Circulating microclots are structurally associated with Neutrophil Extracellular Traps and their amounts are strongly elevated in long COVID patients

doi:10.21203/rs.3.rs-4666650/v1

Download PDF

Article

Circulating microclots are structurally associated with Neutrophil Extracellular Traps and their amounts are strongly elevated in long COVID patients

https://doi.org/10.21203/rs.3.rs-4666650/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

BACKGROUND: The persistence of vasculo-thrombotic complications has been put forward as a possible contributing factor in the long COVID (LC) syndrome.

OBJECTIVES: Given the recently reported separate demonstration of the association of LC with elevated levels of fibrin amyloid microclots (FAM) and with those neutrophil extracellular traps (NETs), markers that are linked to thromboinflammation, this study considers the association of FAM with NETs.

RESULTS: The results show that NETs markers are quantitatively and structurally associated with the size and number of FAM in patients with LC. These markers showed a strong diagnostic performance, both independently and when combined.

CONCLUSIONS: Our study revealed that NETs may be a component of circulating FAM, We suggest that higher NETs formation promotes the stabilization of FAM in the circulation, leading to deleterious effects which contribute causally to the LC syndrome.

Health sciences/Biomarkers/Diagnostic markers

Health sciences/Diseases/Infectious diseases/Viral infection

long COVID

microclots

neutrophil extracellular traps

myeloperoxidase

circulating DNA

immunothrombosis

neutrophil elastase

In the aftermath of the global SARS-CoV-2 pandemic, long COVID (LC) has been recognized as a widespread and challenging condition, with symptoms that impair multiple organ systems and affect millions worldwide. Current estimates indicate that approximately 10–30% of individuals infected with SARS-CoV-2 will experience some form of LC, underscoring the urgent need for effective diagnostic markers and a deeper understanding of its pathophysiology(1–3). Amidst this backdrop, our research details the significant association and potential mechanistic link between fibrin amyloid microclots (FAM) (4) (5) and neutrophil extracellular traps (NETs) (6, 7). These findings not only suggest novel diagnostic pathways but also deepen our understanding of the thromboinflammatory processes underlying long COVID. By exploring the synergistic and potentially orthogonal relationships between FAM and NET markers, we argue for their combined diagnostic potential, paving the way for more targeted and personalized management strategies to confront this enduring health issue.

Increased risks of thrombotic complications such as pulmonary embolism and deep vein thrombosis have been reported more than six months after SARS-CoV2 infection, particularly in patients who have previously suffered severe illness (2, 3). This suggests that and how hypercoagulation may contribute to the varied symptoms experienced in patients with LC, through mechanisms such as capillary blockage, chronic ischaemia-reperfusion injury, and autoantibody formation. Generally speaking, the state of hypercoagulation can affect multiple organs. Although it is known that LC is associated with factors leading to immune dysregulation and tissue damage, the aetiological relationship between LC and coagulation is still under investigation.

The mechanism of healthy blood clotting, culminating in the self-assembly of soluble fibrinogen molecules cleaved by thrombin to make insoluble fibrin is well-understood(9). The resulting fibrin mesh is made up of long fibers, typically with a diameter in the 50–100 nm range. As early as 2008 we suggested that fibrin ultrastructural changes could be useful in identifying disease patterns(9), and referred to the anomalous clots observed as ‘dense matted deposits’(9). We also found that these dense matted deposits were resistant to fibrinolysis, and that this anomalous form could be stained with amyloid stains such as Thioflavin T (ThT) (4). Whereas our initial focus was on the study of clotting pathology in the presence of thrombin, over the years this has evolved, ultimately leading to the discovery that direct protein-protein interactions between circulating inflammatory molecules and soluble fibrinogen are the main drivers of clotting pathologies. The discovery of circulating amyloid fibrin(ogen) marked a crucial milestone, not least due to its significant prevalence in acute COVID-19 cases (5). This shifted the focus to the biochemical and structural characteristics of fibrin(ogen), highlighting a conformational change into an amyloid structure triggered by the presence of circulating inflammatory molecules. These initial discoveries illuminated how such molecules could directly influence pathological clotting, as well as erythrocyte and platelet pathology. Building on this, we coined the term 'fibrin amyloid microclots' (FAM) after identifying these dense amyloid deposits in LC plasma, which are resistant to degradation and could clearly block erythrocyte transport and hence O₂ transport to tissues.

We also showed that these fibrinolytic-resistant FAM entrap various inflammatory molecules, including proteins that prevent clot breakdown (alpha-2-antiplasmin, for example) (10, 11), and that the extent of the FAM, with their trapped inflammatory molecules, is a key influencer in LC pathology. We also showed that the spike protein S1 from SARS-CoV-2 is sufficient induce fibrinolytic-resistant FAM. Recognising that a reliable, quantitative method was needed for measuring FAM in platelet poor plasma, we developed a cell-free imaging flow cytometer method for the detection and quantification of FAM (12). Microscopy methods are also effective in the analysis of such microclots, as observed in sepsis (13).

NETs (7) (6) (14) have also been found to be associated with COVID (15–20) and LC (21), and have long been associated with inflammatory diseases (22). NETs formation (NETosis) is produced by activated neutrophils and is part of the innate immune response. NETs are composed of chromatin and of several proteins, derived notably from granules (e.g. Neutrophil elastase, Myeloperoxidase, bactericidal peptides, cathepsin G, lactoferrin and matrix metalloproteinase-9), each contributing to the elimination of bacteria (22, 23). This allows NETs to eliminate microbes within the first hours of infection, doing so physically by means of the DNA fibers and chemically by means of potent enzymes (NE and MPO, for instance). However, uncontrolled and excessive NETs formation may be deleterious, and has been associated with various sterile and non-sterile disorders, called NETopathies (23) (24) (25). These are mainly inflammatory, and include microbial infection, autoimmune diseases, cancer, diabetes, arthritis, and others (17, 22). DNA is the main NETs component, and the principal NETs byproducts consist of specifically degraded or fragmented DNA that are shed into the bloodstream (circulating DNA, cirDNA) (26). Given that NETs formation has been found to be associated with cirDNA in various inflammatory diseases(22, 27, 28), and that the production of cirDNA was recently shown to derive directly from NETs degradation (26), cirDNA may represent a biomarker of NETs formation. The participation of NETs in thrombus formation has previously been suggested, based on the observation of thrombotic events in various NETopathies (17, 22, 29). In addition, the ex vivo release of intertwined fibrin and NETs scaffolds by isolated, stimulated neutrophils in fibrin clots has been observed. For instance, NETs markers have been detected in a coronary thrombus removed with percutaneous coronary thromboaspiration from a patient with acute myocardial infarction (30).

