C1-Inhibitor Protects From Brain Ischemia-Reperfusion Injury by Combined Antiinflammatory and Antithrombotic Mechanisms
Background and Purpose—Inflammation and thrombosis are pathophysiological hallmarks of ischemic stroke still unamenable to therapeutic interventions. The contact-kinin system represents an interface between inflammatory and thrombotic circuits and is involved in stroke development. C1-inhibitor counteracts activation of the contact-kinin system at multiple levels. We investigated the therapeutic potential of C1-inhibitor in models of ischemic stroke.
Methods—Male and female C57Bl/6 mice and rats of different ages were subjected to middle cerebral artery occlusion and treated with C1-inhibitor after 1 hour or 6 hours. Infarct volumes and functional outcomes were assessed between day 1 and day 7, and findings were validated by magnetic resonance imaging. Blood–brain barrier damage, thrombus formation, and the local inflammatory response were determined poststroke.
Results—Treatment with 15.0 U C1-inhibitor, but not 7.5 U, 1 hour after stroke reduced infarct volumes by ≈60% and improved clinical scores in mice of either sex on day 1. This protective effect was preserved at later stages of infarction as well as in elderly mice and in another species, ie, rats. Delayed C1-inhibitor treatment still improved clinical outcome. Blood–brain barrier damage, edema formation, and inflammation were significantly lower compared with controls. Moreover, C1-inhibitor showed strong antithrombotic effects.
Conclusions—C1-inhibitor is a multifaceted antiinflammatory and antithrombotic compound that protects from ischemic neurodegeneration in clinically meaningful settings.
- blood–brain barrier
- kallikrein-kinin system
- middle cerebral artery occlusion
The pathology of brain ischemia-reperfusion injury is complex and involves a myriad of distinct molecular and cellular pathways. Among these inflammation is one of the most relevant processes.1,2 Activation of the cerebral endothelium early after the ischemic event triggers upregulation of cellular adhesion molecules and successive trafficking of inflammatory cells (neutrophils, macrophages, T cells) from the circulation into the brain parenchyma. Those cells recruited from the periphery in concert with locally activated cell populations (endothelial cells, microglia, astrocytes) produce an array of highly active mediators such as cytokines and chemokines that perpetuate the inflammatory circuits, thereby causing direct or indirect tissue damage. Another characteristic of persisting ischemia is structural disintegration of the blood–brain barrier, which in consequence leads to the formation of brain edema.3,4 Excessive edema can harm otherwise healthy brain regions by mechanical compression and is a frequent cause of worsening of neurological symptoms in stroke patients. Until now, convincing pharmacological strategies to combat inflammation or edema formation in acute ischemic stroke are lacking.5
Current pathophysiological concepts also emphasize the importance of progressive thrombus formation in the cerebral microvasculature for secondary infarct growth.6 We could show recently that blocking of either platelet adhesion receptors or coagulation factors reliably protects from ischemic neurodegeneration.7–9 Most interestingly, there is increasing evidence of a tightly regulated interplay between thrombotic and inflammatory mechanisms during the course of an ischemic insult, and this thromboinflammation might be accessible to specific therapeutic interventions.10
The contact-kinin system constitutes a framework of serially connected serine proteases, namely coagulation factor XII (FXII), kininogen, and plasma kallikrein, and takes a central position in the pathophysiology of acute ischemic stroke.11 This system fosters vascular permeability and stroke-related inflammation by the formation of short-lived kinins while at the same time is linked to thrombus formation via the FXII-driven intrinsic coagulation cascade.12,13 Therefore, the contact-kinin system and its different molecular constituents represent a promising multifunctional target for potential stroke therapies.
C1-inhibitor is a 478 amino acid glycoprotein belonging to the superfamily of serine protease inhibitors called serpins.14 Its designation originates from the initial description as the only known physiological inhibitor of the classical complement pathway in blood and tissue. However, C1-inhibitor is also a major regulator of the contact-kinin system by blocking of activated FXII (FXIIa) and plasma kallikrein.14 Moreover, C1-inhibitor is known to directly interfere with the attraction of circulating leukocytes during inflammatory reactions and application of C1-inhibitor has been proven to be beneficial in ischemia-reperfusion injury in a variety of organs.15,16
We show that plasma-derived C1-inhibitor protects from reversible brain ischemia in mice and rats in several clinically relevant scenarios by a combined antiinflammatory and antithrombotic mode of action.
