCE is required for sustained MRGPRX2-mediated Ca²⁺ responses (Chaki et al. 2022) — damping IP3R would reduce both the initial Ca²⁺ spike and the sustained SOCE. The connection between (a) and (b) is entirely unattested — no study has tested lithium’s effect on mast cell degranulation at any concentration. One 1978 study found LiCl did NOT cause mast cell degranulation in rat mesentery (Nahmod et al. 1978), consistent with a stabilizing rather than activating effect, but this is a weak negative control (n small, 1978, indirect).

If the NCS-1/InsP3R1 → mast cell degranulation link operates in ME/CFS, ultra-low-dose lithium (~2 mg/day elemental Li⁺) could act as a functional mast cell stabilizer via a mechanism distinct from existing agents: cromolyn (mechanism incompletely understood; calcium influx inhibition), ketotifen (H1 antagonist + mast cell stabilization; mechanism unclear), and omalizumab (IgE sequestration). Lithium would target the GPCR→PLC→IP3→Ca²⁺ arm of mast cell activation — particularly relevant if GPCR autoantibodies (anti-M3, anti-β2AR) chronically engage this pathway in mast cells, lowering the degranulation threshold. Notably, both FcεRI (ITAM→Syk→PLCγ) and GPCR (Gαq→PLCβ) pathways converge on IP₃→InsP₃R₁→Ca²⁺ mobilization, so NCS-1 amplification via PI4Kβ — which supplies the PIP₂ substrate for both PLC isoforms — could theoretically influence either route. However, the relative dependence on NCS-1 may differ: the Gαq-PLCβ pathway may rely more heavily on PI4Kβ-mediated PIP₂ resynthesis because PLCβ is more sensitive to PIP₂ availability than PLCγ, which can be recruited to membrane phosphotyrosine clusters. Consequently, lithium should reduce IgE-mediated degranulation less than MRGPRX2-mediated degranulation, rather than leaving it unaffected. This partial-effect prediction is more defensible than a binary affected/unaffected claim and can be resolved experimentally by comparing the dose-response curves for both stimuli. Whether lithium ion concentrations reach mast cells in tissue at sufficient levels after 2 mg/day oral dosing is completely unknown.

Clinical context and risk. Currently zero clinical data exist for lithium as a mast cell stabilizer in any condition. Even at 2 mg/day, lithium carries risks including thyroid suppression (goiter, hypothyroidism, requiring TSH and fT3 monitoring), renal clearance variability (requiring serum creatinine monitoring), and potential for toxicity if clearance is impaired — standard lithium monitoring (TSH, fT3, creatinine, serum lithium at steady state) applies. Established mast cell stabilizers with extensive safety data (cromolyn sodium, ketotifen) should be trialed first before considering lithium for this indication. The NCS-1 hypothesis provides a mechanistic rationale for a trial but does not constitute a treatment recommendation.

Testable predictions. (1) NCS-1 expression in ME/CFS-derived mast cells (cultured from peripheral blood CD34⁺ progenitors) will be at least 2-fold elevated vs healthy controls by qPCR (at least 1.5-fold by Western blot). Falsified if NCS-1 is not elevated. (2) Lithium at 1-10 microM pre-incubation reduces beta-hexosaminidase release from patient-derived mast cells stimulated with MRGPRX2 agonists (compound 48/80, substance P) by 30% or more. Falsified if lithium has no effect or increases release. (3) Lithium effect is graded by pathway dependence on NCS-1: GPCR-mediated degranulation (MRGPRX2 agonists compound 48/80, substance P) is reduced by ≥30%; IgE-mediated degranulation (anti-DNP-IgE + DNP-BSA) is reduced by ≥10% but ≤25% (partial effect). Falsified if the two pathways show identical reduction magnitude or if IgE-mediated reduction exceeds GPCR-mediated reduction — either outcome would indicate a NCS-1-independent mechanism. (4) In MCAS-positive ME/CFS patients, serum tryptase decreases by 20% or more after 4 weeks of lithium 2 mg/day. Falsified if tryptase unchanged. If baseline tryptase is normal (less than 11.5 ng/mL), substitute urine 11beta-PGF2alpha or serum PGD2 (25% or more reduction predicted). Falsified if neither declines. (5) PBMC lithium concentration after 2 mg/day oral dosing reaches 1 microM or more free Li⁺ by ICP-MS, confirming tissue penetration to the in vitro-effective range. Falsified if steady-state PBMC Li⁺ is less than 0.5 microM.

