Keywords
polyanions - contact activation - NETs - polyP
Physiological Polyanions
Biological polyanions are highly abundant, negatively charged molecules that exist
ubiquitously in various forms in nature. For decades it has been established that
polyanions participate in blood coagulation and exert either procoagulant or anticoagulant
activities that contribute to normal hemostasis or pathological thrombosis, respectively.
Polyanionic carbohydrates, glycosaminoglycans (GAGs) provide anticoagulant activities
on the cell surface of most eukaryotic cells. Heparan-, chondroitin-, and dermatan-sulfate-type
GAGs interfere with clot formation at the interface of blood and vascular cells by
amplifying antithrombin and heparin cofactor II activities.[1] Mast cell-derived heparin shares structural similarity with heparan sulfate. Unfractionated
heparin, low-molecular weight heparin, and heparin-derived agents are commonly used
therapeutically as injectable anticoagulants.[2] In contrast to negatively charged polysaccharides, DNA, a key component of neutrophil
extracellular traps (NETs), and polyphosphate (polyP) have procoagulant activities
and promote blood clotting with implications for thrombosis ([Fig. 1]).
Fig. 1 Common mechanistic and structural features of NETs and platelet polyP. The lower
part of the left image shows a neutrophil that is releasing NETs (depicted by long,
dark blue DNA strands entangled with histones and other granular proteins) and the
right-hand side shows a platelet decorated with polyP on its surface. PolyP can be
composed of a few hundreds to thousands of phosphate units, which also make up the
phosphate backbone of DNA. The phosphate backbone serves as a structural support and
energy source for both of these molecules. NETs and polyP are polyanionic, immunomodulatory
structures that can activate platelets, FXII, and other factors of the contact pathway,
which will lead to further downstream events of the coagulation cascade. Eventually,
NETs and polyP can interact with fibrin and fibrinogen, reinforcing the fibrin meshwork.
C3/C5, complement components; FXII, factor XII; FXIIa, activated FXII; Mac-1: macrophage-1
antigen/CD11b/CD18; NET; neutrophil extracellular trap; polyP, polyphosphate; PSGL-1:
P-selectin glycoprotein ligand-1; TF: tissue factor; TFPI: tissue factor pathway inhibitor;
TLR: toll-like receptor; vWF, von Willebrand factor.
Extracellular DNA
Circulating extracellular DNA in human plasma was described as early as 1948.[3] In response to stimulation, an array of cells, including leukocytes, mast cells,
senescent cells, and tumor cells, release their DNA into the extracellular space either
as chromatin (histones complexed with DNA), naked double-stranded DNA (dsDNA), or
mitochondrial DNA.[4] In addition to actively released nucleic acids, DNA from disintegrating bacteria
and viruses is also detectable in circulation.[5] Small amounts of extracellular DNA are present in plasma and serum of healthy individuals;
however, levels are largely elevated in pathological conditions, suggesting the use
of extracellular DNA as a prognostic biomarker.[6]
[7]
[8]
[9]
[10]
Neutrophils are the predominant leukocyte in human blood and present the major source
of extracellular DNA. In response to various inflammatory stimuli, activated neutrophils
cast out their DNA, forming NETs. NETs were originally described as components of
the innate immune response to microbial infections that trap invading microorganisms,
thus interfering with pathogen dissemination. Furthermore, it has been shown that
inducers of phagocytosis trigger NET formation suggesting a cooperative effect of
