DNAse Mediated Dissolution of Neutrophil Extracellular Traps (NETs) Accelerates in vitro Thrombin Generation Kinetics in Trauma Patients

Abstract

Introduction:

Neutrophil extracellular traps (NETs) trigger thrombin generation. We aimed to characterize the effects of DNAse on NET components (cell-free DNA [cfDNA] and histones) and thrombin generation after trauma.

Methods:

Citrated plasma samples were collected from trauma patients and healthy volunteers. Thrombin generation (calibrated automated thrombogram) was measured as lag time (LT, minutes), peak height (PH, nM), and time to peak (ttPeak, minutes). Citrullinated histone 3 and 4 were measured by ELISA; cfDNA by PicoGreen© (all ng/mL). Samples analyzed +/− DNAse (1000 u/mL). Results expressed as median and quartiles [Q1, Q3], Wilcoxon testing, p < 0.05 significant.

Results:

We enrolled 46 patients (age 48 [31, 67], 67% male) and 21 volunteers (age 45 [28, 53], 43% male). DNAse treatment of trauma plasma led to: shorter LT (3.11 [2.67, 3.52]; 2.93 [2.67, 3.19]), shorter ttPeak (6.00 [5.30, 6.67];5.48 [5.00, 6.00]), greater PH (273.7 [230.7, 300.5]; 288.7 [257.6, 319.2]), decreased cfDNA (576.9 [503.3, 803.1]; 456.0 [393.5, 626.7]), decreased CitH3 (4.54 [2.23, 10.01]; 3.59 [1.93, 7.98]), and increased H4 (1.30 [0.64, 6.36]; 1.75 [0.83, 9.67]), all p < 0.001. The effect of DNAse was greater on trauma patients as compared to volunteers for LT (ΔLT −0.21 min vs. −0.02 min, p = 0.007), cfDNA (ΔcfDNA −133.4 ng/mL vs. −84.9 ng/mL, p < 0.001) and CitH3 (ΔCitH3 −0.65 ng/mL vs. −0.11 ng/mL, p = 0.004).

Conclusion:

DNAse treatment accelerates thrombin generation kinetics in trauma patient samples as compared to healthy volunteers. These findings suggest that NETs may contribute to the hypercoagulable state observed in trauma patients.

Introduction:

Trauma patients are at increased risk for venous thromboembolism (VTE), a potentially devastating complication, for up to 90-days after injury and it is known that accelerated thrombin generation kinetics after injury contribute to this risk. Plasma thrombin generation kinetics provide a dynamic profile of a patient’s coagulation status, which is not reflected by static measures such as prothrombin time (PT), international normalized ratio (INR), and activated partial thromboplastin time (aPTT). Compared to other means of measuring thrombin levels such as ELISA assays for thrombin-anti-thrombin (TAT) complexes, assessing thrombin generation kinetics provides a dynamic picture of an individual’s thrombin generation at a point in time, giving a more descriptive picture.¹ Prior work from our group has shown that trauma patients have accelerated (hypercoagulable) thrombin generation kinetics compared to healthy volunteers, and that these changes occur within 12 hours from time of injury (TOI) and correlate to injury severity, and others have also demonstrated increased thrombin levels after trauma.^1–3 Furthermore, early accelerated thrombin generation parameters, particularly time to peak (ttPeak), are an independent predictor of both early transfusion requirement and symptomatic VTE up to 90 days after injury.^3,4

Physiologic responses after trauma, including systemic release of damage-associated molecular patterns (DAMPs), increased endothelial permeability, and neutrophil priming, are thought to contribute to disruptions to normal coagulation.^5,6 Neutrophils, in particular, have been shown over the last two decades, to have a significant role in innate immunity, through the formation of neutrophil extracellular traps (NETs).⁷ The formation of NETs (NETosis) depends on citrullination of histone N-terminal tails, the resultant weakening of histone-DNA interactions, with unfolding of tightly wrapped chromatin which is extruded in web-like strands.^8–11 A number of in vivo animal studies have demonstrated that NETosis contributes to disseminated intravascular thrombosis, deep vein thrombosis (DVT), and cancer associated thrombus.^12–15 Cell-free DNA (cfDNA), a component of NETs, has been found at increased levels in trauma patients and may be regulated by deoxyribonuclease (DNAse); however, the cellular origin of this cfDNA is unknown.¹⁶ In a recent prospective cohort study, plasma markers of NETosis, both citrullinated free histones and citrullinated nucleosomes, were found to be increased in trauma patients compared to volunteers.¹⁷ In addition, citrullinated nucleosome levels correlated to accelerated thrombin generation kinetics, suggesting a link between NETosis and hypercoagulability early after trauma.¹⁷