Dendritic cells (DCs) are professional antigen-presenting cells that initiate adaptive immune responses. ME/CFS patients show altered DC maturation with abnormal expression of co-stimulatory molecules . Changed cytokine production skews toward pro-inflammatory profiles, while impaired antigen presentation may contribute to inadequate pathogen clearance. Plasmacytoid DCs display abnormalities in type I interferon production .

Petrov et al. found that ME/CFS patients show reduced dendritic cell frequency and impaired CCR7 expression on dendritic cells compared to healthy controls, in contrast to long COVID patients who showed dendritic cell expansion with reduced HLA-DR expression (consistent with immune exhaustion) (Petrov et al. 2026). This divergence — DC contraction in ME/CFS, DC expansion with exhaustion in long COVID — may reflect distinct pathophysiological processes, though it could equally represent different disease stages (years of illness in ME/CFS vs months in long COVID) rather than fundamentally distinct mechanisms. Reduced peripheral DC frequency could also reflect tissue redistribution rather than systemic depletion. Certainty: 0.50. As with the monocyte findings above, the T cell activation features from the same study caution against a uniform ‘’immune suppression’’ characterization of the DC compartment.

References

Agrati, C., A. Sacchi, V. Bordoni, E. Cimini, S. Notari, D. Mitarotonda, C. Montesano, G. Ippolito, A. Antinori, and C. Castilletti. 2020. “Expansion of Myeloid-Derived Suppressor Cells in COVID-19 Patients.” Frontiers in Immunology 11: 583574. https://doi.org/10.3389/fimmu.2020.583574.

Asaba, Chinaza N., Tingyu Yan, Shangkun Jin, et al. 2026. “Neutrophil Extracellular Traps in Viral Infections: Regulation, Immune Consequences, and Pathogenic Outcomes.” Cells 15.

Augustin, Max, Lea Picard, Dominic Rauschning, et al. 2026. “Persistent Gut-Immune Axis Dysregulation in Long-Term Post-COVID Syndrome: Insights from a Prospective, Observational, Cross-Sectional Case-Control Study.” Mucosal Immunology. https://doi.org/10.1016/j.mucimm.2026.03.002.

Autaa, Gaëlle, Laura Papagno, Takuto Nogimori, Andrea Boizard-Moracchini, Daniil Korenkov, Maeva Roy, Koichiro Suzuki, et al. 2025. “Aging and Inflammation Limit the Induction of SARS-CoV-2-Specific CD8\(^+\) T Cell Responses in Severe COVID-19.” JCI Insight 10 (4): e180867. https://doi.org/10.1172/jci.insight.180867.

Bragée, B, P Li, D Meadows, A Widgren, P Sjögren, PH Ghatan, BC Bertilson, W Xiao, and J Bergquist. 2026. “Proteomic Signatures in Cerebrospinal Fluid and Their Clinical Associations in Patients with ME/CFS.” Scientific Reports. https://doi.org/10.1038/s41598-026-46965-1.

Brenu, Ekua W., Mieke L. van Driel, Don R. Staines, Kevin J. Ashton, Sandra B. Ramos, James Keane, Nancy G. Klimas, and Sonya M. Marshall-Gradisnik. 2011. “Immunological Abnormalities as Potential Biomarkers in Chronic Fatigue Syndrome/Myalgic Encephalomyelitis.” Journal of Translational Medicine 9: 81. https://doi.org/10.1186/1479-5876-9-81.

Briese, Thomas, Rafal Tokarz, Lucinda Bateman, Xiaoyu Che, Cheng Guo, Komal Jain, Vishal Kapoor, et al. 2023. “A Multicenter Virome Analysis of Blood, Feces, and Saliva in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome.” Journal of Medical Virology 95 (8): e28993. https://doi.org/10.1002/jmv.28993.