NETosis and phagocytosis in host defense.[11] NETs are composed of long DNA strands that are bound to histones and neutrophil
granular-derived proteins.[12] DNA-intercalating dyes stain NETs; however, the signal is lost upon efficient digestion
by deoxyribonuclease-1 (DNase1) indicating that polyanionic dsDNA is the major component
of NETs.[13] Since their discovery, NETs have been implicated in a plethora of pathophysiologic
conditions offering a novel link between inflammation and thrombosis in an emerging
field in biomedicine.[14]
[15]
[16]
NET formation is a multistep process ([Table 1]). Upon neutrophil activation, nuclear chromatin starts to decondense, leading to
a loss of the typical lobulated morphology of the neutrophils' nucleus. During classical
NET formation, the multimeric NADPH-oxidase assembles on cellular membranes and produces
reactive oxygen species, which in turn activate the enzyme peptidyl-arginine deiminase
4 (PAD4).[17]
[18] PAD4 citrullinates histones, neutralizing their net positive charge and thus reducing
their affinity for binding to the negatively charged DNA polyanion, thereby facilitating
chromatin decondensation.[18] Additionally, neutrophil elastase (NE) and myeloperoxidase (MPO) are released from
neutrophil granules and translocate to the nucleus, where they degrade histones and
promote further unfolding of chromatin.[19] Consequently, the nuclear membrane breaks up, and chromatin is released into the
cytosol, where it binds to granular and cytosolic proteins. The mechanism by which
the plasma membrane ruptures to release NETs is not completely understood; however,
recent studies indicate that the pore-forming protein gasdermin D might play a role.[20]
Table 1
Formation, binding partners, cellular origin, detection, and degradation of neutrophil
extracellular traps and polyphosphate
|
Neutrophil extracellular traps
|
Polyphosphate
|
|
Formation
|
• Microbial/inflammatory stimuli such as LPS, TLRs, cytokine, Fc, or complement receptors[13]
[112]
[113]
[114]
• Synthetic compounds like phorbol myristate acetate (PMA), A23187, or ionomycin[114]
[115]
• Platelet neutrophil interaction[113]
|
• Polyphosphate kinase 1 and 2 (PPK1, PPK2) and homologs, e.g., DdPPK[71]
[116]
• Vacuolar transporter chaperone cleaves ATP γ-phosphate residues[117]
• Formation and secretion are induced by agonists such as thrombin, thrombin receptor-activating
peptide 6 (Trap6), collagen, and ADP[31]
|
|
Binding partners
|
• Binds platelets via glycoprotein Ibα, P-selectin, and high-mobility group box 1
(HMGB1)[62]
[118]
[119]
• Extracellular histones, predominantly H3 and H4, cause platelet aggregation[120] and induce platelets to secrete short-chain polyP from α-granules[65]
[94]
• DNA and histones individually promote thrombin generation; histones have shown to
do so in a polyP-dependent manner[49]
[65]
[94]
• NETs can bind and activate FXII, which then induces the activation of the kallikrein–kinin
system[61]
• Neutrophil elastase (NE) cleaves prothrombin, releasing small peptides that exert
antibacterial and immunomodulatory effects[121]
• TFPI and thrombomodulin can be inactivated by myeloperoxidase (MPO) and serine proteases[122]
• NE and cathepsin G contribute to fibrin formation on NETs, also by degradation of
TFPI[95]
• Intertwined fibrin–NET fibrils may reinforce NETs to prevent pathogen spread[123]
• NETs can amplify tissue factor[124]
[125]
• vWF binds to isolated DNA in vitro, potentially acting as a linker for leukocyte
adhesion to endothelial cells[126]
|
• Accelerates the generation of FXIa and thrombin[127]
• Amplifies thrombin-mediated activation of FXI[128]