We have separately revealed both the presence of FAM (1, 4, 7), and elevated cirDNA and NETs in LC, and we here assess the extent to which they are linked/associated. We used the same cohort of healthy individuals (HI) and individuals with LC (HI, Table S1 and LC, Table S2) show the demographics and co-morbidities of LC and both HI cohorts). We first determined the number of FAM and their size, and simultaneously determined the plasma concentrations of NETs protein markers (MPO, NE) and total cirDNA, and performed correlation tests. We then assessed whether NETs constituents were present within the FAM, seeking to determine a possible structural association between the two entities.

Circulating FAM and NETs markers quantitative association

Elevated FAM count in LC patients

We previously developed an imaging flow cytometry method to quantify FAM in platelet poor plasma (12). As determined by imaging flow cytometry, the total number of FAM is statistically much higher in patients with LC than in healthy individuals (P = 1.97x10^− 12, 19.7-fold median difference; Fig. 1A and B, Table S3-11). For all arbitrarily selected microclot size ranges (< 10, 10–20, 20–30, 30–40, and > 40 µM), the median number of FAM is statistically and significantly higher in patients with LC than in healthy individuals (Fig. 2A-C). Note, the number of FAM is greatest within the 10–20 µm size range and lowest in the size fraction > 40 µm (Fig. 2A). This is in accordance with our previous results on controls and LC (12). All patients with LC and almost all HI have some FAM of up to 30 µm (Table S3). In contrast, it is notable that 68.4% and 13.1% of healthy individuals have in their plasma FAM in the 30–40 µm range and > 40 µm, respectively (Table S11). In individuals with LC, the proportion of patients showing FAM in these two size ranges is much higher: 98% and 60%, respectively (1.4- and 4.6-fold greater, respectively; Table S11), indicating that the FAM may be greater in size as well as in numbers in LC patients. The imaging flow cytometry results presented here reveal both a significant microclot presence and an increase in both numbers and sizes in the LC samples (Fig. 1A and B and 2A). This is in line with previous results (12).

NETs markers and FAM counts are quantitatively associated and discriminate LC from HI individuals

The plasma concentrations of circulating DNA, MPO and NE are statistically much higher in patients with LC than in healthy patients (5.7-, 3.5- and 14.9-fold; Fig. 2B and Fig. S1 and S2). This confirms our previously published results (21). The Spearman analysis-based correlation study shows that in healthy individuals only the concentrations of NE and MPO are statistically associated, as has been demonstrated previously(16, 21). On the other hand, the concentrations of NE, MPO, and cirDNA, as well as the total number of FAM, are all clearly associated in the plasma of individuals with LC (Fig. 2C and D). The number in the 10–20µm size range showed the highest discrimination between LC and HI and the highest correlation value with cirDNA, MPO and NE (Fig. 2C and D). The number of FAM appears to be especially discriminating between LC and HI: a comparison of the cohorts of healthy individuals and individuals with LC shows the area under the ROC curves to be 0.90; 0.95; 0.90; 0.89; 0.88 and 0.74 for the total number of FAM, the number determined within size ranges of 0–10, 10–20, 20–30, 30–40 and > 40 µm, respectively (Fig. S3). Furthermore, the concentrations of cirDNA, MPO, NE and the three combined markers are also very discriminating (AUROC 0.90; 0.94; 0.85 and 0.95; respectively; Fig. S4), confirming our previous results (21). When scatter plotting the markers with the highest AUROC (MPO and the 10–20µm size range FAM count), HI and LC patients are distinguishable (Fig. 2E).

Structural association of NETs with circulating FAM

The characteristic NETs filaments released by activated neutrophil can be observed using fluorescence microscopy and the Hoechst DNA marker, as can a more dense, whitish/milky chromatin outside cells (see arrows) (Fig. 3A-D). We utilized staining techniques to visualize the co-incidence of microclot formation and NETs in acute COVID-19 and LC, compared to controls. The micrographs in Fig. 3E and F highlight the use of Thioflavin T (ThT) for FAM, alongside Hoechst or SYTO stains for DNA, indicating NETs presence (Hoechst in blue, SYTO in red, and ThT in green, with a combined overlay in the third column). The bottom row (Fig. 3G) shows plasma samples treated with ThT, SYTO and myeloperoxidase (MPO) stains (SYTO in red, ThT in green and MPO in blue). MPO, ThT and DNA clearly co-localized within FAM. and may also be observed alone in a few localizations in the FAM image. Arrows highlight DNA localizations, illustrating NETs surrounding or outside FAM, or between two FAM (Fig. 3 bottom right image; enlargement in Fig. 4B). Note, previous studies showed MPO and DNA only co-localized when neutrophils are activated (31), or within a coronary thrombus (30). In addition, in vitro colocalization of neutrophils, extracellular DNA and coagulation factors such as fibrinogen was reported during NETosis (32).

In their structural, biological, physiological, and pathophysiological characteristics, FAM differ from traditional blood clots (or fibrin mesh formation) in several fundamental ways. Structurally, FAM tend to be smaller and are found as free entities in circulation. They also exhibit significant structural heterogeneity. Healthy fiber diameters are more uniform and clot into a gel mesh, typically only induced during a physiological clotting event such as following a cut, and under the action of thrombin and the activation of the clotting cascade. Biologically, clots formed in such ‘physiological’ circumstances are composed largely of platelets and fibrin designed to stop bleeding. By contrast FAM are characterized by an amyloid-like nature, which can be revealed by staining them with suitable fluorogenic dyes. In addition, we have shown that FAM contain not only fibrinogen molecules, but also entrap various inflammatory molecules and even antibodies (10, 12). An important factor initiating platelet activation and the formation of intravascular FAM is endothelial damage triggering widespread endothelial damage (endothelialitis). The presence of intravascular FAM in small blood vessels perpetuates endothelial cell damage (endothelialitis), which consequently adopt the pro-adhesive phenotype, thus exposing those FAM to microorganisms that favour the elimination and limit the dissemination of the pathogen (33, 34).