Chaki, Shaswati, Ibrahim Alkanfari, Saptarshi Roy, Aetas Amponnawarat, Yvonne Hui, Carole A Oskeritzian, and Hydar Ali. 2022. “Inhibition of Orai Channel Function Regulates Mas-Related G Protein-Coupled Receptor-Mediated Responses in Mast Cells.” Frontiers in Immunology 12: 803335. https://doi.org/10.3389/fimmu.2021.803335.

Ewing, Andrew G., Spela Salamon, Etheresia Pretorius, David Joffe, Greta Fox, Stephane Bilodeau, and Yaneer Bar-Yam. 2024. “Review of Organ Damage from COVID and Long COVID: A Disease with a Spectrum of Pathology.” Cardiovascular Pathology 71: 107631. https://doi.org/10.1016/j.carpath.2024.107631.

Fluder, Natalia, Morgane Humbel, Emeline Recazens, Alexis A Jourdain, Camillo Ribi, et al. 2026. “Mitochondrial Dysfunction Drives Natural Killer Cell Dysfunction in Systemic Lupus Erythematosus.” JCI Insight. https://doi.org/10.1172/jci.insight.195170.

Garcia, Geoffrey, Sylvie Labrouche-Colomer, Alexandre Duvignaud, et al. 2024. “Impaired Balance Between Neutrophil Extracellular Trap Formation and Degradation by DNases in COVID-19 Disease.” Journal of Translational Medicine 22: 246. https://doi.org/10.1186/s12967-024-05044-7.

Glass, Katherine A., Ludovic Giloteaux, Sheng Zhang, and Maureen R. Hanson. 2025. “Extracellular Vesicle Proteomics Uncovers Energy Metabolism, Complement System, and ER Stress Response Dysregulation Postexercise in Males with ME/CFS.” Clinical and Translational Medicine 15 (5): e70346. https://doi.org/10.1002/ctm2.70346.

Heng, Ruiwen Benjamin, Bavani Gunasegaran, Shivani Krishnamurthy, Sonia Bustamante, Ananda Staats, Sharron Chow, Seong Beom Ahn, et al. 2025. “Mapping the Complexity of ME/CFS: Evidence for Abnormal Energy Metabolism, Altered Immune Profile, and Vascular Dysfunction.” Cell Reports Medicine 6 (12): 102514. https://doi.org/10.1016/j.xcrm.2025.102514.

Hetland, Geir, Magne Kristoffer Fagerhol, Markus Karl Hermann Wiedmann, et al. 2022. “Elevated NETs and Calprotectin Levels After ChAdOx1 nCoV-19 Vaccination Correlate with the Severity of Side Effects.” Vaccines 10 (8): 1267. https://doi.org/10.3390/vaccines10081267.

Hoel, August, Fredrik Hoel, Sissel Elisabeth Dyrstad, Henrique Chapola, Ingrid Gurvin Rekeland, Kristin Risa, Kine Alme, et al. 2026. “Charting the Circulating Proteome in ME/CFS Using Cross-System Profiling to Uncover Mechanistic Insights.” Cell Reports Medicine 7 (March): 102647. https://doi.org/10.1016/j.xcrm.2026.102647.

Kappel, S, D H Bhatt, C Letts, M Bissonnette, T W Abrams, A Helli, S Laporte, et al. 2003. “Neuronal Calcium Sensor-1 and Phosphatidylinositol 4-Kinase Beta Regulate IgE Receptor-Triggered Exocytosis in Cultured Mast Cells.” Journal of Cell Science 116 (Pt 20): 4151–62. https://doi.org/10.1242/jcs.00716.

———, et al. 2006. “Neuronal Calcium Sensor-1 and Phosphatidylinositol 4-Kinase Beta Stimulate Extracellular Signal-Regulated Kinase 1/2 Signaling by Accelerating Recycling Through the Endocytic Recycling Compartment.” Molecular Biology of the Cell 17 (8): 3579–91. https://doi.org/10.1091/mbc.e06-01-0040.

Kappel, S, R Bissonnette, A Helli, S A Laporte, and D H Bhatt. 2007. “The Mast Cell: Where Endocytosis and Regulated Exocytosis Meet.” In Molecular Mechanisms of Exocytosis, 237–50. Springer. https://doi.org/10.1007/978-0-387-69086-7_19.