• Accelerates FV activation by FXa and thrombin[128]
• Enhances the binding of platelets to von Willebrand factor[129]
• Activates FXII[127] thereby also triggering inflammation via FXIIa-mediated activation of the kallikrein–kinin
system[52]
• Inactivates TFPI, abrogating its anticoagulant function[130]
• Integrates into the fibrin clot, making it more resistant to fibrinolysis[131]
• Binds extracellular histones and activates platelets[132]
|
|
Cellular origin
|
• Leukocytes[13]
[133]
• Mast cells[134]
• Tumor cells[4]
|
• Ubiquitously found in various species including bacterial, plant, and mammalian
cells[21]
• Platelet dense granules[29]
• Mast cells[34]
• Astrocytes[135]
• Tumor cells[27]
|
|
Detection
|
• Microscopy[13]
• Flow cytometry[81]
• Flow chamber[113]
• ELISA[82]
• Western blotting[19]
• Sytox Green/PicoGreen staining[12]
|
• DAPI, Hoechst 33342, toluidine blue O, methylene blue, tetracycline, neutral red,
malachite green[85]
[86]
[87]
• Flow cytometry[79]
[87]
• Urea-polyacrylamide gel electrophoresis[84]
• Chromatography[84]
• 32P-NMR[84]
• Fourier transform-infrared (FT-IR)[84]
• Mass spectrometry[84]
• Microscopy[84]
|
|
Degradation
|
• Endonucleases DNase1 and DNase1 like-3[59]
• Human monocyte-derived macrophages and dendritic cells[69]
[70]
• Opsonization by complement factors[70]
|
• Endopolyphosphatases, e.g., Ppn1, Ppn2, Ddp1[74]
[75]
[136]
• Exopolyphosphatases, e.g., Ppx1[72]
[73]
• Diphosphoinositol polyP phosphohydrolases (DIPPs) may degrade polyP in mammals[76]
|
Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; ELISA, enzyme-linked immunosorbent
assay; LPS; lipopolysaccharides; NET; neutrophil extracellular trap; 32P-NMR, phosphorus-31 nuclear magnetic resonance; polyP, polyphosphate; TFPI, tissue
factor pathway inhibitor; TLR: toll-like receptor; vWF, von Willebrand factor.
The extracellular DNA forms supramolecular web-like structures both in the vasculature
and surrounding tissues.[17] High-resolution scanning electron microscopy (SEM) revealed that NETs are made of
fine thread-like structures, composed of long and sticky DNA strands.[13]
Polyphosphate
In contrast to NETs, polyP is a purely inorganic polymer composed of linear chains
of orthophosphates that are connected by energy-rich phosphoanhydride bonds ([Table 1]). PolyP is abundant in the environment, synthetic polyP is used in multiple technical
processes (e.g., as water softener, food ingredient, or fire extinguisher), and physiological
polyP is found in every cell in nature. The polymer is evolutionarily conserved among
bacterial, fungal, plant, and animal cells.[21] The high-energy phosphoanhydride bonds in the polyP chain are equivalent to those
in ATP and bacteria and yeast use the polymer as a chemical energy storage pool during
starvation and environmental stress.[22]
Prokaryotic and lower eukaryotic microorganisms have intracellular polyP molecules
ranging in chain length from a few hundreds to thousands of phosphate units. The polymer
is stored in subcellular organelles called acidocalcisomes, along with high concentrations
(in the molar range) of divalent metal cations, mostly Ca2+ but also Mg2+ and Zn2+.[21] Ca2+-ions bind to the phosphate units in the polyP backbone with high affinity. In vivo,
physiological polyP is complexed with metal ions. Ion-bound polyP has a different
structure and biophysical properties. Ca2+-polyP has little if any solubility in plasma, challenging the predictive value of
coagulation studies based on soluble synthetic polyP. Despite detailed information
on polyP metabolism in yeast, not much is known about polyP regulation in mammals.