The mechanisms intertwining increased NETosis and accelerated in vitro thrombin generation kinetics in trauma patients have not been well described. Studies co-incubating murine and human neutrophils stimulated to form NETs with normal plasma have shown that NETs promote thrombin generation in a Factor XII dependent manner.^18,19 Cell-free DNA plays a dose-dependent role in thrombin generation through platelet-independent mechanisms and is thought to weakly activate the contact pathway, which may have pathophysiological significance.^19–22 Meanwhile, the other major components of NETs, histones, are able to trigger thrombin generation through platelet dependent mechanisms as well as through platelet independent effects on protein C activation, with recombinant H3 and H4 having the most pronounced procoagulant effects.^20,23,24 It has also been shown that NET components, cfDNA and histones, rather than intact NETs themselves are responsible for accelerated thrombin generation.²⁰ It is unknown how intact NETs, cfDNA, and histones interact within a milieu of circulating DAMPs that are released after trauma, contribute to dysregulated coagulation. In this study, we sought to elucidate these relationships in order to better describe the mechanism for the accelerated thrombin generation kinetics consistently seen after traumatic injury. To our knowledge, this has not been previously explored in clinical samples from trauma patients. We hypothesized that DNAse would increase trauma-induced hypercoagulable state via dissolution of NETs and release of its prothrombotic components.

Methods

Study Design and Sample Source:

This study was approved by the Mayo Clinic Institutional Review Board. Samples were collected from January 2020 to January 2021. Patients presenting to the Mayo Clinic Emergency Department (ED) as trauma activations were considered for inclusion. Exclusion criteria included refusal or inability to obtain informed consent, ongoing systemic anticoagulation (e.g., heparin, warfarin, or novel oral anticoagulants) other than anti-platelet agents (e.g., aspirin, clopidogrel, NSAIDs), known preexisting coagulopathy, cirrhosis, active malignancy, sepsis, renal failure requiring dialysis, burn injuries, major surgery, or another significant trauma within the last year, pregnant women, or prisoners. The time of injury (TOI) was determined by the pre-hospital medical providers based on information at the injury scene. Trauma patients or their Legal Authorized Representative (LAR) were consented for this study after the collection of one or more blood samples. If the patient or LAR could not be consented for study participation or declined consent, the sample was destroyed, and the patient excluded. Demographic and clinical characteristics for each patient were obtained by reviewing the electronic medical record.

Sample Collection, Processing, and Storage:

Samples were collected within 6 hours after the documented TOI. A total of 18mL whole blood was collected by venipuncture or from existing indwelling catheters into 4.5 mL citrated Vacutainer tubes (0.105M buffered sodium citrate, 3.2% Becton Dickinson, Franklin Lakes, NJ) and processed to platelet poor plasma by double centrifugation (3000g, 15 minutes) at room temperature (20–24°C) as recommended by the International Society of Thrombosis and Haemostasis Vascular Biology Standardization Subcommittee and stored in multiple aliquots at −80°C until analysis.²⁵

Calibrated Automated Thrombogram (CAT) Analyses:

Thrombin generation was measured with the Calibrated Automated Thrombogram (CAT, Thrombinoscope BV, Maastricht, Netherlands), utilizing a Fluoroskan Ascent plate reader (390 nm excitation, 460 nm emission, Thermo Electron Corp, Vantaa, Finland), as previously described.^3,26,27 Assays of trauma patient samples were performed in triplicate. Corn trypsin inhibitor (50μg/mL final concentration), a selective Factor XIIa inhibitor which minimizes the effect of the contact pathway on in vitro thrombin generation, was added to each plasma sample prior to analysis.²⁸ Samples were incubated for 15 minutes at 37°C with recombinant deoxyribonuclease (DNAse) I (MilliporeSigma Roche 4716728001) at final concentration of 1000 units/mL or incubation buffer (40 mM Tris-HCl, 10 mM NaCl, 6 mM MgCl₂, 1 mM CaCl₂, pH 7.9) at equivalent volume as a negative control. Thrombin generation was initiated with addition of 20 μL of agonists (5 pM tissue factor and 4 μM phospholipids, Diagnostica Stago, Parsippany, NJ) reagent. Then, 80 μL of citrated platelet poor plasma was added to each well of U-bottom 96-well microtiter plates (Nunc, Thermo Fischer Scientific, Rochester, NY) using a single channel pipette. After an additional incubation period (10 minutes at 37 °C), 20 μL of warmed FLUCA reagent (Fluca kit, TS50, Thrombinoscope, BV), which contains the fluorogenic substrate and CaCl₂ was added to each well via an automated dispenser. Thrombin generation curves were recorded continuously for 90 minutes at a rate of three readings per minute. Separate wells containing the thrombin calibrator, which corrects for inner filter effects and quenching variation among individual plasmas were also measured in parallel. A dedicated software program, Thrombinoscope (Diagnostica Stago, Maastricht, The Netherlands) was used to calculate thrombin activity over time. The parameters derived were: lag time (LT – time to initiation of thrombin generation, minutes), peak height (PH – maximum thrombin generated at a given point during the assay, nM), and time to peak (ttPeak – time to peak thrombin generation, minutes).

In vitro Experiments to Study Effects of NET Components on Thrombin Generation

The following reagents were obtained from EpiCypher, Inc. (Durham, NC): Recombinant Human Mononucleosomes (SKU 16-0009), Recombinant Human Histone 2A (SKU 15-0301), Recombinant Human Histone 2B (SKU 15-0302), Recombinant Human Histone 3.1 (SKU 15-0303), Recombinant Histone 4 (SKU 15-0304), and synthetic Nucleosome Assembly 601 Sequence DNA (SKU 18-0006). Pooled normal plasma (Cryocheck, Precision BioLogic, Bethesda, MD) was incubated with each of these components, to assess their individual effects on plasma thrombin generation kinetics, at a concentration of 20 μg/mL, a dose previously shown to impact thrombin generation, with or without DNAse as above.^20,23,24 CAT assay was performed to quantify in vitro thrombin generation kinetics. All assays were run in triplicate.

Cell Free DNA (cfDNA) and Histone Quantification

Cell free DNA (cfDNA) was quantified using Quant-iT^™ PicoGreen© dsDNA kit (ThermoFisher Scientific, P7581) according to the manufacturer’s instructions at 1:10 plasma dilutions. Citrullinated Histone H3 ELISA kit (Cayman Chemical, Clone 11D3) was used to measure CitH3 following the manufacturer’s instructions at 1:4 dilutions of plasma in assay buffer. Human H4 Sandwich ELISA kit (LS Bio LS-F32393) was used to measure histone 4 (H4) following the manufacturer’s instructions at 1:2 dilutions of plasma in dilution buffer. Plasma samples were incubated with DNAse I at final concentration 1000 units/mL or incubation buffer at equivalent volume. This dosage was used after confirming an effect on thrombin generation (data not shown) and is consistent with prior dosing used in vivo and in vitro.^12,18

Statistical Analyses

Data analysis was performed using SAS, version 9.4 (SAS Institute Inc., Cary, NC). Descriptive statistics are presented as median values and quartiles [Q1, Q3]. Wilcoxon rank-sum testing was used to detect differences between trauma patients and volunteers. Wilcoxon matched-pairs signed rank testing was used to detect differences with and without addition of DNAse. A p-value < 0.05 was considered significant.

