Keywords
neutrophil extracellular traps - acute ischemic stroke - thrombolysis resistance -
deoxyribonuclease I - intravenous thrombolysis
A class of cerebrovascular illnesses known as strokes has a high prevalence of disability
and death. Stroke is the third most common cause of mortality and the fourth most
common source of disease burden worldwide, according to the most recent statistics
from the Global Burden of Disease Study.[1]
[2] Among various types, the most frequent kind is ischemic stroke.[3] As an arterial thromboembolic event, early reperfusion is essential for acute ischemic
stroke (AIS) treatment to preserve the ischemic penumbra; the sole FDA-approved thrombolytic
medication for AIS at the moment is recombinant tissue plasminogen activator (rt-PA).[4] However, its therapeutic efficacy is significantly limited by a narrow therapeutic
time window and numerous contraindications, with less than 15% of new-onset AIS patients
receiving intravenous thrombolysis (IVT) in most European countries.[5]
More concerning, the early recanalization rate with rt-PA is only about 33%.[6] Although rt-PA continues to dissolve fibrin, platelets may still be recruited to
the dissolved thrombus, potentially leading to secondary occlusion.[7] This phenomenon, known as thrombolysis resistance, was initially attributed to factors
such as thrombus location,[6] thrombus length,[8] post-translational modification of fibrin,[9] and clot contraction.[10] However, recent advancements in mechanical thrombectomy (MT) have provided technical
support for acquiring pathological thrombi, enabling deeper analysis of thrombus composition
and structure. This research has revealed that non-fibrin components, such as neutrophil
extracellular traps (NETs), leukocytes, and von Willebrand factor (vWF), are also
critical contributors to thrombolysis resistance.[11]
[12]
[13]
Neutrophils are the first to be drawn to the infection site and are crucial protectors
of the host's innate immune response, where they perform immune defense functions
through three primary mechanisms: phagocytosis of pathogens, release of antimicrobial
granules, and formation of NETs.[14] Beyond their role in infectious diseases, neutrophils are also a part of the aseptic
inflammation that occurs when tissue is damaged.[15] Studies have shown that neutrophils extravasate from the leptomeningeal vessels
after stroke and gradually accumulate at the lesion site. Notably, neutrophils show
indications of activity, such as histone H3 citrullination, chromatin decondensation,
and the release of cellular contents indicative of NETs formation.[16]
[17] NETs have been found in AIS patients' plasma and thrombi, and their concentrations
are highly connected with the severity of the illness and a bad prognosis.[18]
[19]
[20] Further research has demonstrated that NETs contribute to thrombosis through mechanisms
such as platelet interaction,[21] promotion of atherosclerosis,[22] and activation of the coagulation cascade.[23] Additionally, NETs provide scaffolds for platelet and erythrocyte aggregation,[24] alter the fibrin architecture, and form a hard thrombus shell,[25] which acts as a barrier to thrombolysis.
The processes of NETs generation and their function in AIS will be covered in this
review, with a focus on their involvement in thrombolysis resistance. We will also
explore the potential of targeting NETs as an adjunctive strategy in AIS thrombolytic
therapy.
Basics of NETs
Origins of NETs
In 2004, Brinkmann et al first observed neutrophils forming extracellular network
structures under an electron microscope after stimulation with interleukin-8, phorbol
myristate acetate (PMA), or lipopolysaccharide. These activated neutrophils released
histones, DNA, and granule proteins—such as neutrophil elastase (NE), myeloperoxidase
(MPO), and cathepsin G (CG)—which captured and killed bacteria. This structure was
named “NETs,” and “NETosis” was the name given to the process by which they formed.[26] Due to characteristic chromatin decondensation and the inability of both NADPH oxidase
(NOX) inhibitor-treated neutrophils as well as neutrophils from chronic granulomatous
disease patients (which involves a defective NOX gene) to form NETs under PMA stimulation,[27] NETosis was initially considered a NOX-dependent process distinct from apoptosis
and necrosis.[28] Further studies have demonstrated that a range of stimuli binding to distinct receptors
can cause NETosis, including bacteria and their derivatives,[29]
[30]
[31]
[32] fungi,[33] viruses,[34] immune complexes,[35] specific cytokines (e.g., IL-1β, IL-8, and TNF-α),[36] and crystals.[37]
Formation Mechanism of NETs
Suicidal NETosis is the first identified and most studied form of NETs formation.
It is characterized by its dependence on NOX and is accompanied by neutrophil death.