The polymer is enriched in various subcellular compartments including lysosomes, mitochondria,
and nuclei; however, it is also found in association with the cytoskeleton and in
the cytoplasm. Some cells such as astrocytes, mast cells, tumor cells, and platelets
have the capacity to actively release polyP from secretory vesicles.[23]
[24]
[25]
[26]
[27]
[28] Platelet dense granules are specialized secretory organelles similar to acidocalcisomes
found in microorganisms. Dense granules appear as dark vesicles in electron microscopy
images because of their high local concentration of polyP (∼130mM) that is complexed
with Ca2+, Mg2+, and Zn2+ ions.[29]
[30]
PolyP in Blood Coagulation
PolyP in Blood Coagulation
Patients with a defect in platelet dense granules show significantly lower polyP levels
and defective factor XII (FXII)-dependent clotting in platelet-rich plasma (PRP; Hermansky–Pudlak
syndrome, delta storage pool diseases, Chediak–Higashi syndrome).[24]
[31]
[32] The addition of exogenous polyP to PRP of Hermansky–Pudlak syndrome patients restores
their clotting capacity, consistent with the notion that platelet polyP triggers coagulation
in a FXII-dependent manner.[31] Yeast cells lacking inositol hexakisphosphate kinase (a key enzyme involved in polyP
synthesis) are devoid of polyP. Consistently, platelets of inositol hexakisphosphate
kinase-1-deficient (Ip6k1−/−
) mice have reduced polyP levels, compromised FXII-triggered coagulation, and are
protected from platelet-driven lethal pulmonary embolisms.[33] In addition to defective polyP levels, Ip6k1−/−
mice show an array of other severe phenotypes including infertility and heart problems
making them a challenging model to study platelet polyP in vivo. Xenotropic and polytropic
retrovirus receptor 1 (XPR1) is a transmembrane protein that was originally described
as a cellular docking site for retroviruses but also functions as a phosphate exporter.
Recent systems biology-based studies have identified XPR1 as the major, if not exclusive,
phosphate exporter in platelets. Pharmacologic and genetic targeting of XPR1 activity
increased intracellular phosphate levels and led to polyP accumulation. Conditional
ablation of the Xpr1 gene in mouse platelets accelerated arterial thrombosis and activated platelet-driven
pulmonary embolism, but did not affect hemostasis.[34] The data identify XPR1 as the first specific regulator of polyP in platelets and
possibly other cells and indicate a fundamental role of phosphate metabolisms for
thromboembolic diseases.
For years, it was believed that platelets secrete soluble short-chain 50–100mer polyP
upon activation. However, this hypothesis was based on polymer purifications from
the supernatant of activated platelets using a phenol–chloroform extraction method
that selects for short-chain water soluble molecules.[31] Follow-up studies using anion-exchange isolation methods from complete cell lysates
confirmed the presence of small amounts of short-chain polyP in platelets. Additionally,
it was revealed that, similar to other mammalian cells, the vast majority of platelet
polyP consists of long-chain polymers. As platelets store polyP together with high
concentrations of Ca2+ ions in dense granules, the released polyP is complexed with calcium.[35] Ca2+–polyP has a very low solubility and readily precipitates into nanoparticles independent
of its chain length.[36] PolyP nanoparticles are stable in physiologic buffers for several hours.[37] Real-time imaging using polyP-specific probes showed that only minor portions of
the soluble polyanion fraction are released into the supernatant while the majority
remains anchored to the platelet plasma membrane.[38]
[39] Flow cytometry-based methods have been established to quantify polyP on the surface
of activated platelets suggesting a potential use of polyP as a biomarker in thrombotic
diseases.[40] Platelet-bound polyP nanoparticles drive coagulation in a FXII-dependent manner,
while soluble polymers have the capacity to drive other FXII-independent coagulation
reactions. Consistent with the notion that polyP operates by activating FXII, a series
of classical studies has shown the contribution of FXII in activated platelet-driven
coagulation/clot formation.[41]
[42]
[43]
[44] Vice versa, ablation of FXII or polyP impairs platelet-driven thrombosis in murine
models.[33]
[45]
[46]
NETs in Blood Coagulation
NETs in Blood Coagulation
Similar to polyP, multiple in vivo studies have shown NETs to be implicated in thrombotic