Kennedy, G., V. Spence, C. Underwood, and J. J. F. Belch. 2004. “Increased Neutrophil Apoptosis in Chronic Fatigue Syndrome.” Journal of Clinical Pathology 57 (8): 891–93. https://doi.org/10.1136/jcp.2003.015511.

Kim, H. et al. 2018. “S100A8/TLR4 Signaling.” Scientific Reports.

Klein, Jon et al. 2024. “Persistent Complement Dysregulation in Long COVID.” Science.

Krinsky, Nitzan, Shiran Levy, Yaara Zisman-Ilani, Michal Mandelboim, Amos Nili, et al. 2023. “NETosis Induction Reflects COVID-19 Severity and Long COVID: Insights from a 2-Center Patient Cohort Study in Israel.” Journal of Thrombosis and Haemostasis 21 (9): 2561–75. https://doi.org/10.1016/j.jtha.2023.02.033.

Li, Shuang, Jing Wu, Shan Zhu, Yong-Jun Liu, and Jingtao Chen. 2017. “Disease-Associated Plasmacytoid Dendritic Cells.” Frontiers in Immunology 8: 1268. https://doi.org/10.3389/fimmu.2017.01268.

Maher, K. J., N. G. Klimas, and M. A. Fletcher. 2005. “Chronic Fatigue Syndrome Is Associated with Diminished Intracellular Perforin.” Clinical and Experimental Immunology 142 (3): 505–11. https://doi.org/10.1111/j.1365-2249.2005.02935.x.

Mihatsch, M., K. Schmidt, A. Linnemann, M. Riecker, N. Bannert, C. Kerkau, and S. Rieg. 2026. “Myeloid-Derived Suppressor Cell Expansion During Acute Epstein-Barr Virus Infectious Mononucleosis.” Journal of Virology 96 (5): e01934–21. https://doi.org/10.1128/JVI.01934-21.

Monsalve, Diana M., Yeny Acosta-Ampudia, Nicolás Guerrero Acosta, Mariana Celis-Andrada, Ali Şahin, Ahsen Morva Yilmaz, Yehuda Shoenfeld, and Carolina Ramírez-Santana. 2025. “NETosis: A Key Player in Autoimmunity, COVID-19, and Long COVID.” Journal of Translational Autoimmunity 10: 100280. https://doi.org/10.1016/j.jtauto.2025.100280.

Nahmod, V E, J C Fasciolo, S Carrizo, and J Moguilevsky. 1978. “Mesenteric Mast Cell Degranulation Is Not Essential for Conditioned Taste Aversion.” Pharmacology Biochemistry and Behavior 9 (5): 633–37. https://doi.org/10.1016/0091-3057(78)90215-8.

Nijs, Jo, Jessica Van Oosterwijck, Mira Meeus, et al. 2010. “Unravelling the Nature of Postexertional Malaise in ME/CFS: The Role of Elastase, Complement C4a and Interleukin-1\(\beta\).” Journal of Internal Medicine 267 (4): 418–35. https://doi.org/10.1111/j.1365-2796.2009.02178.x.

Nunes, M., M. Vlok, A. Proal, D. B. Kell, and E. Pretorius. 2024. “Data-Independent LC-MS/MS Analysis of ME/CFS Plasma Reveals a Dysregulated Coagulation System, Endothelial Dysfunction, Downregulation of Complement Machinery.” Cardiovascular Diabetology 23: 260. https://doi.org/10.1186/s12933-024-02315-x.

Perez Mazzali, Marina, Francisco Pérez-Cózar, Paloma Cal-Sabater, et al. 2026. “Persistent t Cell Phenotypic Alterations and Early Innate Immune Dysregulation as Potential Biomarkers of Long COVID.” The Journal of Infection 92: 106731. https://doi.org/10.1016/j.jinf.2026.106731.

Pérez-Gómez, Alberto, Joana Vitallé, Carmen Gasca-Capote, et al. 2021. “Dendritic Cell Deficiencies Persist Seven Months After SARS-CoV-2 Infection.” Cellular & Molecular Immunology 18: 2128–39. https://doi.org/10.1038/s41423-021-00728-2.