Results

Demographics and Clinical Features

There were no significant differences in median age between trauma patients and volunteers (48 years [31, 67] vs. 45 years [28, 53], p = 0.262). There were similarly no significant differences in the portion of male patients (31 (67%) vs. 9 (43%), p = 0.067), Body Mass Index (27.0 kg/m² [22.9, 31.3] vs. 29.4 kg/m² [24.7, 34.9], p = 0.067), or the number of patients on anti-platelet medications (11 (24%) vs. 1 (5%), p = 0.086) between trauma patients and volunteers. Of the trauma patients, blunt trauma was the predominant mechanism of injury, with median injury severity score (ISS) 12 (Table 1). Only 11 of the 46 patients (24%) required blood product transfusion within 24 hours of injury, while 18 (39%) underwent operative intervention requiring general anesthesia within 24 hours from TOI. Patients requiring early operative intervention underwent a variety of procedures including fracture fixation, laparotomy, wound washouts, and craniotomy. Of note, four patients underwent multiple procedures within the first 24 hours from TOI. VTE chemoprophylaxis was started within 24 hours, as recommended by our institutional guidelines, in 19 (41%) of trauma patients. The majority of patients (80%) had a discharge disposition to home after admission. There were two patients who died within 90 days and there were no VTEs.

TABLE 1.

Clinical characteristics of trauma patients (n = 46)

Mechanism (% Blunt)	44 (96%)
Injury Severity Score (ISS)	12 [9, 17]
Traumatic Brain Injury (%)	14 (30%)
Long bone fracture (%)	13 (28%)
Spinal cord injury (%)	2 (4%)
On Presentation
Heart Rate (BPM)	102 [90, 110]
Systolic Blood Pressure (mm Hg)	111 [96, 120]
International Normalized Ratio (INR)	1.0 [1.0, 1.1]
Platelet Count (x 10^9/L)	251 [222, 279]
Hemoglobin (g/dL)	14.2 [13.3, 14.9]
GCS < 15 (%)	12 (26%)
First 24 hours
TXA (%)	4 (9%)
Transfusion of any blood product (%)	11 (24%)
Surgery (%)	18 (39%)
VTE Chemoprophylaxis (%)	19 (41%)
Intensive Care Unit (%)	22 (48%)
Length of Stay (days)	5 [3, 13]
Discharge Disposition
Home (%)	37 (80%)
Acute Rehab (%)	4 (9%)
Nursing Home (%)	4 (9%)
In-hospital mortality (%)	1 (2%)
90-day VTE (%)	0 (0%)
90-day Mortality (%)	2 (4%)

All values expressed as medians and quartiles [Q1, Q3] or counts and percentages.

BPM, beats per minute; GCS, Glasgow Coma Scale; TXA, tranexamic acid.

In vitro Effects of NET Components on Thrombin Generation

To assess the effects of NET components, including nucleosomes, histones, and DNA on thrombin generation, CAT was performed in pooled normal plasma spiked with 20 μg/mL of intact recombinant human mononucleosomes, H2A, H2B, H3.1, H4, or nucleosome assembly DNA (Figure 1). H3.1 and H4 led to accelerated thrombin generation, evidenced by shorter LT, greater PH, and shorter ttPeak, as compared to pooled normal plasma alone. H2A and H2B let to prolonged initiation of thrombin generation (longer LT), while intact mononucleosomes and nucleosome assembly DNA had minimal impact on thrombin generation kinetics. For all conditions, the addition of DNAse led to increased LT, greater PH, and shortened or unchanged ttPeak.

FIG. 1.

Thrombin generation kinetics for pooled normal plasma (Cryocheck) with NET components +/− DNAse (1,000 U/mL).

Thrombin Generation Kinetics, Cell Free DNA, and Histone Levels in Trauma Patients

Baseline plasma thrombin generation kinetics (no DNAse treatment) revealed no significant differences in CAT parameters or H4 levels for trauma patients as compared to volunteers (Table 2), though trauma patients had a trend toward greater PH and shorter ttPeak. Trauma patients had greater levels of both cfDNA and circulating CitH3 as compared to volunteers.

TABLE 2.