and inflammatory reactions.[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54] Following vascular injury, neutrophils immediately migrate to the lesion site preceding
platelets.[50] At the site of injury, activated platelets and endothelial cells activate neutrophils
to induce NET formation (NETosis). NETs in turn stimulate platelet aggregation and
trigger fibrin formation in vitro.[48] NETs are abundant in thrombi from experimental animal models and infusion of DNase
interferes with thrombus formation.[48]
[55] NETs are also enriched in venous and arterial thrombi of patients who suffered from
a heart attack,[56] stroke,[57] and peripheral vascular arterial occlusions.[58] Furthermore, it has been shown that NETs alone are sufficient for vascular occlusions
under septic conditions in the absence of host enzymes DNase1 and DNase1L3.[59] High levels of NET biomarkers, such as DNA/histones complexes, MPO, and S100A8/A9,
are detectable in plasma from patients with thrombotic microangiopathy, indicating
that the ineffective clearance of NETs contributes to the mechanisms of the occlusive
disease.[55]
[60]
Various components of NETs have been identified as initiators or propagators of coagulation
activity, including histones and granule proteins. Soluble DNA purified from neutrophils,
as well as NETs (induced by glucose oxide or interleukin-8 [IL-8] stimulation), can
assemble and activate FXII in vitro.[61] SEM of NETs induced by platelet-activated neutrophils showed that the DNA backbone
of NETs binds FXII and its substrate of the intrinsic coagulation pathway, factor
XI (FXI).[62] However, whether NETs directly trigger FXII contact activation or merely act as
a scaffold for the assembly of FXII activators and coagulation factors is still unclear.[63] Thrombin generation triggered by the addition of NETs is reduced in FXII- and FXI-deficient
plasma, indicating that the procoagulant activity of NETs is mediated by the FXII–FXI
axis at least in vitro.[49] Besides contact-activating FXII, DNA acts as a surface in thrombin-dependent FXI
activation.[64]
Many of the studies on the procoagulant nature of NETs examined purified NET DNA and
the various components of NETs individually, and thus the overall procoagulant activity
of NETs was largely omitted. Recently it was shown that human neutrophil-purified
DNA and recombinant histones H3 and H4 triggered coagulation in plasma individually,
whereas intact NETs did not. Histone–histone and histone–DNA interactions within the
nucleosome unit and supercoiled chromatin in NETs neutralize the negative charges
of the polyanion and thereby dampen the procoagulant activity of NET–DNA.[65] The precise mechanisms of NETs in thrombus formation are the subject of ongoing
studies; however, NETs appear to stimulate both platelets and the coagulation system.[49]
Degradation of NETs and polyP
Degradation of NETs and polyP
Despite their functional and structural similarities, the degradation pathways of
NETs and polyP seem to be quite different. Defective NET clearance triggers proinflammatory
and autoimmune conditions; however, the underlying mechanisms are still under investigation.
Degradation of NETs is an intricate process involving the activity of various enzymes.
While endonucleases efficiently degrade extracellular DNA, other NET components such
as histones and NE remain intact in a murine model of bacterial infection.[66] Although coagulation inhibitor activated protein C (APC) cleaves histones and interferes
with their cytotoxic activity in a purified system, it has no effect on NET-induced
cytotoxicity suggesting that histone-dependent cytotoxicity is protected from APC
degradation.[67]
[68]
Complete degradation of NETs in vivo requires the concerted activity of two secreted
host endonucleases, DNase1 and DNase1-like 3.[59] The reconstitution of either DNase1 or DNase1-like 3 was sufficient to rescue the
lethal phenotype of a chronic inflammation model in Dnase1
−/−/Dnase1-like3
−/− mice. In addition to extracellular DNases, the cytosolic exonuclease TREX1 (DNase
III) has the capacity of clearing NETs in vitro.[69] Furthermore, NETs can be engulfed by monocyte-derived macrophages and dendritic
cells in a cytochalasin D-dependent manner, implying a role of active endocytosis
in NET clearance.[69]
[70] Following internalization by macrophages, NETs are degraded in lysosomal compartments
in an immunologically silent manner. Hence, NET clearance does not evoke the release
of proinflammatory cytokines, maintaining homeostasis in tissues.[70]
In contrast to NETs, not much is known about polyP metabolism in mammalian systems.