Petrov, Steliyan, Martina Bozhkova, Mariya Ivanovska, Teodora Kalfova, Dobrina Dudova, Yana Todorova, Radostina Dimitrova, et al. 2026. “Comprehensive Immunophenotyping of Monocytes and Dendritic Cells Suggests Distinct Pathophysiology in Chronic Fatigue Syndrome and Long COVID.” International Journal of Molecular Sciences 27 (10): 4488. https://doi.org/10.3390/ijms27104488.

Polli, Andrea et al. 2019. “Exercise-Induced Hyperalgesia, Complement System and Elastase Activation in ME/CFS.” Scandinavian Journal of Pain 19 (1): 183–92. https://doi.org/10.1515/sjpain-2018-0075.

Querec, Troy D, Jin-Mann S Lin, Yang Chen, Brian Helton, Andreas M Kogelnik, Nancy G Klimas, Daniel L Peterson, et al. 2023. “Natural Killer Cytotoxicity in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS): A Multi-Site Clinical Assessment of ME/CFS (MCAM) Sub-Study.” Journal of Translational Medicine 21: 242. https://doi.org/10.1186/s12967-023-03958-2.

Romano, Alessandra, Nunziatina Laura Parrinello, Martina Barchitta, et al. 2022. “In-Vitro NET-Osis Induced by COVID-19 Sera Is Associated to Severe Clinical Course in Not Vaccinated Patients and Immune-Dysregulation in Breakthrough Infection.” Scientific Reports 12: 7485. https://doi.org/10.1038/s41598-022-11506-1.

Roy, Saptarshi, Chalatip Chompunud Na Ayudhya, Manoj Thapaliya, Venkata Deepak, and Hydar Ali. 2021. “Multifaceted MRGPRX2: New Insight into the Role of Mast Cells in Health and Disease.” Journal of Allergy and Clinical Immunology 148 (2): 293–308. https://doi.org/10.1016/j.jaci.2021.03.049.

Sanhueza, Sergio, Camilo Cabrera, Romina Quiroga, et al. 2026. “De Novo COVID-19-Associated Insulin Resistance Drives Dysregulated Neutrophil Extracellular Trap Formation (NETosis) Four Months After Infection.” Frontiers in Immunology 17: 1787799. https://doi.org/10.3389/fimmu.2026.1787799.

Schlecker, C., W. Boehmerle, A. Jeromin, B. DeGray, A. Varshney, Y. Sharma, K. Szigeti-Buck, and B. E. Ehrlich. 2006. “Neuronal Calcium Sensor-1 Enhancement of InsP3 Receptor Activity Is Inhibited by Therapeutic Levels of Lithium.” Journal of Clinical Investigation 116 (6): 1668–74. https://doi.org/10.1172/JCI22466.

Schönrich, Günther, and Martin J. Raftery. 2016. “Neutrophil Extracellular Traps Go Viral.” Frontiers in Immunology 7: 366. https://doi.org/10.3389/fimmu.2016.00366.

Sorensen, Barbara, John E. Streib, Mark Strand, et al. 2003. “Complement Activation in a Model of Chronic Fatigue Syndrome.” Journal of Allergy and Clinical Immunology 112 (2): 397–403. https://doi.org/10.1067/mai.2003.1615.

Thierry, Alain R., Tom Usher, Cynthia Sanchez, et al. 2025. “Circulating Microclots Are Structurally Associated with Neutrophil Extracellular Traps and Their Amounts Are Elevated in Long COVID Patients.” Journal of Medical Virology 97: e70613. https://doi.org/10.1002/jmv.70613.

Van der Sluis, Renée Marije, Christian Kanstrup Holm, and Martin Roelsgaard Jakobsen. 2022. “Plasmacytoid Dendritic Cells During COVID-19: Ally or Adversary?” Cell Reports 40 (4): 111148. https://doi.org/10.1016/j.celrep.2022.111148.

Veras, Flavio P., Giovanni F. Gomes, Bruna M. S. Silva, Diego B. Caetíte, Cicero J. L. R. Almeida, Camila Meirelles S. Silva, Ayda H. Schneider, et al. 2023. “Targeting Neutrophils Extracellular Traps (NETs) Reduces Multiple Organ Injury in a COVID-19 Mouse Model.” Respiratory Research 24 (1): 66. https://doi.org/10.1186/s12931-023-02336-2.

Yang, H. et al. 2013. “HMGB1/TLR4 Signaling.” Molecular Medicine.