Using the classical stimulus PMA as an example, activation of neutrophil surface receptors
increases intracellular calcium levels, which activates protein kinase C (PKC) and
peptidylarginine deiminase 4 (PAD4). This cascade subsequently induces the generation
of reactive oxygen species (ROS) from NOX2 via the Raf/MEK/ERK signaling pathway.[38] ROS then triggers the dissociation of NE from the azurophilic granule complex into
the cytoplasm, where F-actin is degraded as a result of MPO activating NE's proteolytic
action. After that, NE moves into the nucleus, where it breaks down histones and aids
in chromatin decondensation.[39]
[40] PAD4 is essential to the chromatin decondensation mechanism described above, and
one of the possible mechanisms is that PAD4 reduces the positive charge on the histone
surface by catalyzing the citrullination of histones, thereby weakening its electrostatic
interaction with DNA.[41]
[42] Ultimately, a pore-forming protein, gasdermin D (GSDMD), creates holes inside the
plasma and nuclear membranes, permitting citrullinated histones to be released, DNA
and released proteins in the extracellular space such as granules come together to
form NETs.[31]
[43]
The second type, termed vital NETosis, is characterized by independence from NOX,
and neutrophils always maintain plasma membrane integrity and functional activity.
This process has been shown to be triggered by Staphylococcus aureus or Gram-negative bacteria via Toll-like receptors (TLRs) and complement receptors.[44]
[45]
Staphylococcus aureus, a highly invasive pathogen, is rapidly captured by neutrophils, which form NETs
to contain its spread. Upon stimulation of Toll-like receptor 2 (TLR2) and complement
receptors by Staphylococcus aureus, neutrophils undergo rapid changes: their multilobular nuclei round-up and concentrate,
and the nuclear membranes' inner and outer layers split apart. Nuclear DNA and granule
proteins sprout from the outer nuclear membrane and form vesicles, which are exocytosed
onto the plasma membrane and ultimately assembled into NETs in the extracellular space.
The remaining neutrophil nucleus, though empty, retains its defensive role.[32]
[46]
Mitochondrial NETosis, first identified and described by Yousefi et al, occurs after
stimulation with granulocyte-macrophage colony-stimulating factor (GM-CSF), lipopolysaccharide
(LPS), or complement factor 5a. In this process, neutrophils release DNA from mitochondria
in a way that is dependent on ROS to form NETs without inducing neutrophil death.[47] Recent studies have shown that the mechanism of mitochondrial ROS and NETs formation
involves the opening of the mitochondrial permeability transition pore.[48]
Pathologic Role of NETs in AIS
Pathologic Role of NETs in AIS
Kinetics of NETs during AIS
Following the onset of acute ischemic stroke (AIS), the hypothalamic–pituitary–adrenal
(HPA) axis and sympathetic nervous system are rapidly activated. Glial cells and damaged
neurons release damage-associated molecular patterns (DAMPs), which induce endothelial
cells to express chemokines. As a result, neutrophils are mobilized from the bone
marrow and spleen into the peripheral circulation. These circulating neutrophils then
undergo a well-orchestrated sequence of events, including intravascular rolling, adhesion,
transendothelial migration, and ultimately infiltration into the brain parenchyma.[49]
In the absence of reperfusion, peripheral neutrophils predominantly access the brain
via leptomeningeal vessels, traversing along the Virchow–Robin space into the perivascular
space, and subsequently penetrating the brain parenchyma. In fact, neutrophils are
already activated prior to their entry into the brain parenchyma.[17] Notably, neutrophil infiltration into the brain occurs in a time-dependent manner,
suggesting that the rate of infiltration is closely associated with the extent of
basement membrane disruption.[50]
Based on the synthesis of relevant studies ([Table 1]), we preliminarily propose the following timeline: NETs begin to form intravascularly
within 0.5 to 6 hours post-stroke, infiltrate the brain parenchyma between 6 and 24 hours,
and peak at 2 to 5 days. However, there is considerable variability among the findings
of different studies. We attribute this variability to several factors: (1) Differences
in animal species and experimental models, notably, neutrophil infiltration appears
to be more pronounced following pMCAO compared with tMCAO.[51] (2) Variability in the sensitivity of detection methods. (3) Some studies initiated
measurements too late and employed prolonged intervals between time points, lacking
continuous early-phase dynamic monitoring.[52]
[53]
[54]
[55]
Table 1
Compilation of NETs kinetic studies during AIS
|
Animals
|
Model
|
Methods
|
Viewpoints
|
Note
|
Reference
|
|
ICR mice
|
Carotid occlusion
|
IF
|
The count of NETs within thrombi began to increase starting at 0.5 hour after occlusion
|
None
|
[52]
|
|
WT C57BL/6 mice
|
tMCAO
|
QFM
|
NETs formed in cerebral vasculature by 6 hours post-stroke, spread to brain parenchyma
by 12 hours, and peaked at 24 hours
|
None
|
[53]
|
|
SD rats
|
pMCAO
|
IF, WB
|
NETs enter via the leptomeninges at 6 hours post-stroke, appear in the cortex and
peripheral blood by 12 hours, and reach the striatum by 24 hours
|
The earliest detection time points were set at 6 hours for vasculature and 12 hours
for the parenchyma
|
[54]
|
|
WT C57BL/6 mice
|
pMCAO
|
WB
|
NETs appear in the ischemic cortex by 1 day post-infarction, peaking at 3–5 days
|
The detection time points spanned days 1, 3, and 5
|
[55]
|
|
Balb/C mice
|
pMCAO
|
IF
|
NETs can be detected in the capillary lumen, perivascular space, and parenchyma before
24 hours after ischemia
|
Lack of specific early detection time points
|
[17]
|
|
WT C57BL/6 mice
|
tMCAO
|
FM,
SEM
|
NETs formed in the cortex by 6 hours and striatum by 12 hours post-stroke, peaking
at 2–3 days
|
The earliest detection time point was at 6 hours
|
[16]
|
Abbreviations: FM, fluorescence microscopy; IF, immunofluorescence; pMCAO, permanent
middle cerebral occlusion; QFM, quantitative fluorescence microscopy; SEM, scanning
electron microscope; tMCAO, transient middle cerebral artery occlusion; WB, Western
blot.