In prokaryotes polyP is synthesized by polyP kinase (PPK), which reversibly transfers
γ-phosphate units from ATP and guanosine diphosphate onto the polymer chain.[22]
[71] Depolymerization of polyP into free Pi residues is catalyzed by exopolyphosphatase (Ppx).[72] Three distinct polyP phosphatases have been described in Saccharomyces cerevisiae: exopolyphosphatase [Ppx1], endopolyphosphatases [Ppn1], and diadenosine and diphosphoinositol
phosphohydrolase [Ddp1].[73]
[74]
[75] Mammalian homologs for these polyP phosphatases have not yet been identified; however,
diphosphoinositol polyP phosphohydrolases seem to participate in polyP degradation
under alkaline conditions.[75] Mammalian alkaline phosphatase (AP) from calf intestine is a potent exopolyphosphatase
and cleaves polyP.[76] Ca2+–polyP has a half-life in plasma of approximately 90 minutes, before it gets degraded
by polyphosphatases, such as AP. Exopolyphosphatase (Ppx1)-mediated degradation of
polyP improved cardiomyocyte function in cell culture[77] and alleviated Ca2+ accumulation in mitochondria and Ca2+-induced cell death processes related to myocardial infarction and ischemia-reperfusion
injury.
Extracellular RNA is considered to promote blood coagulation based on the fact that
infusion of RNase interferes with arterial thrombosis in a murine FeCl3-driven vascular injury model.[78] RNase readily hydrolyzes polyP, offering an alternative explanation for the thromboprotective
effects conferred by the enzyme.[79]
Detection of polyP and NETs
Detection of polyP and NETs
NETs and polyP are detected by similar dyes and techniques. Imaging of NETs in vitro
is mainly based on immunofluorescence microscopy, transmission electron microscopy,
and SEM. The DNA-intercalating dyes SYTOX Green/PicoGreen and 4′,6-diamidino-2-phenylindole
(DAPI), as well as antibodies against NET-specific structures such as citrullinated
histones (H3cit) and histone–MPO complexes are typically used for microscopy.[13] In recent years, the occasional bias in microscopic imaging of NETs has been criticized,
hence more automated software tools for image-based NET quantification are currently
being developed.[80] Granular proteins and other NET components can also be targeted with flow cytometry,
Western blotting, and enzyme-linked immunosorbent assays.[19]
[81]
[82] To further standardize quantification of NETs, especially in clinical settings,
the ISTH (International Society on Thrombosis and Haemostasis) Vascular Biology Subcommittee
has recently started a collaborative effort to investigate and harmonize NET quantification
techniques. Despite successful imaging of NETs in vitro, visualization of the polyanion
in vivo still poses significant hurdles. However, during the last few years, imaging
by intravital microscopy has strongly facilitated the in vivo evaluation of NET formation
and degradation.[64]
[83]
PolyP can be stained with dyes such as toluidine blue O or methylene blue to be visualized
by phase-contrast, bright-field, and electron microscopy.[79] Because polyP is mainly stored in membrane-enclosed compartments in eukaryotes,
the dyes neutral red and tetracycline can detect polyP with nondestructive methods,
such as light microscopy and flow cytometry.[84] Similarly to NETs, DNA-intercalating dyes such as DAPI[85] and Hoechst 33342[86] stain polyP. However, DAPI bound to polyP emits a bright yellow-green fluorescence,
distinct from the blue fluorescence emitted by DNA.[87] Toluidine blue O and DAPI only detect polyP with a chain length longer than 15-mers.[88]
[89] Sophisticated flow cytometry analyses using DAPI or tetracycline staining[87] or recombinant polyP-specific probes based on the polyP-binding domain of Escherichia coli exopolyphosphatase[40] are also being employed.