Impact of NETs and FAM association on LC complications

Abnormal blood clotting and FAM formation thus appear as important factors that may explain the development and persistence of symptoms in LC. Indeed, several researchers have signalled the exacerbation by a dysregulated thromboinflammatory response of the interplay between coagulation and innate immunity(29, 30, 35, 36). Platelet hyperactivation and endothelial dysfunction would therefore appear to be a unifying pathway underlying the wide range of symptoms seen in LC (2, 11).

Mechanistically, cell viral infection and the resulting inflammation leads to the stimulation of neutrophils, which leads in turn to the release of tissue factors and the degradation of anticoagulants by monocytes, promoting coagulation(25). As a result, the interplay between the innate immune response and blood coagulation (immunothrombosis) leads to the formation of microthromboses (29, 35, 37).

Existing evidence points to a central role played by NETs in immunothrombosis (27, 29, 33, 37, 38). Thus, activated platelets are an endogenous stimulus capable of inducing NETs (22, 25, 39). NETs interact physically with platelets,both through their binding with the P-selectin glycoprotein ligand and through the binding of the high mobility group box 1 (HMGB1) nuclear protein to the RAGE or TLR4 receptors (39). There is also other evidence of the mechanistic link between NETs and coagulation, such as the binding of: (i) the coagulation factor fibrinogen with DNA-rich NETs; (ii), the coagulation factors prothrombin, FX and FVIIa with activated neutrophils during NETosis; and (iii), the anticoagulant activated (40) protein C (APC) with activated neutrophils and DNA-rich NETs (29, 41).

Neutrophils and platelets activate each other mutually, and their physical interaction constitutes a central loop of amplification of immuno-thrombosis: it induces platelet-derived HMGB1, which in turn triggers NETs formation, and it amplifies NETosis via the binding of platelets to neutrophils via glycoprotein 1B (GPIb) (39). This amplification loop is thus closely associated with that of inflammation, due in particular to the deterioration or activation of the endothelial cells by FAM, which in turn triggers the inflammation process (40, 42). Consequently, exacerbated and deleterious NETs production is not only maintained by the NETs/thrombosis feedback loop but also by the positive NETs/inflammation feedback loop (42). This may cause a persistent stimulation of neutrophils, which may in turn lead to the pathogenic microthrombosis phenomena observed in LC. Recent proteomic analysis showed neutrophil activation as well as neutrophil signatures in LC patients associated with clotting cascade members. Our observations point to the possibility that, in addition to this hypercoagulation state, a persistent imbalance in NETs formation, plausibly caused by viral persistence (43), may lead to a high structural association of NETs with FAM. This would impair clot lysis, contributing to their higher numbers in the circulation and a lower clearance level. The protective role of NETs in infection, therefore, would come at a high cost in the case of SARS-CoV2 infection, as well as may be in certain other inflammatory diseases.

Co-detection of FAM and NETs markers as a diagnostic test

The ability to detect these markers efficiently would pave the way for better disease monitoring, understanding the etiology of inflammation, and personalized treatment strategies. More specifically, a detection strategy that could demonstrate the combined presence of high microclot numbers and elevated levels of both NETs markers and cirDNA (either by microscopy, cytometry or ELISA) would offer an efficient means of tackling the underdiagnosis of LC, thereby improving diagnostic accuracy and appropriate intervention. In order to further evaluate the performance of an LC detection test in using markers of both entities, we employed Decision Tree-Gini, K-nearest Neighbors Classifier, and Random Forests for training by machine learning. Out of 88 samples (50 and 38 from LC and HI individuals, respectively), training models were created from 79 samples (90%) and evaluated on an unseen independent test set with 9 samples (10%) (Fig. 4C, and Fig. S5). In these models, we utilized all markers for training and. Five-fold cross-validation was used for each classifier using Stratified-KFold for preventing sample bias (Fig. S6). The mean feature importance in cross-validation Random Forest model indicates that MPO, FAM 0–10, FAM 10–20, FAM objects, microclot counts, and cirDNA contribute to the discrimination between healthy and LC subjects (Fig. 4D, and Fig. S5). Random Forest models provide the more interesting performance value in training with accuracy of 0.958, AUC of 0.950, 1.000 of sensitivity, and 0.900 specificity, and for the independent test an accuracy, AUC, sensitivity, and specificity of 1.000, respectively (Fig. 4E and F, and Fig. S5). This evaluation result indicates the effectiveness in avoiding overfitting in Random Forest models. By excluding the least important markers one-by-one, we found using a Random Forest model that the most important markers in most models are MPO and FAM 10–20 showing accuracy, AUC, sensitivity and specificity of 1.0 in the validation test and 0.95, 0.95, 0.96 and 0.94, respectively in the cross-validation test (Stratified K-fold) (Fig. 4, and Fig. S6). These data showed that combining a NETs marker (MPO) and microclot numbering marker (10–20 µm size) are able to obtain a very high level of discriminative/screening power (Fig. S7) as illustrated by scatter plot presentation of MPO and 10–20 µm FAM concentration values (Fig. 2E).

Plasma from healthy individuals contains low numbers of FAM of low sizes as observed here (Fig. 1A and B, Fig. 3; and Fig. S2 and S3) and previously (7), but this study revealed they may be covered at least partially with NETs. We suggest that FAM are bound/entrapped by floating NET parts in the blood stream, and that physico-chemical properties of both entities enable close structural association inhibiting their degradation in the circulation. Thus, NET parts may now be conceived as a potential microclot constituent in addition to the insoluble fibrin amyloid and other molecules like cytokines or antibodies(10, 12).

FAM frequently appear in chronic inflammatory states and are more resistant to fibrinolysis, showing resistance even to powerful proteolytic enzymes such as trypsin (9). Pathophysiologically, their presence and characteristics are closely related to the severity of various diseases, such as Alzheimer’s, Parkinson’s, rheumatoid arthritis, and COVID-19 (9). This observation adds to the evidence that FAM play a significant role in disease progression. Consequently, we suggest that imbalanced NETs formation as observed in various inflammatory diseases such as COVID-19 or LC leads to higher FAM stability and therefore their elevated number in the circulation. Such resistance to normal breakdown mechanisms and the variability in their formation, even in the absence of typical clotting triggers such as thrombin, indicates that FAM may have more widespread and clinically deleterious effects.