Involvement of NETs in Thrombosis
An increasing body of research has shown that innate immune cells, including neutrophils
and monocytes, contribute to immune defense by participating in thrombosis. In 2013,
this physiological process was referred to as immunothrombosis by Engelmann and Massberg.[56] Although immunothrombosis helps to limit the spread of pathogens, its dysregulation
can lead to thrombotic diseases, including cerebral infarction, myocardial infarction,
and deep vein thrombosis.[57] The discovery of NETs in both venous and arterial thrombosis offers direct evidence
of their involvement in these processes.[58]
[59] Laridan et al analyzed thrombi taken from AIS patients undergoing MT. They found
that neutrophils were found in every thrombi, and the existence of NETs was shown
by the colocalization of citrullinated histone H3 (CitH3) with extracellular DNA.[60] Similarly, Ducroux et al demonstrated that 108 AIS thrombi samples had NETs present
in large quantities, primarily localized to the outer layer of the thrombus.[61] Importantly, the content of NETs in thrombi has been identified as a potential predictor
of AIS severity and poor functional outcomes.[19]
[20]
NETs Interact with Platelets to Promote Thrombosis
Under normal physiological conditions, neutrophils and platelets do not interfere
with each other. However, in pathological states, their interaction leads to mutual
activation, influencing both NETs formation and thrombus development.[21]
[62] Upon platelet activation, P-selectin and CD40 ligand (CD40L) expression on the platelet
surface is upregulated.[63] These molecules then bind to P-selectin glycoprotein ligand-1 (PSGL-1) and CD40
on neutrophils, respectively, activating macrophage-1 antigen (Mac-1) through a tyrosine
kinase-dependent mechanism.[64] Once activated, Mac-1 binds to platelets via GPIbα and intercellular adhesion molecule
2 (ICAM2), promoting neutrophil–platelet adhesion.[65]
[66] Additionally, activated platelets may promote the recruitment and activation of
neutrophils by releasing serotonin, chemokines, and high mobility group box-1 protein
(HMGB1).[67]
[68]
[69] Conversely, activated neutrophils promote platelet activation and thrombosis through
the release of cathelicidins and NETs.[70]
[71]
[72]
[73] Fuchs et al induced neutrophils to form NETs using PMA and subsequently perfused
the NETs with platelets. Under electron microscopy, they observed time-dependent platelet
aggregation on the NETs, along with platelet activation.[24] Then, they perfused the NETs with blood either treated or untreated with deoxyribonuclease
I (DNase I). In the DNase I-treated group, NETs were rapidly degraded and platelet
aggregates did not form. After 10 minutes, DNase I was added to the untreated group,
resulting in the rapid clearance of both NETs and platelet aggregates. These findings
suggest that NETs serve as scaffolds for platelet aggregation.
Recent studies highlight HMGB1 as a key mediator of the crosstalk between NETosis
and thrombosis: during the acute phase of cerebral ischemia, HMGB1 expression on platelets
is significantly upregulated, mediating platelet aggregation, activation, and thrombosis
via the TLR4/MyD88 and cGMP/PKG pathways.[74]
[75] Activated platelets release HMGB1, it subsequently attaches to neutrophils' TLR4
and receptor for advanced glycation end products (RAGE), causing NETosis and encouraging
thrombus development.[67]
[75]
[76]
[77]
NETs Promote the Coagulation Cascade
NETs can promote the coagulation cascade through several mechanisms. Key processes
include NETs components enhancing thrombin production by activating platelets, triggering
endogenous coagulation pathways, and degrading inhibitors of exogenous coagulation
pathways.[23] Extracellular histones can induce platelet activation and procoagulant phenotype
expression through TLR2 and TLR4, which in turn promote the generation of plasma thrombin.