Baseline thrombin generation kinetics, cfDNA, and histone levels for volunteers versus trauma patients

	Volunteers (n = 21)	Trauma (n = 46)	P
LT (min)	3.18 [2.67, 3.67]	3.11 [2.67, 3.52]	0.973
PH (nM)	227.3 [206.1, 288.7]	273.7 [230.7, 300.5]	0.084
ttPeak (min)	6.83 [5.67, 7.67]	6.00 [5.33, 6.67]	0.074
cfDNA (ng/mL)	438.8 [381.4, 502.3]	576.9 [503.3, 803.1]	< 0.001
CitH3 (ng/mL)	0.73 [0.40, 3.62]	4.54 [2.23, 10.01]	0.003
H4 (ng/mL)	0.96 [0.41, 91.25]	1.30 [0.64, 6.36]	0.752

LT, lag time; PH, peak height; ttPeak, time to peak; CitH3, citrullinated histone 3; H4, histone 4.

All values expressed as median and quartiles [Q1, Q3]. Wilcoxon rank-sum (unpaired) testing.

Significant P values (P < 0.05) appear in bold.

Thrombin Generation Kinetics, Cell Free DNA, and Histone Levels with and without DNAse I

Treatment of trauma patient samples with DNAse resulted in accelerated thrombin generation profiles as compared to untreated samples (Table 3a), with shorter LT and ttPeak and greater PH. When volunteer samples were treated with DNAse, there was no change in LT, shorter ttPeak, and decreased PH as compared to untreated samples (Table 3b). In both trauma and volunteer samples, DNAse treatment resulted in decreased levels of circulating cfDNA and CitH3 levels. Notably, DNAse treatment led to greater levels of H4 in both trauma and volunteer samples. Additionally, the effect of DNAse was greater on trauma patients as compared to volunteers for LT (ΔLT −0.21 min vs. −0.02 min, p = 0.007), cfDNA (ΔcfDNA −133.4 ng/mL vs. −84.9 ng/mL, p < 0.001) and CitH3 (ΔCitH3 −0.65 ng/mL vs. −0.11 ng/mL, p = 0.004) (Supplemental Table 1).

TABLE 3A.

Trauma patients (n = 46)—thrombin generation kinetics, cfDNA, and histone levels (+/−DNAse) for trauma patients (n = 46) with and without the addition of DNAse I (1,000 U/mL)

	(−) DNAse	(+) DNAse	Median difference	Median % change	P
LT (min)	3.11 [2.67, 3.52]	2.93 [2.67, 3.19]	−0.21	−5.9	< 0.001
PH (nM)	273.7 [230.7, 300.5]	288.7 [257.6, 319.2]	20.7	7.9	< 0.001
ttPeak (min)	6.00 [5.33, 6.67]	5.48 [5.00, 6.00]	−0.55	−8.6	< 0.001
cfDNA (ng/mL)	576.9 [503.3, 803.1]	456.0 [393.5, 626.7]	−133.4	−22.3	< 0.001
CitH3 (ng/mL)	4.54 [2.23, 10.01]	3.59 [1.93, 7.98]	−0.65	−14.4	< 0.001
H4 (ng/mL)	1.30 [0.64, 6.36]	1.75 [0.83, 9.67]	0.44	32.8	< 0.001

LT, lag time; PH, peak height; ttPeak, time to peak; CitH3, citrullinated histone 3; H4, histone 4.

All values expressed as median and quartiles [Q1, Q3]. Wilcoxon rank-sum (unpaired) testing.

Significant P values (P < 0.05) appear in bold.

TABLE 3B.

Volunteers (n = 21)—thrombin generation kinetics, cfDNA, and histone levels (+/−DNAse) for volunteers (n = 21) with and without the addition of DNAse I (1,000 U/mL)

	(−) DNAse	(+) DNAse	Median difference	Median % change	P
LT (min)	3.18 [2.67, 3.67]	3.22 [2.83, 3.50]	−0.02	−0.6	0.891
PH (nM)	227.3 [206.1, 288.7]	245.5 [208.0, 280.3]	7.6	2.0	0.284
ttPeak (min)	6.83 [5.67, 7.67]	6.33 [5.67, 7.22]	−0.36	−5.3	0.002
cfDNA (ng/mL)	438.8 [381.4, 502.3]	355.9 [297.0, 400.1]	−84.9	−20.3	< 0.001
CitH3 (ng/mL)	0.73 [0.40, 3.62]	0.63 [0.40, 3.31]	−0.11	−4.9	0.045
H4 (ng/mL)	0.96 [0.41, 91.25]	1.19 [0.57, 202.13]	0.24	42.8	0.002

All values expressed as median and quartiles [Q1, Q3]. Wilcoxon matched-pairs signed rank test.