This work shows a robust association between biomarkers indicative of thromboinflammatory activity and LC. Our study revealed that NETs may be a component of circulating FAM, suggesting that higher NETs formation promotes the stabilization of FAM in the circulation, leading to deleterious effects which (in part) may contribute to the symptoms of LC. The discovery of these biomarker linkages not only presents a possible novel diagnostic methodology but also novel therapeutic targets, offering prospects for future markedly improved clinical management. Consequently, our findings present a significant advancement in the understanding of the interactions between NETs and FAM in Long COVID.

Study cohort

In an attempt to establish a potential connection between the prevalence of fibrin amyloid microclots (FAM) in Long COVID (LC) patients and elevated markers for neutrophil extracellular traps (NETs), we compiled two comparison groups: one comprised of controls from South Africa [SA controls, (n = 14)] and the other composed of controls from France [EFS controls, (n = 24)]. These control cohorts consisted of healthy individuals exhibiting no symptoms of LC at the time of blood collection. By incorporating two subsets of controls exposed to different cultural and disease contexts, we aimed to uncover potential similarities or differences in their inflammatory profiles within the collected data. LC patients were recruited with help from our clinical collaborators, who identified post-acute COVID-19 individuals still experiencing persistent symptoms associated with LC. The self-reported symptoms and co-morbidities of LC patients were compared with both control populations. Ethical clearance was obtained before initiating the study, and we ensured that informed consent was obtained from all participants before collecting a small blood donation. The subsequent sections provide a more detailed discussion of the demographics and considerations for all sample cohorts.

Ethical clearance

The investigation received ethical approval from the Health Research Ethics Committee (HREC) at Stellenbosch University in South Africa, under the references: N19/03/043 and project ID: 9521 with annual re-approval. The French ethics approval number is EFS-PM N° 21PLER2018-0069. We have a data and material sharing agreement between Stellenbosch University and INSERM (S008426). Participants were recruited based on the information they provided to their clinicians in collaboration with the study. Prior to sample collection, participants were extensively briefed on the study’s objectives, potential risks, and details of the entire process to gain their informed consent. Throughout all study procedures there was mandatory adherence to ethical guidelines and principles, as detailed by the Declaration of Helsinki, South African Guidelines for Good Clinical Practice, and Medical Research Council Ethical Guidelines for Research.

Sample demographics and considerations

Whole blood was obtained from randomized selection of LC patients, healthy French controls, and healthy South African control individuals. Platelet poor plasma was cold-shipped to South Africa and to France. Blood samples were gathered from 14 healthy volunteers in South African as the first healthy control cohort [7 males; 7 females; mean age 28 (26–42)]. The control participants had no previous medical history of cardiovascular disease or coagulation disorders, did not use prescription anticoagulant medications, were non-smokers, and were not pregnant at the time of sample collection. It is important to mention that two control participants had hypercholesterolemia and one had hypertension; however, both conditions were effectively managed with treatment and diet.

A second cohort of 24 controls was drawn from a database of healthy blood donors in France, serving as our French control cohort. The specific details regarding their gender distribution [15 males; 9 females; median age 45.5 (19–64)], and other demographic information was outlined. Similar to the South African cohort, these individuals were selected due to the absence of LC symptoms, and the same exclusion criteria were applied to this group. None of these individuals had a previous history of cardiovascular disease or coagulation complications. The co morbidities of this cohort were documented and compared with those of the disease cohort. In addition, the study incorporated blood samples from 50 LC patients [18 males; 32 females; mean age 50 (± 17)]. The LC volunteers completed a questionnaire or report, either during their clinic visit in consultation with their attending physician or as part of their documentation for blood sample donations. The questionnaire gathered information on volunteers’ age, gender, pre-existing comorbidities, and chronic medications use. Furthermore, volunteers provided details on their COVID 19 vaccination history, a report on their acute COVID 19 infection, including its severity and diagnosis, and a self-reported list of persistent LC symptoms.

Blood sample preparation

Blood samples were obtained from all participants. The South African samples were prepared by isolating platelet poor plasma (PPP) and haematocrit. Whole blood (WB) samples underwent a centrifugation step at 3000×g for 15 minutes at room temperature. Following centrifugation, the PPP supernatant was extracted, platelets in the hematocrit fraction and FAM in the plasma fraction were immediately analysed. The rest of the PPP was stored in 1.5mL Eppendorf tubes, and the frozen at − 80°C. A portion of the frozen PPP was cold shipped to the INSERM laboratory in France.

Fluorescence microscopy of platelets

The haematocrit of the fresh samples were used for analysis of the South African controls and LC samples. CD62PE (P-selectin on platelet surface) (IM1759U, Beckman Coulter, Brea, CA, USA) and to PAC-1 (activated glycoprotein (GP) IIb/IIIa) (340507, BD Biosciences, San Jose, CA, USA) was added to the hematocrit. P-selectin is released from the cellular granules during platelet activation and then moves to the surface of the platelet membrane. The antibody PAC-1 detects the neoepitope of active GPIIb/IIIa. PAC-1 antibody binding is correlated with platelet activation. After the samples was incubated at room temperature for 30 minutes, samples were also viewed using the Zeiss Axio Observer 7 fluorescence microscope with a Plan-Apochromat 63×/1.4 Oil DIC M27 objective (Carl Zeiss Microscopy, Munich, Germany). We used excitation wavelength 450 to 488 nm and the emission at 499 to 529 nm (PAC-1), and excitation 540 nm to 570 nm and the emission 577 nm to 607 nm (CD62PE).

Fluorescence microscopy of FAM in platelet-poor plasma samples

For this analysis, PPP was used from all the 3 cohorts (controls and LC samples from South Africa and controls form France). Microclot analysis was done on stored PPP samples, by exposing 49 µL PPP to 1 µL Thioflavin T (ThT) followed by incubation for 30 minutes at room temperature (measures taken to shield the samples from light exposure). Following incubation, a 3 µL smear of stained PPP sample was prepared on a microscope slide, covered with a glass coverslip. For microscopy analysis the excitation wavelength was chosen to be between 450 nm to 488 nm, paired with an emission wavelength ranging from 499 nm to 529 nm. Visualising and evaluating these samples utilised a Zeiss Axio Observer 7 fluorescent microscope with a Plan Apochromat 63x/1.4 Oil DIC M27 objective (Carl Zeiss Microscopy, Munich, Germany).