The presence of DNA further enhances histones' ability to stimulate thrombin production.[71]
[72] Additionally, NETs' negatively charged DNA backbone can bind and activate factor
XII, thereby enhancing the endogenous coagulation pathways and shortening clotting
time when DNA is added to human plasma.[78]
[79] Interestingly, Noubouossie et al shown that coagulation activation is not directly
promoted by intact NETs in vitro, which may be due to interactions between histones
and DNA within the nucleosome that neutralize the negative charge of the NETs' DNA
surface.[80] Furthermore, NE and CG in NETs can inactivate tissue factor pathway inhibitor (TFPI),
thereby enhancing the tissue factor (TF)-induced coagulation pathway.[81] Zhou et al demonstrated that NETs contribute to the hypercoagulability in AIS patients
with internal carotid artery occlusion. They found phosphatidylserine (PS)-bearing
NETs in the plasma and thrombi of these patients, which enhanced platelet aggregation
and coagulation factor deposition, promoting thrombin and fibrin formation. NETs-derived
proteases and histones also exert toxic effects on vascular endothelial cells (ECs),
inducing a procoagulant phenotype in ECs by promoting PS exposure and TF expression.[82]
[83]
[84]
NETs Involved in Atherosclerosis
Arterial occlusion caused by atherosclerotic thrombus is a significant cause of ischemic
stroke.[85] Recent research has revealed that NETs exist in atherosclerotic plaques.[86] In ApoE knockout (Apoe −/−) mice fed a high-fat diet for 3 weeks, NETs were detected
in the atherosclerotic lesions of the aortic root.[87] Blocking NETs formation in this model by injecting Cl-amine, a peptidylarginine
deiminase (PAD) inhibitor, reduced the formation of carotid atherosclerotic plaques.[88] Liu et al constructed PAD4 gene knockout mice, which further confirmed that the
specific deletion of PAD4 reduced NET formation and vascular inflammation and caused
a significant reduction in atherosclerotic burden in Apoe−/− mice.[87] However, Franck et al found that while PAD4 gene defects reduced EC damage and plaque
erosion by inhibiting NETosis, they did not significantly impact atherosclerosis formation
or progression in hypercholesterolemic mice.[89]
NETs may contribute to atherosclerosis through two primary mechanisms: inflammatory
stimulation and immune activation, with oxidized low-density lipoprotein (ox-LDL)
and cholesterol crystallization playing key roles. First, NETs can directly induce
EC death, leading to collagen exposure and platelet aggregation, which further triggers
NETosis and exacerbates local inflammation, creating a vicious cycle that promotes
atherosclerosis.[90]
[91] On the one hand, MPO can induce the oxidative modification of LDL, and ox-LDL is
subsequently phagocytosed by macrophages and forms foam cells, which contribute to
the formation of atherosclerotic plaque by accumulating underneath the artery intima.[92]
[93] Conversely, ox-LDL can activate TLR-PKC-IRAK-MAPK and NADPH oxidase pathways, stimulating
the NETs generation and aggravating chronic inflammation in the arterial intima.[94] Additionally, cholesterol crystal-induced NETs can trigger the release of IL-1β
by macrophages, which in turn activates T helper 17 (Th17) cells and upregulates the
expression of IL-17. This cascade amplifies immune cell recruitment to the lesion
site, promotes vascular inflammation and endothelial dysfunction, and ultimately increases
the development and spread of atherosclerotic plaques.[95] Recent studies also show that low shear stress, resulting from hemodynamic changes,
can induce NETs formation via Piezo1, a mechanically gated ion channel, thereby exacerbating
atherosclerosis.[96]
NETs and Thrombolysis Resistance
NETs and Thrombolysis Resistance
NETs as Thrombus Components Affect Thrombolytic Efficacy
IVT has offered hope for reperfusion therapy of AIS, but it is not equally effective
against all thrombi, particularly in cases of cardioembolic stroke.[97] The composition and structure of thrombus are now considered to be the critical
factors influencing thrombolytic efficacy.[98] Although AIS thrombi exhibit considerable heterogeneity, they can generally be categorized
into erythrocyte-rich and platelet-rich regions. The erythrocyte-rich regions typically
have a simpler structure, primarily consisting of erythrocytes embedded in a thin
fibrin network. Conversely, areas that are rich in platelets are distinguished by
a thick network of fibrin that acts as a scaffold, along with non-fibrin components
such as vWF, leukocytes, extracellular DNA, or NETs. NETs are often localized along
the thrombus surface, especially in places that are rich in platelets or at the intersection
of regions that are rich in erythrocytes and platelets.[12]
[99]
[100] Numerous studies have shown that cardioembolic thrombi, compared with atherosclerotic
thrombi, have a higher fibrin/platelet ratio, contain more leukocytes and NETs, and
have fewer erythrocytes.[101]
[102]
[103] This compositional difference may help explain the variability in thrombolytic efficacy.
In vitro experiments have demonstrated that erythrocyte-rich thrombi are more susceptible
to rt-PA than thrombi that are rich in platelets, a finding that aligns with clinical
observations.[98]
[104]
Mechanism of NETs Leading to Thrombolysis Resistance
Fuchs et al's study further demonstrated that, beyond activating platelets and serving
as scaffolds for platelet and erythrocyte aggregation, NETs also promote fibrin formation
and deposition by interacting with vWF, fibronectin, and fibrinogen, thereby enhancing
thrombus stability.[24] They also compared the sensitivity of NETs and fibrin to thrombolysis in in vitro
clots. It was found that rt-PA could effectively remove fibrin but could not prevent
subsequent thrombus formation. In rt-PA-resistant clots, platelets and erythrocytes
remained bound together via the DNA scaffold of NETs. Only when rt-PA was used in
combination with DNase I could thrombus formation be inhibited. Therefore, NETs can
provide a thrombus scaffold that is independent of fibrin and resistant to rt-PA.