Significant P values (P < 0.05) appear in bold.

Thrombin Generation Kinetics, Cell Free DNA, and Histone Levels in Trauma Patients vs. Volunteers after DNAse Treatment

Plasma thrombin generation profiles for samples treated with DNAse demonstrated greater PH and shorter ttPeak in trauma patients as compared to volunteers (Table 4), suggesting accelerated thrombin generation kinetics in trauma patients as compared to volunteers. As prior to treatment with DNAse (Table 2), trauma patient samples still had greater levels of circulating cfDNA and CitH3 in comparison to volunteers when samples were treated with DNAse.

TABLE 4.

Thrombin generation kinetics, cfDNA, and histone levels for volunteers versus trauma patients after incubation with DNAse (1,000 U/mL)

	Volunteers (n = 21)	Trauma (n = 46)	P
LT (min)	3.22 [2.83, 3.50]	2.93 [2.67, 3.19]	0.111
PH (nM)	245.5 [208.0, 280.3]	288.7 [257.6, 319.2]	0.017
ttPeak (min)	6.33 [5.67, 7.22]	5.48 [5.00, 6.00]	0.013
cfDNA (ng/mL)	355.9 [297.0, 400.1]	456.0 [393.5, 626.7]	< 0.001
CitH3 (ng/mL)	0.63 [0.40, 3.31]	3.59 [1.93, 7.98]	0.006
H4 (ng/mL)	1.19 [0.57, 202.13]	1.75 [0.83, 9.67]	0.844

LT, lag time; PH, peak height, ttPeak, time to peak; CitH3, citrullinated histone 3; H4, histone 4.

All values expressed as median and quartiles [Q1, Q3].Wilcoxon rank sum (unpaired) testing.

Significant P values (P < 0.05) appear in bold.

Discussion

In this pilot study, we demonstrate that addition of DNAse to plasma from trauma patients accelerates thrombin generation kinetics in vitro, as compared to without DNAse. This effect of DNAse (decreased ttPeak and increased PH) was greater on trauma patients as compared to volunteers despite decreases in plasma CitH3 and cfDNA with DNAse treatment. This is the first study to assess the role of DNAse mediated NET dissolution on thrombin generation kinetics in clinical samples from trauma patients, who are known to have accelerated NETosis and accelerated thrombin generation kinetics. It has been shown that purified human neutrophil DNA, but not purified intact NETs (complexes of neutrophil-derived extracellular DNA, histones, and other proteins originating from neutrophil granules) or synthetic nucleosomes, can trigger thrombin generation in platelet poor plasma.²⁰ Additionally, another murine model of laser induced thrombosis showed that in vivo treatment with DNAse I significantly inhibited accumulation of polymorphonuclear neutrophils, neutrophil elastase secretion, and platelet thrombus formation following injury, and that the neutrophils present at the laser injury site did not form NETs.²⁹ These studies provide mechanistic insight but do not reflect the pro-inflammatory milieu of cytokines and DAMPs found in plasma from trauma patients. Interestingly, histones on their own or reconstituted with neutrophil DNA also do not trigger in vitro thrombin generation. However, the combination of histones with inorganic polyphosphate (polyP) is able to trigger in vitro thrombin generation, suggesting a synergistic effect.^20,24