Hoechst NETS analysis

In a subset of samples, we exposed 49 µL PPP to 1 µL ThT and 50 µL of Hoechst 33342 (20mg/mL) (Cayman Chemical Company, Item#: 15547). The samples were incubated at room temperature for 30 minutes, followed by a microscopy slide preparation as described previously. The microscope settings were kept as before for the ThT, with the settings for the Hoechst set at 370 nm to 400 nm excitation wavelength and the emission wavelength was set at 410 nm to 440 nm. The Zeiss Axio Observer 7 fluorescent microscope with a Plan Apochromat 63x/1.4 Oil DIC M27 objective (Carl Zeiss Microscopy, Munich, Germany) was again used to view the samples.

Imaging flow cytometry

For this analysis, PPP from the SA control group (n = 14), French control group (n = 24), and the LC group (n = 50) samples were thawed at room temperature after removal from the − 80°C freezer. Upon thawing at room temperature, 47 µL of PPP was incubated for 30 minutes at room temperature with 3 µL ThT (final concentration: 0.005 mM), (Sigma Aldrich, St. Louis, MO, USA). Our investigation utilised imaging flow cytometry with the Amnis® FlowSight® Imaging Flow Cytometer (Luminex, Seattle, WA, USA). This method integrates fluorescence microscopy for single-cell morphology analysis with the capabilities of imaging flow cytometry, enabling precise discrimination of heterogenous cell populations and high flow rate capacity while maintaining data quality.

Prior to sample acquisition, a gate was established to exclusively collect data on ThT positive events within the gate. This approach significantly reduced the size of acquisition files, minimised background noise, and enhanced the speed of data analysis for each file. To achieve this, a template was constructed, utilising the 405 nm and 488 nm lasers. The gate was established using a negative control (comprising water and ThT), a standard unstained control (PPP without ThT), and a standard control (PPP stained with ThT) as reference points. Utilising information from these reference samples, a gate was constructed to exclude any regions with abnormal fluorescent signals (trapped air bubbles in the sample syringe). Channel 1 displayed images captured by the brightfield microscope, while channel 7 showed all ThT positive events using the 405 nm laser. Sample acquisition, performed using the ThT positive template, was conducted over a 5 minute duration for each sample run.

The flow cytometer underwent daily calibration using FlowSight® Calibration Beads (400300, Luminex, Seattle, WA, USA), with a sterilisation step involving 100% bleach solution and a subsequent water run to ensure no beads remained in the sample load line. Before sample acquisition began, a negative control (comprising water and ThT) was run to confirm no ThT positive events were recorded post calibration. The subsequent raw data files underwent analysis using IDEAS 6.2 software, and the results were presented using GraphPad Prism 10 (version 10.2.0).Heathy individuals control samples

Healthy Individuals (HI, n = 24) were obtained from the Etablissement Français du Sang (EFS), which is Montpellier’s blood transfusion center (Convention EFS-PM N° 21PLER2018-0069). Blood sample from healthy individuals was collected in 5-milliliter Streck tubes. No difference in terms of NETs markers and cirDNA was found when comparing EDTA vs Streck tubes.

Plasma isolation and cirDNA extraction

Blood sample from healthy individuals was centrifuged at 1200g at 4°C for 10 minutes, and the plasma supernatants were immediately centrifuged at 16,000g at 4°C for 10 minutes. Afterward, the plasma was either immediately used for DNA extraction or stored at -20°C. CirDNA was extracted from 50 µL of plasma using the Maxwell® RSC Blood DNA Kit (Promega) according to the “RSC ccfDNA plasma” in an elution volume of 100 µL. Until they were used, DNA extracts were stored at -20°C. The same extraction procedure was used for the LC-19 patients and group control.

Quantification of CirDNA

Quantitative PCR IntPlex® method was employed to analyze CirDNA following MIQE recommendations (Bustin, 2010; Bustin et al., 2009), as previously described (Mouliere et al., 2014; Thierry et al., 2014). Total CirDNA concentration in each sample was measured using amplification of a 67 bp wild-type KRAS gene sequence (amplification primers were 5’CCTTGGGTTTCAAGTTATATG3’ sense and 3’CCCTGACATACTCCCAAGGA5’ antisense). Q-PCR was conducted in a 25 µl reaction volume on a CFX96 instrument with CFX management software 3.0 (Bio-Rad). The PCR reaction mixture included Bio-Rad Super mix SYBR Green PCR mix, amplification primers, PCR-analyzed water, and DNA extract. Thermal cycling involved a 3-minute Hot-start Polymerase activation denaturation step at 95°C, followed by 40 cycles at 95°C for 10 s and 60°C for 30 s. Melting curves were obtained from 55°C to 90°C with a plate reading every 0.2°C. Quantifications were calibrated using serial dilutions of DIFI cell line genomic DNA, and sample concentrations were extrapolated using the standard curve. All experiments included duplicate samples, and DNA quantification data were validated using an internal control with a known concentration.

Myeloperoxidase and Neutrophil Elastase assay

MPO and NE concentrations were determined using ELISA according to the manufacturer's instructions (Duoset R&D Systems, DY008, DY3174, and DY9167-05),. Captured antibodies were diluted with Reagent Diluent (RD) from supplementary reagent kits (DY008), coated on 96-well microplates, and incubated overnight at room temperature. After clearing and washing with Wash Buffer (WB), microplates were blocked with RD for two hours. After washing, 100µL of diluted 1/10 plasma samples, standards, and controls were added and incubated for one hour at room temperature. The standard curve ranged from 0.12 to 8 ng/mL. After removal and washing, detection antibodies diluted in RD were added for one hour. Following another wash, microplates were incubated for 30 minutes at room temperature with streptavidin-HRP. After three washes, substrate solution (100µL per well) was added, and after a 10-minute incubation, the PHERAstar FS instrument, in conjunction with the PHERAstar control software, promptly read the Optical Density (O.D) at 450 nm for each well.

Machine learning assistance

Samples were randomly divided into a training-validation test (90% − 79 samples) and an independent test set (10% − 9 samples). The former comprised 70% training samples and 30% validation samples. The team initially trained Decision Tree-Gini, K Neighbors Classifier, and Random Forest for all markers using scikit-learn 1.4.2. Parameters – criterion and random state – were fine-tuned through manual search. Other supported criteria remained at their defaults. Upon training, with the correlation matrix heatmap and feature importances algorithm, the team identified the seven most crucial markers, including cirDNA (ng/mL), NETs biomarkers (MPO (ng/mL) and NE (ng/mL)), and three FAM biomarkers (FAM count, FAM in 0–10µm range, FAM in 10–20µm range, and FAM in 20–30µm range).