As the primary constituents of NETs, histones, and extracellular DNA can alter the
fibrin architecture in thrombi and make them thicker fibrin fibers, accompanied by
lower permeability, they can exert anti-fibrinolytic effects.[105]
[106] Zhang et al found that NETs contribute to blood hypercoagulability, leading to microthrombosis
and consumption of rt-PA. Importantly, NETs make it easier for PS to reach platelet
and EC surfaces, enhancing their procoagulant activity and promoting the release of
vWF and plasminogen activator inhibitor-1 (PAI-1), which further contribute to thrombolysis
resistance.[107] Di Meglio et al examined the thrombi of 199 AIS patients with large vessel occlusion
using scanning electron microscopy and immunohistochemistry. They observed that all
these thrombi exhibited a shell composed of dense thrombus components (including fibrin,
vWF, platelets, extracellular DNA, and NETs), which formed a barrier that hindered
thrombus dissolution by rt-PA.[25] One of the mechanisms may be that the NETs in the shell limit the binding of rt-PA
to its substrate (fibrin). Recent proteomic studies support these findings, showing
that the amount of fibrin(ogen) within AIS thrombi correlates positively with the
presence of NETs. The dense fibrin cap created by this interaction reduces fibrinolytic
activity by decreasing thrombus permeability. Furthermore, the large number of neutrophils
and NETs surrounding the thrombus impairs fibrinolysis by covering the thrombus surface,
ultimately contributing to thrombolysis resistance.[108] The current knowledge of NETs involved in rt-PA-resistant thrombosis is summarized
in [Fig. 1].
Fig. 1 Mechanisms of neutrophil extracellular traps (NETs) involved in rt-PA-resistant thrombosis.
① Platelets and neutrophils interact through specific ligands and receptors, leading
to mutual activation. High mobility group box-1 protein (HMGB1) released by activated
platelets binds to the receptor for advanced glycation end products (RAGE) on neutrophils,
inducing NETosis. ② NETs damage endothelial cells (Ecs), thereby inducing Ecs to expose
collagen and phosphatidylserine (PS) and express tissue factors (TF), which cause
platelet aggregation and procoagulant phenotype expression. ③ Activated platelets
with a procoagulant phenotype, along with the DNA backbone of NETs, activate factor
XII (FXII), initiating the endogenous coagulation pathway. NETs further enhance coagulation
by inactivating tissue factor pathway inhibitor (TFPI). ④ Macrophages engulf NETs-induced
oxidized low-density lipoprotein (ox-LDL) to form foam cells and develop into atherosclerotic
plaques, and NETs aggravate inflammation and endothelial dysfunction by initiating
IL-1β/Th17, resulting in unstable plaque rupture and arterial thrombosis. ⑤ NETs provide
scaffolds for platelet and erythrocyte aggregation and promote fibrin fiber thickening,
thus enhancing thrombus stability. Additionally, NETs stimulate the release of von
Willebrand factor (vWF) and PAI-1 from platelets and ECs. Together with the platelets
and NETs, this forms a dense shell together with NETs and platelets, ultimately exacerbating
thrombolysis resistance (created with Adobe illustrator).
Current Strategies for Improving Thrombolysis Resistance
Current Strategies for Improving Thrombolysis Resistance
Although MT is a valuable treatment, it remains limited in availability, making IVT
the preferred reperfusion therapy for AIS in most regions. Therefore, improving the
effectiveness of IVT remains of critical clinical importance. The composition of an
AIS thrombus is highly complex, and since rt-PA primarily targets fibrin, its “one-size-fits-all”
approach is insufficient to achieve optimal recanalization. Recent trials targeting
non-fibrin components of AIS thrombus have provided valuable insights into overcoming
thrombolysis resistance ([Table 2]).