Materials and Methods
A detailed description of the surgical procedures, the stroke study population, and the methods is provided in the Online Data Supplement.
Three hundred sixty-nine C57Bl/6 mice (349 males, 20 females) and 33 male CD rats were included in the study, which was approved by institutional panels on animal care and governmental authorities (Regierung von Unterfranken, Würzburg, Germany; Regierungspräsidium Giessen, Germany). Focal cerebral ischemia was induced for 60 minutes (mice) or 90 minutes (rats) by transient middle cerebral artery occlusion (tMCAO) using the intraluminal filament technique.17,18 Animals were anesthetized with 2.5% isoflurane (Abbott) in a 70% N2O/30% O2 mixture. For permanent middle cerebral artery occlusion (pMCAO), the filament was left in situ. Mice were controlled for several physiological parameters that can critically affect stroke outcome (cerebral blood flow, blood pressure, heart rate, arterial blood gases; Supplemental Figures I, II, and Supplemental Table I). We calculated edema-corrected infarct volumes from coronal 2,3,5-triphenyltetrazolium chloride–stained brain slices. All stroke experiments were performed in accordance with the recently published ARRIVE guidelines (http://www.nc3rs.org/ARRIVE). Animals were randomly assigned to the treatment groups by an independent person not involved in data acquisition and analysis. We performed surgery and evaluation of stroke size, edema volume, thrombus formation, behavioral analysis including mortality assessment, and the extent of the local inflammatory response while being blinded to the experimental groups.
One hour or 6 hours after the induction of MCAO, mice received a single intravenous injection of human C1-inhibitor (Berinert-P; CSL Behring, GmbH) at a dose of 7.5 U or 15.0 U.19,20 In rats, C1-inhibitor was intravenously injected 90 minutes after the induction of tMCAO at a dose of 20 U/kg body weight. Control animals received equal volumes of isotonic saline (vehicle).
Functional Outcome Tests
We assessed the Bederson score21 and the grip test score22 on day 1 after MCAO to monitor global neurological function, motor function, and coordination in mice. In rats, a composite neuroscore23 was calculated. This functional assay is based on a battery of different neurological tests in which the sum of all subitems constitutes the total neurological score (maximum score is 28, with a score of 0 indicating no neurological deficit). In detail, the test battery evaluates: postural signs with “forelimb flexion” (degree of limb flexion when animal is held by tail; 0–2 points) and “thorax twisting” (degree of body rotation when animal is held by tail; 0–2 points); gait disturbances with “circling” (straight walking=0 points, walking toward contralateral side=1 point, alternate circling and walking straight=2 points, alternate circling and walking toward paretic side=3 points, circling and/or other gait disturbance [backing, crawling, walking on digits]=4 points, and constant circling toward paretic side=5 points), and “climbing” (ability to climb up an inclined board [45 deg]; 0–1 points); limb placing with “forelimb placing” (normal, weak, or no placing; 0–2 points) and “hind limb placing” (normal, weak, or no placing; 0–2 points); balance on a cylindrical beam (10 cm above the floor, 3 times): 0 or 1 times falling off the beam with or without attempt to stay on the beam=0 points, 2 times falling off the beam with or without attempt to stay on the beam=1 to 2 points, 3 times falling off the beam with or without attempt to stay on the beam=3 to 4 points; symmetry of muscle tone/strength with “lateral resistance” (degree of resistance against lateral push; 0–2 points) and “grasping strength” (symmetry of grasping strength onto wire cage; 0–1 points); sensory function with “grasping reflex of forepaw” (grasping onto tube when gently touched; 0–1 points) and “touching reflex” (withdrawal of forelimb when touched by needle; 0–1 points); and motility/spontaneous activity (1-minute observation): normal or slightly reduced exploratory behavior=0 to 1 points, moving limbs without proceeding=2 points, moving only to stimuli=3 points, unresponsive to stimuli with normal muscle tone=4 points, and severely decreased tone/premortal signs=5 points.