Since Random Forest performed better evaluation, this approach was ultilized for those most critical markers with 1000 trees (n_estimators = 1000). Following training, to prevent overfitting, we applied the independent test set. In each round, we re-analyzed mean feature importances and excluded the least significant marker until the final two most crucial ones. Moreover, to present unbiased performance, we conducted a five-fold cross-validation using StratifiedKFold (n = 5).

We computed major evaluation metrics, including accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC). Models were performed and optimized using Python 3 (ipykernel 6.19.2) in JupyterLab 3.6.3 (Anaconda Navigator 2.4.3, MacBook Pro with Apple M1 chip and 16 GB RAM).

Statistical analysis

The choice of statistical tests was made based on the nature of the data and the specific objectives of the analysis. For comparing groups with non-parametric data, the Mann-Whitney U test was selected. This test is appropriate when the data deviate from a Gaussian distribution, as confirmed by the Shapiro-Wilk test.

For exploring relationships between variables, correlation analysis was conducted using the Spearman test. The Spearman test evaluates the strength and direction of association between variables without assuming linearity. It is particularly appropriate when the relationship between variables may be monotonic but not necessarily linear, or when outliers are present in the data. This test calculates the Spearman correlation coefficient, which ranges from − 1 to 1, with values closer to 1 indicating a strong positive correlation, values closer to -1 indicating a strong negative correlation, and values around 0 indicating no correlation.

All statistical analyses were performed using Graph Pad Prism 8.3.1 software. A significance level of 0.05 was chosen to determine statistical significance, meaning that results with a p-value less than 0.05 were considered statistically significant. To provide a clear indication of the level of significance, p-values were reported with asterisk notation: *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

1. Z. Al-Aly, E. Topol, Solving the puzzle of Long Covid. Science 383, 830–832 (2024).

2. A. Nalbandian, K. Sehgal, A. Gupta, M. V. Madhavan, C. McGroder, J. S. Stevens, J. R. Cook, A. S. Nordvig, D. Shalev, T. S. Sehrawat, N. Ahluwalia, B. Bikdeli, D. Dietz, C. Der-Nigoghossian, N. Liyanage-Don, G. F. Rosner, E. J. Bernstein, S. Mohan, A. A. Beckley, D. S. Seres, T. K. Choueiri, N. Uriel, J. C. Ausiello, D. Accili, D. E. Freedberg, M. Baldwin, A. Schwartz, D. Brodie, C. K. Garcia, M. S. V. Elkind, J. M. Connors, J. P. Bilezikian, D. W. Landry, E. Y. Wan, Post-acute COVID-19 syndrome. Nat Med 27, 601–615 (2021).

3. H. E. Davis, L. McCorkell, J. M. Vogel, E. J. Topol, Long COVID: major findings, mechanisms and recommendations. Nat Rev Microbiol 21, 133–146 (2023).

4. E. Pretorius, S. Mbotwe, J. Bester, C. J. Robinson, D. B. Kell, Acute induction of anomalous and amyloidogenic blood clotting by molecular amplification of highly substoichiometric levels of bacterial lipopolysaccharide. J R Soc Interface 13, 20160539 (2016).

5. E. Pretorius, C. Venter, G. J. Laubscher, P. J. Lourens, J. Steenkamp, D. B. Kell, Prevalence of readily detected amyloid blood clots in “unclotted” Type 2 Diabetes Mellitus and COVID-19 plasma: a preliminary report. Cardiovasc Diabetol 19, 193 (2020).

6. L. Stiel, C. Mayeur-Rousse, J. Helms, F. Meziani, L. Mauvieux, First visualization of circulating neutrophil extracellular traps using cell fluorescence during human septic shock-induced disseminated intravascular coagulation. Thromb Res 183, 153–158 (2019).

7. V. Brinkmann, U. Reichard, C. Goosmann, B. Fauler, Y. Uhlemann, D. S. Weiss, Y. Weinrauch, A. Zychlinsky, Neutrophil extracellular traps kill bacteria. Science 303, 1532–1535 (2004).

8. D. B. Kell, E. Pretorius, Are fibrinaloid microclots a cause of autoimmunity in Long Covid and other post-infection diseases? Biochem J 480, 1217–1240 (2023).

9. E. Pretorius, D. B. Kell, A Perspective on How Fibrinaloid Microclots and Platelet Pathology May be Applied in Clinical Investigations. Semin Thromb Hemost, doi: 10.1055/s-0043-1774796 (2023).

10. A. Kruger, M. Vlok, S. Turner, C. Venter, G. J. Laubscher, D. B. Kell, E. Pretorius, Proteomics of fibrin amyloid microclots in long COVID/post-acute sequelae of COVID-19 (PASC) shows many entrapped pro-inflammatory molecules that may also contribute to a failed fibrinolytic system. Cardiovasc Diabetol 21, 190 (2022).

11. E. Pretorius, M. Vlok, C. Venter, J. A. Bezuidenhout, G. J. Laubscher, J. Steenkamp, D. B. Kell, Persistent clotting protein pathology in Long COVID/Post-Acute Sequelae of COVID-19 (PASC) is accompanied by increased levels of antiplasmin. Cardiovasc Diabetol 20, 172 (2021).

12. S. Turner, G. J. Laubscher, M. A. Khan, D. B. Kell, E. Pretorius, Accelerating discovery: A novel flow cytometric method for detecting fibrin(ogen) amyloid microclots using long COVID as a model. Heliyon 9, e19605 (2023).

13. J. Schofield, S. T. Abrams, R. Jenkins, S. Lane, G. Wang, C.-H. Toh, Microclots, as defined by amyloid-fibrinogen aggregates, predict risks of disseminated intravascular coagulation and mortality. Blood Adv 8, 2499–2508 (2024).

14. S. Yousefi, D. Stojkov, N. Germic, D. Simon, X. Wang, C. Benarafa, H.-U. Simon, Untangling “NETosis” from NETs. Eur J Immunol 49, 221–227 (2019).

15. M. Pastorek, M. Dúbrava, P. Celec, On the Origin of Neutrophil Extracellular Traps in COVID-19. Front Immunol 13, 821007 (2022).