Table 2
Summary of potential drug targets for improving thrombolysis resistance
|
Target
|
Classification
|
Drugs
|
Mechanisms
|
Clinical trials
|
|
NETs
|
DNA degrader
|
Dornase Alfa
|
Recombinant human form of DNAse 1
|
NETS-target, NCT04785066; EXTEND-IA DNase, NCT05203224
|
|
PAD4 inhibitor
|
GSK484
|
Inhibits NETs formation and thrombosis
|
None
|
|
GSK199
|
Reduces NETs formation and infarct volume
|
None
|
|
nNIF
|
Reduces NETs formation and infarct volume
|
None
|
|
ROS inhibitor
|
Vitamin C
|
Reduces NETs formation
|
PSIOM, NCT03543917
|
|
Edaravone
|
Reduces BBB damage
|
TASPE, NCT06248242
|
|
vWF
|
vWF degrader
|
N-acetylcysteine
|
Cleaves disulfide bonds within vWF polymers
|
NAC-S, NCT04920448; NCT04918719
|
|
ADAMTS13
|
Specific VWF-cleaving metalloprotease
|
NCT02219035
|
|
RNA aptamer
|
BB-031
|
Inhibits vWF activity
|
RAISE, NCT06226805
|
|
Platelet
|
GPIIb/IIIa inhibitor
|
Eptifibatide
|
Prevents platelet aggregation
|
CLEAR-ER, NCT00894803; CLEAR-FDR, NCT01977456
|
|
Monoclonal antibody of GPVI
|
Glenzocimab
|
Inhibits platelet adhesion
|
ACTISAVE, NCT05070260; GALICE, NCT06437431
|
|
P2Y12 inhibitor
|
Cangrelor
|
Inhibits platelet aggregation and activation
|
REPERFUSE, NCT04667078
|
|
Fibrin
|
TAFI inhibitor
|
DS-1040b
|
Promotes fibrinolysis
|
ASSENT, NCT02586233; NCT03198715
|
|
PAI-1 modulator
|
THR-18
|
Promotes fibrinolysis
|
NCT01957774; NCT02572336
|
|
Monoclonal antibody of α2-AP
|
TS23
|
Inhibits endogenous plasminogen activator
|
SISTER, NCT05948566
|
Abbreviations: BBB, blood–brain barrier; NETs, neutrophil extracellular traps; PAI-1,
plasminogen activator inhibitor-1; TAFI, thrombin activatable fibrinolysis inhibitor;
vWF, von Willebrand factor; α2-AP, α2-antiplasmin.
Strategies Targeting NETs
Therapeutic strategies targeting NETs mainly involve inhibiting NETs formation or
promoting their degradation. The activation of PAD4 and the generation of ROS are
central to NETosis. Thus, inhibiting PAD4 and ROS production is key to suppressing
NETs formation. GSK484, a specific and reversible PAD4 inhibitor, has been shown to
inhibit thrombosis and alleviate inflammatory damage in preclinical models of subarachnoid
hemorrhage by blocking NETs formation.[109]
[110] Moreover, neonatal NET-inhibitory factor (nNIF) and GSK199, new PAD4 inhibitors,
have been demonstrated to reduce NETs levels in the plasma of tMCAO mouse models and
significantly decrease infarct volume, thereby improving stroke outcomes.[111] In the context of AIS, oxygen radical scavengers that target ROS, such as vitamin
C and edaravone, have shown promise in inhibiting NETs formation.[112] A recent meta-analysis demonstrated that vitamin C intervention reduced the risk
of stroke (RR = 0.77, 95% CI 0.70–0.85).[113] Furthermore, edaravone dexborneol (Eda.B), a formulation comprising edaravone (30 mg)
and (+)-borneol (7.5 mg), was approved by the National Medical Products Administration
of China in July 2020 for the clinical treatment of AIS. Studies have shown that Eda.B
can reduce NETs levels in serum samples of AIS patients and tissue samples from MCAO
mouse models, improve blood–brain barrier (BBB) permeability, and exert neuroprotective
effects.[114]
Pulmozyme (dornase alfa), a recombinant human deoxyribonuclease I (rhDNase), has received
FDA approval to treat cystic fibrosis (CF) clinically.[115] Dornase alfa reduces sputum viscosity by hydrolyzing extracellular DNA released
from degenerating neutrophils in the sputum of CF patients, thereby facilitating mucus
clearance.[116] Currently, dornase alfa is recommended only for nebulized inhalation. Under this
route, it acts primarily at the local pulmonary level, with minimal systemic absorption.
Its half-life is approximately 3 to 4 hours in the bronchi and 8 to 11 hours in the
lungs, with anticipated metabolism by proteases present in biological fluids.[117] According to the prescribing information of Pulmozyme, the elimination half-life
of dornase alfa in human plasma following intravenous injection is approximately 3
to 4 hours. Notably, there have been no reports of hemorrhagic complications associated
with dornase alfa treatment to date, which provides a safety foundation for its subsequent
clinical application in stroke.
Two ongoing phase II clinical trials are investigating its role in improving early
reperfusion rates in AIS. The NETs-target trial (NCT04785066) aims to evaluate the
efficacy of intravenous dornase alfa in improving vascular recanalization following
thrombectomy in AIS patients. The study plans to enroll adult stroke patients with
occlusions in the internal carotid artery, M1 segment, or M2 segment of the middle
cerebral artery. After receiving standard therapy (IVT and MT), participants will
receive adjunctive intravenous dornase alfa (specific dose has not been disclosed).
The EXTEND-IA DNase trial (NCT05203224) seeks to assess whether adjunctive intravenous
dornase alfa can improve early reperfusion in large vessel ischemic stroke, and to
explore the optimal single bolus dose of dornase alfa. Compared with NETs-target,
this study additionally enrolled patients with basilar artery occlusion. Following
IVT and/or MT, participants received a single intravenous bolus of dornase alfa at
escalating tiers (0.125, 0.25, 0.5, and 1 mg/kg). In both trials, the primary endpoint
is the achievement of significant recanalization on post-treatment angiography without
symptomatic intracranial hemorrhage. It is worth mentioning that no adverse events
have been reported to date in either trial. The findings of these investigations may
offer clinical evidence for the use of DNase I in AIS.