Malachite green dye binds free orthophosphates and can be used to quantify phosphate
monomers in solution.[79] Degradation of polyP with Ppx, allows for quantifying polyP concentration. The malachite
green assay fails to detect the chain length of the polyanion; however, it has a high
sensitivity and measures polyP up to the picomolar range. As Ppx only digests polyP
with a chain length of greater than 38 phosphate subunits, the malachite green assay
is insensitive for short-chain polyP. PolyP can furthermore be visualized by microscopy,[79] electrophoresis, chromatography, 32P-NMR, Fourier transform-infrared, and mass spectrometry.[84]
Crosstalk of NETs and polyP
Crosstalk of NETs and polyP
Inflammation and thrombosis are mediated by a complex interplay involving neutrophils
and platelets. During coagulation, FXII is activated by a unique mechanism triggered
by binding (“contact”) to negatively charged polyanionic surfaces (“contact activation”).
Activated FXII initiates the intrinsic pathway of coagulation and the bradykinin-producing
kallikrein–kinin system, leading to coagulation and inflammation.[90]
[91]
[92] Extracellular DNA and polyP activate FXII and promote thrombosis by the intrinsic
pathway of coagulation in vivo.[87]
[93]
[94]
While polyP initiates coagulation via FXII, NETs also contribute to the activity of
the tissue factor (TF)-driven extrinsic coagulation pathway. NET-associated TF and
granular protease NE and cathepsin G inhibit the TF pathway inhibitor (TFPI).[95]
[96] Activation of neutrophils with cytokines upregulates their TF mRNA expression and
TF deposition on NETs.[97] Furthermore, NETs contribute to mechanical clot stability by slowing down plasminogen–plasmin
conversion by tissue plasminogen activator (t-PA) on clot surfaces. They also bind
fibrin degradation peptides and delay their release from fibrin clots, as well as
intercalate into fibrin fibers and delay plasmin-mediated lysis of plasma clots.[98]
[99]
Neutrophils and platelets interact with each other via platelet glycoprotein Ibα binding
to neutrophil MAC-1 and platelet P-selectin binding to neutrophil P-selectin glycoprotein
ligand-1 (PSGL-1).[100]
[101]
[102]
[103] NETs promote coagulation in a platelet-dependent manner. High-resolution confocal
intravital microscopy revealed that NET-triggered coagulation is a result of collaborative
interaction between multiple components of NETs including DNA, histones and proteases
with platelets, and platelet polyPs. Histone H4 on NETs perforates platelets causing
the release of procoagulant polyP.[94] Neutralization of polyP with monoclonal blocking antibody (PP2055) significantly
reduced NET-initiated thrombin formation in an experimental sepsis model.[64] Furthermore, the procoagulant effect of NETs in PRP was attenuated by addition of
bovine AP, providing additional evidence that polyP plays a role in the procoagulant
activity of NETs.[49] In an acute ST-segment elevation myocardial infarction model, platelet polyP led
to NET formation by mTOR inhibition and autophagy induction. Treatment with IL-29
counteracted the effect of polyP on NET formation.[104] Together, these studies indicate that polyP interacts with NETs and that the polyP–NETs
crosstalk is important in coagulation.