16. S. Kapoor, L. Mihalovičová, E. Pisareva, B. Pastor, A. Mirandola, B. Roch, J. Bryant, A. P. Princy, S. Chouaib, A. R. Thierry, Association of vascular netosis with COVID-19 severity in asymptomatic and symptomatic patients. iScience 27, 109573 (2024).

17. A. R. Thierry, B. Roch, Neutrophil Extracellular Traps and By-Products Play a Key Role in COVID-19: Pathogenesis, Risk Factors, and Therapy. JCM 9, 2942 (2020).

18. Y. Zuo, S. Yalavarthi, H. Shi, K. Gockman, M. Zuo, J. A. Madison, C. N. Blair, A. Weber, B. J. Barnes, M. Egeblad, R. J. Woods, Y. Kanthi, J. S. Knight, Neutrophil extracellular traps in COVID-19. JCI Insight, doi: 10.1172/jci.insight.138999 (2020).

19. E. A. Middleton, X.-Y. He, F. Denorme, R. A. Campbell, D. Ng, S. P. Salvatore, M. Mostyka, A. Baxter-Stoltzfus, A. C. Borczuk, M. Loda, M. J. Cody, B. K. Manne, I. Portier, E. S. Harris, A. C. Petrey, E. J. Beswick, A. F. Caulin, A. Iovino, L. M. Abegglen, A. S. Weyrich, M. T. Rondina, M. Egeblad, J. D. Schiffman, C. C. Yost, Neutrophil extracellular traps contribute to immunothrombosis in COVID-19 acute respiratory distress syndrome. Blood 136, 1169–1179 (2020).

20. H. Ng, S. Havervall, A. Rosell, K. Aguilera, K. Parv, F. A. von Meijenfeldt, T. Lisman, N. Mackman, C. Thålin, M. Phillipson, Circulating Markers of Neutrophil Extracellular Traps Are of Prognostic Value in Patients With COVID-19. Arterioscler Thromb Vasc Biol 41, 988–994 (2021).

21. E. Pisareva, S. Badiou, L. Mihalovičová, A. Mirandola, B. Pastor, A. Kudriavtsev, M. Berger, C. Roubille, P. Fesler, K. Klouche, J. Cristol, A. R. Thierry, Persistence of neutrophil extracellular traps and anticardiolipin auto‐antibodies in post‐acute phase COVID‐19 patients. Journal of Medical Virology 95 (2023).

22. V. Papayannopoulos, Neutrophil extracellular traps in immunity and disease. Nat Rev Immunol 18, 134–147 (2018).

23. S. Van Bruggen, K. Martinod, The coming of age of neutrophil extracellular traps in thrombosis: Where are we now and where are we headed? Immunol Rev 314, 376–398 (2023).

24. M. J. Kaplan, M. Radic, Neutrophil Extracellular Traps: Double-Edged Swords of Innate Immunity. J.I. 189, 2689–2695 (2012).

25. S. K. Jorch, P. Kubes, An emerging role for neutrophil extracellular traps in noninfectious disease. Nat Med 23, 279–287 (2017).

26. E. Pisareva, L. Mihalovičová, B. Pastor, A. Kudriavtsev, A. Mirandola, T. Mazard, S. Badiou, U. Maus, L. Ostermann, J. Weinmann-Menke, E. W. I. Neuberger, P. Simon, A. R. Thierry, Neutrophil extracellular traps have auto-catabolic activity and produce mononucleosome-associated circulating DNA. Genome Med 14, 135 (2022).

27. S. T. Abrams, B. Morton, Y. Alhamdi, M. Alsabani, S. Lane, I. D. Welters, G. Wang, C.-H. Toh, A Novel Assay for Neutrophil Extracellular Trap Formation Independently Predicts Disseminated Intravascular Coagulation and Mortality in Critically Ill Patients. Am J Respir Crit Care Med 200, 869–880 (2019).

28. B. Pastor, J.-D. Abraham, E. Pisareva, C. Sanchez, A. Kudriavstev, R. Tanos, A. Mirandola, L. Mihalovičová, V. Pezzella, A. Adenis, M. Ychou, T. Mazard, A. R. Thierry, Association of neutrophil extracellular traps with the production of circulating DNA in patients with colorectal cancer. iScience 25, 103826 (2022).

29. C. Thålin, Y. Hisada, S. Lundström, N. Mackman, H. Wallén, Neutrophil Extracellular Traps: Villains and Targets in Arterial, Venous, and Cancer-Associated Thrombosis. Arterioscler Thromb Vasc Biol 39, 1724–1738 (2019).

30. I. Varjú, K. Kolev, Networks that stop the flow: A fresh look at fibrin and neutrophil extracellular traps. Thromb Res 182, 1–11 (2019).

31. W. Zhao, D. K. Fogg, M. J. Kaplan, A novel image-based quantitative method for the characterization of NETosis. Journal of Immunological Methods 423, 104–110 (2015).

32. L. D. Healy, C. Puy, A. Itakura, T. Chu, D. K. Robinson, A. Bylund, K. G. Phillips, E. E. Gardiner, O. J. T. McCarty, Colocalization of neutrophils, extracellular DNA and coagulation factors during NETosis: Development and utility of an immunofluorescence-based microscopy platform. Journal of Immunological Methods 435, 77–84 (2016).

33. A. Bonaventura, A. Vecchié, A. Abbate, F. Montecucco, Neutrophil Extracellular Traps and Cardiovascular Diseases: An Update. Cells 9, 231 (2020).

34. T. A. Fuchs, J. A. Kremer Hovinga, D. Schatzberg, D. D. Wagner, B. Lämmle, Circulating DNA and myeloperoxidase indicate disease activity in patients with thrombotic microangiopathies. Blood 120, 1157–1164 (2012).

35. L. Nicolai, R. Kaiser, K. Stark, Thromboinflammation in long COVID—the elusive key to postinfection sequelae? Journal of Thrombosis and Haemostasis 21, 2020–2031 (2023).

36. M. C. Woodruff, K. S. Bonham, F. A. Anam, T. A. Walker, C. E. Faliti, Y. Ishii, C. Y. Kaminski, M. C. Ruunstrom, K. R. Cooper, A. D. Truong, A. N. Dixit, J. E. Han, R. P. Ramonell, N. S. Haddad, M. E. Rudolph, S. Yalavarthi, V. Betin, T. Natoli, S. Navaz, S. A. Jenks, Y. Zuo, J. S. Knight, A. Khosroshahi, F. E.-H. Lee, I. Sanz, Chronic inflammation, neutrophil activity, and autoreactivity splits long COVID. Nat Commun 14, 4201 (2023).