Thrombolytic Benefits of Targeting NETs
Since NETs play a crucial role in thrombolytic resistance, the potential advantages
of targeting NETs to improve thrombolytic treatment have been investigated in several
studies. Despite the fact that rt-PA by itself did not lower the amount of NETs in
the thrombus before thrombectomy,[118] DNase I can effectively degrade NETs by cleaving the DNA scaffold. As a result,
DNase I has been proposed as a promising adjuvant in thrombolytic therapy.
To simulate rt-PA-resistant thrombi, Peña-Martínez et al constructed a mouse photothrombotic
stroke model to generate fibrin-free thrombi composed primarily of aggregated platelets.
They found that DNase I administration alone was sufficient to recanalize occluded
vessels, improving stroke outcomes in experimental mice.[119] Additionally, Laridan et al demonstrated that combining DNase I with rt-PA enhanced
the lysis of retrieved AIS thrombi.[60] A recent study further showed that the combination of DNase I and rt-PA resulted
in a 3-fold increase in thrombolysis efficiency in contrast to rt-PA alone, and the
content of histones and DNA in thrombus was related to the lysis sensitivity of DNase
I.[120] Considering that erythrocyte-poor thrombi are often resistant to rt-PA, Vandelanotte
et al compared the effects of rt-PA combined with DNase I on AIS thrombi with varying
erythrocyte content. They found that DNase I overcame the rt-PA resistance of erythrocyte-poor
thrombi, but had no additional effect on erythrocyte-rich thrombi, which are already
susceptible to rt-PA.[98] Interestingly, it was reported that DNase I alone could not lyse ex vivo AIS thrombi.[61] A subsequent study contradicted this, showing that DNase I was even more effective
than rt-PA. This discrepancy may be attributed to two factors: one is the difference
in the type of thrombus used (i.e., cryopreserved versus fresh), and the other is
the variation in the duration of the ex vivo lysis tests (i.e., 1 hour vs. 4 hours).[119]
It is noteworthy that existing studies have demonstrated a synergistic interaction
between DNase I and plasmin. Napirei et al showed that DNase I exhibits high activity
against protein-free plasmid DNA but is nearly inactive against chromatin (DNA complexed
with histones and other proteins). However, upon the addition of serine proteases
like plasmin, DNase I can efficiently degrade chromatin.[121] The mechanism involves plasmin removing DNA-binding proteins (primarily histones),
thereby exposing the DNA strands and providing DNase I access to its substrate. Additionally,
DNase I possesses di-N-glycosylation, which allows it to remain stable and resist
proteolytic cleavage in the presence of plasmin. Desilles et al found that DNase I
enhances rt-PA-mediated thrombolysis at least partially by promoting fibrin degradation.[122] This suggests that while DNase I may exert a direct thrombolytic effect on thrombi
with high NETs burden, it also facilitates indirect fibrinolytic-dependent thrombolysis.
This effect is likely due to the degradation of the DNA backbone of NETs, which disrupts
thrombus stability and consequently enhances plasmin activity. In summary, the synergy
between DNase I and plasmin essentially follows a process of “removing protein barriers
first, then cleaving DNA, and finally enhancing plasmin activity.”
Neutrophil stasis in the cerebral capillaries has been demonstrated as a major cause
of no reflow after thrombolysis.[123] Notably, the aggregation of NETs has been observed in capillaries, potentially promoting
secondary microthrombosis and resulting in microcirculatory dysfunction.[17] Increased risk of hemorrhagic transformation is another defect of rt-PA, which may
be because rt-PA increases neutrophils recruitment and induces NETs formation by upregulating
LDL receptor-related protein 1 and PAD4, aggravating the inflammation and BBB damage
in the lesion.[124]
[125]
[126] Notably, NETs have been shown to activate the cGAS-STING pathway, causing type I
interferon to be produced, which exacerbates the BBB breakdown and cerebral bleeding
brought on by rt-PA. Importantly, targeting NETs with DNase I can reduce BBB damage,
improve vascular remodeling, and enhance microcirculation perfusion, thereby mitigating
rt-PA-associated cerebral hemorrhage complications.[125]
Other Strategies
In addition to platelet glycoprotein receptors, vWF within the thrombus may serve
as another potential therapeutic target. Two promising candidates—N-acetylcysteine
and ADAMTS13—have demonstrated significant thrombolytic efficacy in various experimental
thrombosis models without inducing an increased risk of hemorrhagic transformation.[127]
[128]
[129] Furthermore, targeting endogenous fibrinolytic inhibitors such as PAI-1, thrombin-activatable
fibrinolysis inhibitor (TAFI),[130]
[131] and α2-antiplasmin (α2-AP) may also help overcome rt-PA resistance.[132]
Current Limitations
DNase I holds significant potential for enhancing thrombolysis by degrading NETs.
This represents a promising therapeutic approach, particularly given the current limited
treatment options. However, it is important to note that the present study has several
limitations. First, the thrombi retrieved by MT do not represent the full spectrum
of AIS thrombi. Only those thrombi that are either resistant to spontaneous dissolution
or successfully retrieved after rt-PA treatment are included in research studies,
leaving out thrombi that may be sensitive to rt-PA or resistant to thrombectomy. This
limits the generalizability of findings to the broader population of AIS thrombi.