Purified platelet and bacterial polyP exert high procoagulant activity even in the
presence of inhibitors of the TF-driven extrinsic pathway. However, in the absence
of FXII, polyP fails to trigger procoagulant activities.[31] Consistent with polyP activities in human plasma, infusion of the polyanion into
wild-type mice led to lethal pulmonary embolism, whereas FXII-deficient mice or mice
treated with a FXII inhibitor were protected from polyP-triggered thrombosis. FXII,
FXI, and FXII/FXI-double-deficient mice were similarly protected upon polyP-triggered
thrombosis, indicating that polyP operates via the classical intrinsic coagulation
pathway in vivo.[31]
Therapeutic Targeting of NETs and polyP in Thrombosis
Therapeutic Targeting of NETs and polyP in Thrombosis
NETs play a role in both arterial and venous thrombosis, making them an interesting
target to reduce thrombosis or stimulate thrombolysis. There is a multitude of NET
components (e.g., DNA, PAD4), cellular interactions (e.g., leukocyte–platelet/leukocyte–endothelium),
and signaling pathways (e.g., leukocyte recruitment, NET formation/degradation), that
are currently being investigated and that can be targeted pharmaceutically.[53] For instance, blocking platelet α-granules or Weibel–Palade body release would hamper
tethering of platelets and neutrophils to the vessel wall and also reduce leukocyte
and platelet recruitment upon activation.[105] Treating NET-containing thrombi from ischemic stroke patients ex vivo with t-PA
resulted in partial thrombus dissolution, which was significantly accelerated upon
the addition of DNase1.[105] A similar study showed that treating stroke thrombi with DNase1 alone does not efficiently
resolve the thrombi.[106] Thus, a combination treatment with a fibrinolytic agent, e.g., t-PA and/or ADAMTS13
(the protease specifically cleaving vWF), and a nuclease is recommended to obtain
a sufficient degree of thrombolysis. Besides the recombinant human DNase1 (Dornase
α, Pulmozyme, Roche), which is approved for the treatment of cystic fibrosis, there
are ongoing endeavors to develop improved NET-degrading nucleases.[59] Furthermore, ongoing preclinical studies investigate PAD4 inhibitors as potential
treatment options for multiple myeloma (BMS-P5, Bristol Myers Squibb[107]), rheumatoid arthritis, lung fibrosis, and thrombosis (preclinical PAD4 inhibitors
program, Jubilant Therapeutics).
Based on the structural homology of DNA and polyP, nucleic acid-binding polymers were
analyzed for interference with polyP-mediated coagulation.[108] Polyamidoamine dendrimer, 1,4-diaminobutane core, generation 3 (PAMAM G-3) was shown
to be the most effective polyP-binding molecule and reduced thrombus formation without
increasing the risk of bleeding in both the FeCl3-induced carotid artery injury and collagen/epinephrine-induced pulmonary thromboembolism
models. The notion that targeting polyP interferes with thrombosis while sparing hemostasis
confirms that polyP exerts its procoagulant activity via FXII. FXII is the only coagulation
factor that critically contributes to thrombosis but has no role in hemostatic mechanisms
(reviewed in Renné and Stavrou[109]). Due to concerns regarding the significant toxicities of anti-polyP agents including
dendrimers and other cationic small molecules,[110]
[111] a new nontoxic, thromboprotective dendrimer-like cationic polyP-blocking compound
class was introduced in 2016. Two of these novel universal heparin reversal agents
(UHRAs), UHRA-9 and -10, significantly reduced arterial thrombosis in vivo and did
not indicate any signs of fibrinogen aggregates, inflammation, tissue damage, or necrosis.
UHRA-9 and -10 also displayed a lower bleeding risk compared with therapeutic doses
of heparin. In a more specific approach, recombinant E. coli Ppx was shown to specifically bind and degrade polyP. Targeting polyP with Ppx abolished
polyP procoagulant activity in human plasma and in experimental thrombosis models
in vivo while sparing hemostasis, demonstrating that polyP is procoagulant in a FXII-dependent
manner in vivo.[79]
Summary and Conclusions
-
NETs and polyP are physiologic polyanions with potent procoagulant activity.
-
PolyP triggers coagulation by activating FXII, while both FXII- and TF-driven pathways
contribute to NET-stimulated coagulation.
-
The crosstalk between NETs and polyP plays an important role in coagulation and thrombosis.
-
PolyP forms Ca2+-rich nanoparticles independently of the polyanion chain lengths that are retained
on procoagulant platelet surfaces in vivo.
-
DNase1 digests NETs in vivo and provides a promising strategy to therapeutically target
NETs during thrombosis.
-
Cationic nucleic acid-binding molecules, recombinant exopolyphosphatase mutants, and
universal heparin reversal agent (UHRA) target polyP-driven thrombosis while sparing
hemostasis, indicating that polyP functions via FXII activation in vivo.