The majority of in vitro studies on NETs and thrombin generation have used either no or low (1 pM) tissue factor (TF) to initiate thrombin generation in PPP. A study that used 5 pM TF showed that purified histones, particularly H4, in fact are able to accelerate thrombin generation in PPP through interactions with thrombomodulin, and that these effects persist when DNA is present.²³ In contrast, we used 5 pM TF based on our previous work using CAT assay as predictor of VTE in trauma.³ In our study, DNAse led to decreased in vitro concentrations of CitH3 and cfDNA. It is possible this actually reflects consumption of these NET components leading to accelerated thrombin generation kinetics. On the other hand, we did find that H4 levels were increased with DNAse treatment and it is perhaps this effect in the presence of 5 pM TF that is leading to the accelerated thrombin generation we observed. Interestingly, we did not observe a difference in H4 levels between trauma patients and volunteers, which may be because H4 is largely found within nucleosomes and difficult to quantify in plasma. One additional explanation for our findings is that the procoagulant effects of DNA and histones are enhanced by their anionic charge, which is neutralized within intact NETs.²⁰ Perhaps, DNAse increases prothrombotic potential of histones and cfDNA via dissolution of intact nucleosomes, which are charge neutralized and elevated in trauma patients, as shown in our previous study.¹⁷

To our knowledge, this is the first study to use plasma isolated from trauma patients to study the effect of DNAse on in vitro thrombin generation and NET components, providing a potential link between early innate immune response to injury and downstream effects of TIC in clinical samples. Gould et al. assessed thrombin generation kinetics in plasma samples obtained from septic patients determined to have low, intermediate, and high levels of cfDNA and found that LT and ttPeak were inversely correlated to cfDNA levels while PH was directly correlated to cfDNA levels.¹⁹ Of note, TF was not used to trigger the initiation of thrombin generation in this study. In another study using patient samples to assess the effects of DNAse on plasma thrombin generation kinetics, Kumar et al. found that a one hour incubation with DNAse led to decreased endogenous thrombin potential (ETP – total thrombin generated) in patients with pediatric acute lymphoblastic leukemia.³⁰ Notable differences between this study and ours are that these patients had greater cfDNA but not greater CitH3 levels as compared to healthy volunteers, the authors used low TF (1 pM) for initiation of thrombin generation, and they used significantly lower concentration of DNAse (20 μg/mL which would be approximately 40 u/mL).³⁰ This suggests that our finding of a procoagulant effect of DNAse is related to its effects both on cfDNA as well as histones. It is important to note that studying NETosis in clinical samples is inherently limited by the fact that there are no standardized assays specific for NETosis.³¹ Measuring cfDNA levels is non-specific as cfDNA can be released through cell death¹⁶, and thus we also have measured CitH3 as a marker of NETosis since histone citrullination by peptidylarginine deiminase 4 (PAD4) is a necessary precursor to NETosis.⁸ Our pilot study adds to the limited number of in vitro studies using clinical samples to assess the procoagulant mechanisms of NETs and their components and is the first to study this in trauma patients, who are known to have increased circulating levels of cfDNA, citrullinated nucleosomes, and citrullinated histones.^16,17

Our study has a number of limitations. First, this is a pilot study with a small number of trauma patients and healthy volunteers. The trauma patients included in this manuscript, a representative sample from our Level 1 trauma center, had a relatively low ISS (median 12), and we cannot draw any conclusions about the role of DNAse-mediated NET dissolution on thrombin generation kinetics in the most severely injured patients who are at highest risk for coagulation disturbances. This study did not investigate potential mechanisms by which DNAse itself may be altering thrombin generation. There has been recent work suggesting a DNAse-dependent NET-independent mechanism by which DNAse potentiates thrombus formation in vivo.²⁹ Furthermore, the observed in vitro effects of DNAse on plasma thrombin generation may not be reflective of what is happening in whole blood or in vivo. Future studies are needed to quantify and determine the role of circulating DNAse on altered coagulation after trauma. In this study, samples were only obtained at one time point early after injury, so we cannot determine differences in DNAse effect based on time from injury. Simply quantifying cfDNA and histones as we have done here may not provide the full picture of how cfDNA and histones contribute to thrombin generation in vitro. It is becoming increasingly clear that histone modifications, including methylation, acetylation, and citrullination, may have profound effects on function.³² Further studies are needed to determine how these modifications may contribute to coagulation after traumatic injury and affect assay accuracy for quantification of markers of NETs.

Conclusion

DNAse treatment further accelerates hypercoagulable thrombin generation kinetics in trauma patient samples as compared to healthy volunteers. These findings suggest that production of NETs after injury and their dissolution to NET constituents, may contribute to the hypercoagulable state observed in trauma patients.