Magnetic Resonance Imaging
Serial stroke assessment by magnetic resonance imaging was performed in mice 24 hours and again 7 days after tMCAO on a 1.5-Tesla unit (Vision, Siemens) using T2-weighted imaging sequences and blood-sensitive T2-weighted gradient echo constructed interference in steady-state sequences as previously described.7,8
Real-Time Polymerase Chain Reaction Studies
We determined relative gene expression levels of endothelin-1, interleukin-1β, and tumor necrosis factor-α in the ischemic cortices and basal ganglia by real-time polymerase chain reaction (StepOnePlus Real-Time PCR System; Applied Biosystems) as described.12
Determination of Blood–Brain Barrier Leakage and Brain Edema
Blood–brain barrier leakage after tMCAO was quantified using the vascular tracer Evan's Blue (Sigma Aldrich) as described.12 The free water content of the brains (edema) was calculated from the brain wet/dry weights.12
Invading immune cells were detected by a rat antimouse Ly-6B.2 antibody (neutrophilic granulocytes; MCA771GA, 1:1000; AbD Serotec) and rat antimouse CD11b antibody (microglia/macrophages; MCA711, 1:100; AbD Serotec). For immunofluorescence staining against occludin, a rabbit antimouse occludin antibody (ab 31721, 1:100; Abcam) was applied. DNA stainings were performed using a fluorescent Hoechst dye (Hoechst 33342, 0.4 mg/mL; Sigma-Aldrich). Human C1-inhibitor was detected by a polyclonal sheep antibody against human C1-inhibitor (1:30 000; Abcam). For calculation of the thrombosis index, the whole brain was sliced 24 hours after tMCAO. Hematoxylin and eosin staining was performed according to standard procedures. The number of occluded blood vessels within the ischemic basal ganglia was counted in every tenth slice for control mice and mice treated with 7.5 U C1-inhibitor or 15.0 U C1-inhibitor, respectively, under 40-fold magnification (Axiophot2 microscope; Zeiss). Negative controls for all histological experiments included omission of primary or secondary antibody and produced no signals (not shown).
Immunoreactivity for fibrin(ogen) (anti-Fibrinogen pAb 1:500; Acris Antibodies)24 and occludin (antioccludin pAB 1:1000; Abcam) in the ischemic cortices and basal ganglia was detected and quantified by Western blot.
All results were expressed as mean±standard error of the mean except for ordinal functional outcome scales that were depicted as scatter plots including median with the 25% percentile and the 75% percentile given in brackets in the text. Numbers of animals (n=10) necessary to detect a standardized effect size on infarct volumes ≥0.2 (untreated mice vs mice treated with 15.0 U C1-inhibitor) were determined via a priori sample size calculation with the following assumptions: α=0.05, β=0.2, mean, and 20% standard deviation of the mean (GraphPad Stat Mate 2.0; GraphPad Software). For statistical analysis, the GraphPad Prism 5.0 software package was used. Data were tested for Gaussian distribution with the D‘Agostino and Pearson omnibus normality test and then analyzed by 1-way analysis of variance or in case of measuring the effects of 2 factors simultaneously 2-way analysis of variance with post hoc Bonferroni adjustment for probability values. Nonparametric functional outcome scores were compared by Kruskal-Wallis test with post hoc Dunn multiple comparison test. For comparison of survival curves, the log rank test was used. Rat data were compared by unpaired 2-tailed Student t test (stroke size, brain edema) or nonparametric Mann–Whitney test (functional scores). P<0.05 was considered statistically significant.