37. A. Bonaventura, A. Vecchié, L. Dagna, K. Martinod, D. L. Dixon, B. W. Van Tassell, F. Dentali, F. Montecucco, S. Massberg, M. Levi, A. Abbate, Endothelial dysfunction and immunothrombosis as key pathogenic mechanisms in COVID-19. Nat Rev Immunol 21, 319–329 (2021).

38. F. Denorme, I. Portier, J. L. Rustad, M. J. Cody, C. V. de Araujo, C. Hoki, M. D. Alexander, R. Grandhi, M. R. Dyer, M. D. Neal, J. J. Majersik, C. C. Yost, R. A. Campbell, Neutrophil extracellular traps regulate ischemic stroke brain injury. J Clin Invest 132, e154225 (2022).

39. M. Scherlinger, C. Richez, G. C. Tsokos, E. Boilard, P. Blanco, The role of platelets in immune-mediated inflammatory diseases. Nat Rev Immunol, doi: 10.1038/s41577-023-00834-4 (2023).

40. D. B. Kell, G. J. Laubscher, E. Pretorius, A central role for amyloid fibrin microclots in long COVID/PASC: origins and therapeutic implications. Biochemical Journal 479, 537–559 (2022).

41. J. H. Griffin, P. Lyden, COVID-19 hypothesis: Activated protein C for therapy of virus-induced pathologic thromboinflammation. Res Pract Thromb Haemost 4, 506–509 (2020).

42. Thierry AR and Salmon D., Inflammation-, thrombosis- and autoimmune-feedback loops may lead to persistent neutrophil stimulation in long COVID.

43. A. D. Proal, M. B. VanElzakker, S. Aleman, K. Bach, B. P. Boribong, M. Buggert, S. Cherry, D. S. Chertow, H. E. Davies, C. L. Dupont, S. G. Deeks, W. Eimer, E. W. Ely, A. Fasano, M. Freire, L. N. Geng, D. E. Griffin, T. J. Henrich, A. Iwasaki, D. Izquierdo-Garcia, M. Locci, S. Mehandru, M. M. Painter, M. J. Peluso, E. Pretorius, D. A. Price, D. Putrino, R. H. Scheuermann, G. S. Tan, R. E. Tanzi, H. F. VanBrocklin, L. M. Yonker, E. J. Wherry, SARS-CoV-2 reservoir in post-acute sequelae of COVID-19 (PASC). Nat Immunol 24, 1616–1627 (2023).

Acknowledgments

The authors thank the excellent technical assistance of F. Frayssinoux (IRCM, Institut de Recherche en Cancérologie de Montpellier, INSERM U1194, Université de Montpellier, Institut régional du Cancer de Montpellier, Montpellier, F-34298, France) and Cormac Mc Carthy (Mc Carthy Consultant, Montpellier) for English editing (financial compensation). We are grateful to D. Salmon, M. Ychou, J.P. Cristol, and K. Klouche (Centre Hospitalier Universitaire de Montpellier) for fruitful discussion.

Funding

The study was partially supported by SIRIC Montpellier Cancer Grant INCa_Inserm_DGOS_12553 (ART); South African National Research Foundation (142142); (EP); and the South African Medical Research Foundation (EP), the Novo Nordisk Foundation grant NNF20CC0035580 (DBK) and the Balvi Foundation grant 18 (DBK).

Authors contributions

Conceptualization: ART, EP

Methodology: CS, BP, TH, AM, EEP, ST, MW, AT, TU, CV, CP

Investigation: ART, EP, TU, CP

Visualization: EP, CV, TU

Funding acquisition: ART, EP

Project administration: CV, EP, ART

Supervision: ART, EP

Writing – original draft: ART, TH, EP, CV

Writing – review & editing: EP, ART, CV, DBK, GJL

Competing interests

ART, EEP and BP are author of a patent: NEW METHOD TO DIAGNOSE INFLAMMATORY DISEASES 11194720 PCT application number PCT/EP2022/072147, Date of receipt 05 August 2022. EP is an author of a patent DIAGNOSTIC METHOD FOR LONG COVID PCT application number GB2105644.5. EP is a founding director of Biocode Technologies, a Stellenbosch University start-up company. ST and AT are employees of Biocode Technologies, as well as postgraduate students of EP. All other authors: no competing interests to declare.

Data and materials availability

All data are available in the main text or the supplementary materials. Material and data sharing agreement (signed on the 26th of July 2023, #S008426) between Stellenbosch University (SA) and INSERM (France).

Yes there is potential Competing Interest. ART, EEP and BP are author of a patent: NEW METHOD TO DIAGNOSE INFLAMMATORY DISEASES 11194720 PCT application number PCT/EP2022/072147, Date of receipt 05 August 2022. EP is an author of a patent DIAGNOSTIC METHOD FOR LONG COVID PCT application number GB2105644.5. EP is a founding director of Biocode Technologies, a Stellenbosch University start-up company. ST and AT are employees of Biocode Technologies, as well as postgraduate students of EP. All other authors: no competing interests to declare.

ALLSUPPLEMENTARYFIGURESandTABLESFORTHEMANUSCRIPT.pdf
Supplementary Dataset 1

Download PDF

Version 1

posted

You are reading this latest preprint version

Circulating microclots are structurally associated with Neutrophil Extracellular Traps and their amounts are strongly elevated in long COVID patients

Status:

Version 1

Abstract

Figures

Introduction

Objectives and study design

Circulating FAM and NETs markers quantitative association

Elevated FAM count in LC patients

NETs markers and FAM counts are quantitatively associated and discriminate LC from HI individuals

Structural association of NETs with circulating FAM

Impact of NETs and FAM association on LC complications

Co-detection of FAM and NETs markers as a diagnostic test

General conclusions

Materials and methods

Study cohort

Ethical clearance

Sample demographics and considerations

Blood sample preparation

Fluorescence microscopy of platelets

Fluorescence microscopy of FAM in platelet-poor plasma samples

Hoechst NETS analysis

Imaging flow cytometry

Plasma isolation and cirDNA extraction

Quantification of CirDNA

Myeloperoxidase and Neutrophil Elastase assay

Machine learning assistance

Statistical analysis

References

Declarations

Additional Declarations

Supplementary Files

Status:

Version 1