Second, most experimental studies have been conducted using in vitro AIS thrombi or
synthetic models. It remains unclear whether the mechanisms observed in these controlled
settings also apply to the complex in vivo environment of a human stroke. Future research
must elucidate the precise function of NETs in the pathological processes of AIS and
determine if the findings from laboratory models can be translated to clinical scenarios.
The efficacy and safety of DNase I combined with rt-PA to assist thrombolysis require
further clinical studies. The premise for combining DNase I with thrombolytic therapy
is that NETs have already formed within the therapeutic time window (4.5 hours). Cha
et al established a murine carotid artery occlusion model using FeCl3 and collected thrombi at 0.5, 1, 2, 3, 6, and 24 hours post-occlusion. Their findings
revealed that NETs within thrombi began to increase as early as 0.5 hour.[52] Similarly, Zhang et al enrolled 60 AIS patients who received intravenous thrombolysis
within 4.5 hours of symptom onset. Plasma samples collected before thrombolysis revealed
that NETs levels in AIS patients were significantly higher than in healthy controls.[107] Taken together, NETs form within 4.5 hours in at least a proportion of AIS patients.
Of course, it is also possible that NETs formation begins beyond 4.5 hours. In such
cases, early administration of DNase I could still exert beneficial effects during
thrombolysis. First, early co-administration of DNase I may prevent excessive accumulation
of NETs, thereby reducing thrombus stability. Second, as previously mentioned, early
use of DNase I can mitigate NETs-induced local inflammation and BBB disruption, improve
microcirculatory perfusion, and potentially reduce the risk of rt-PA–associated hemorrhagic
complications. Finally, DNase I has a relatively long plasma half-life, which further
increase the likelihood of early NETs targeting.
Concerns regarding the safety of DNase I primarily include the following points: First,
there is uncertainty about whether the degradation of NETs by DNase I, which releases
components such as DNA, NE, histones, and other procoagulant substances, might increase
the risk of thrombosis. Second, since NETs are part of the host's immune defense system,
reducing NETs may elevate the risk of infection in critically ill individuals. Finally,
it remains unclear whether DNase I could promote genomic instability in damaged neuronal
cells, potentially triggering carcinogenesis. Theoretically, exogenous recombinant
DNase I primarily degrades extracellular DNA and does not easily penetrate cell membranes
to enter the intracellular environment. Currently, there is no reliable evidence supporting
the notion that DNase I induces carcinogenesis. Therefore, long-term and systematic
safety monitoring is required in the future to further address these concerns.
Future Directions
Given that studies have reported higher neutrophil NETs content in cardioembolic thrombi
compared with other subtypes and reduced thrombolytic efficacy,[60]
[103]
[133]
[134]
[135] the NETs' content within thrombi has been proposed as an etiological classifier
and therapeutic response indicator. However, thrombus analysis is relatively complex
and time-consuming, and is only applicable to patients undergoing thrombectomy. Therefore,
the development of rapid and highly specific peripheral blood biomarkers of NETs is
essential, as this would assist neurologists in formulating more precise treatment
strategies.
Previous studies have demonstrated significantly elevated NETs levels in the plasma
of AIS patients.[107] Vallés et al reported that plasma levels of CitH3 and cell-free DNA (cfDNA) were
higher in patients with cardioembolic stroke compared with other subtypes.[18] Importantly, Genchi et al found a correlation between NETs content in thrombi and
NETs levels in plasma (r = 0.62, p ≤ 0.001).[133] Furthermore, Baumann et al observed a negative correlation between thrombus MPO
content and plasma MPO–histone complex concentrations (ρ = –0.237, p = 0.017), while a positive correlation was found between thrombus DNA–histone-1 complexes
and plasma DNase activity (ρ = 0.204; p = 0.037), which may reflect endogenous regulatory mechanisms.[136]
Nevertheless, caution should be exercised when interpreting these correlations, as
further validation in larger cohorts is required. To date, no gold-standard biomarker
for NETs has been established. Given the limitations of individual markers, for example,
CitH3 only detects PAD4-dependent NETosis and cfDNA lacks specificity,[137] we recommend a combined assessment of CitH3, MPO–DNA complexes, and cfDNA to evaluate
NETs levels.
Conclusion
Given the limitations of current IVT options for AIS, there is an urgent need to identify
novel therapeutic targets that can overcome rt-PA resistance mechanisms. NETs are
essential to the pathogenic progression of AIS and contribute to thrombolysis resistance
in thrombi. The combination of rt-PA and DNase I offers a promising strategy by simultaneously
degrading both fibrin and NETs, which improves the efficacy of IVT while reducing
the risk of intracranial hemorrhage. This approach holds significant clinical potential
for enhancing early reperfusion rates and improving long-term outcomes in AIS patients.
However, further clinical studies are necessary to evaluate the efficacy and safety
of combining DNase I with rt-PA in thrombolysis.