C1-Inhibitor Protects From Ischemic Brain Damage in Clinically Relevant Scenarios
To investigate the efficacy of plasma-derived C1-inhibitor in acute ischemic stroke, we chose a model of focal cerebral ischemia in which mice are subjected to tMCAO. This model induces a rapid and strong inflammatory response and massive edema formation within the brain and in addition depends on progressive microvascular thrombosis.7,25 First, 6-week-old male C57Bl/6 mice were subjected to tMCAO and treated with 7.5 U or 15.0 U C1-inhibitor 1 hour after stroke (Figure 1A). Infarct volumes on day 1 after tMCAO as assessed by staining of brain sections with 2,3,5-triphenyltetrazolium chloride were significantly smaller, by >60%, in male mice treated with 15.0 U C1-inhibitor than in vehicle-treated controls (mean 110.5±8.0 mm3 [control] vs 41.9± 8.4 mm3 [15.0 U], respectively; P<0.0001; Figure 1A). The smaller infarct volume was functionally relevant. Compared with control mice, mice receiving 15.0 U C1-inhibitor had significantly better overall neurological function 24 hours after tMCAO (Bederson score: median 3.0 [3.0–4.25] for control vs 1.0 [1.0–3.0] for 15.0 U, respectively; P<0.001) as well as improved motor function and coordination (grip test score: median 3.0 [3.0–4.0] for control vs 4.5 [4.0–5.0] for 15.0 U, respectively; P<0.001; Figure 1B). The 7.5-U C1-inhibitor failed to significantly lower infarct volumes (mean 110.5±8.0 mm3 [control] vs 81.3±10.4 mm3 [7.5 U], respectively; P>0.05) or improve functional outcomes (P>0.05; Figure 1A, B), indicating a dose-dependent effect of C1-inhibitor in brain ischemia.
Ischemic stroke usually is a disease of the elderly and, consequently, it is recommended to verify any stroke-protective effects observed in young laboratory animals in an older cohort, also.26,27 Adult middle-aged mice (age 6 months) also had development of significantly smaller brain infarctions (mean 95.9±4.9 mm3 [control] vs 44.5±5.9 mm3 [15.0 U], respectively; P<0.05) and less neurological deficits (Bederson score median: 3.5 [3.0–4.0] for control vs 3.0 [2.0–3.0] for 15.0 U, respectively; P<0.05; grip test score median 2.0 [0.0–2.25] for control vs 4.0 [3.5–4.0] for 15.0 U, respectively; P<0.0001) when treated with 15.0 U C1-inhibitor 1 hour after the onset of tMCAO, confirming our results in young animals (Figure 1A, B). We also tested old mice with an age of 12 months. Again, infarct volumes were significantly smaller (mean 152.7±10.0 mm3 [control] vs 72.4±21.6 mm3 [15.0 U], respectively; P<0.0001) and neurological deficits were less pronounced (grip test score median 1.0 [0.0–2.0] for control vs 4.0 [3.0–4.0] for 15.0 U, respectively; P<0.001) in the group treated with 15.0 U C1-inhibitor compared with vehicle-treated controls (Figure 1A, B).
Gender can significantly influence stroke outcome in rodents.27,28 Therefore, we also subjected 6-week-old female mice to 60 minutes tMCAO. In line with the results in male mice, C1-inhibitor–treated (15.0 U) female mice likewise had development of significantly smaller infarctions (mean 139.9± 8.1 mm3 [control] vs 87.3±9.6 mm3 [15.0 U], respectively; P<0.001) and less severe neurological deficits (Bederson score median 3.5 [3.0–4.0] for control vs 2.0 [1.75–3.0] for 15.0 U, respectively; P<0.001; grip test score median 3.0 [1.5–1.25] for control vs 4.0 [3.75–4.25] for 15.0 U, respectively; P<0.05) compared with female controls (Figure 1A, B).
Serial magnetic resonance imaging of living mice after tMCAO reaffirmed smaller infarctions on day 1 in animals receiving a single injection of C1-inhibitor (15.0 U) 1 hour after stroke (P<0.001; Figure 1C). Importantly, the infarct volumes did not increase between day 1 and day 7 (P>0.05), thus indicating that C1-inhibitor provides sustained protection against stroke. The alleged shrinkage in stroke size after 1 week in both groups was attributable to infarct maturation and subsequent fogging effects on magnetic resonance imaging rather than true infarct size reduction. Importantly, infarcts always appeared hyperintense on blood-sensitive constructed interference in steady-state sequences. Hypointense areas, which typically indicate intracerebral hemorrhage, were absent from C1-inhibitor–treated mice and vehicle-treated controls (Figure 1C). This finding excludes the possibility of increased bleeding complications caused by an excess of plasma C1-inhibitor.
We also determined the functional outcome and mortality of C1-inhibitor–treated mice and controls over a longer time period after ischemic stroke (Figure 1D, E). Seven days after 60 minutes of tMCAO, 9 out of 10 control mice (90%) had died, which is in line with previous reports.29 In contrast, 7 out of 10 mice (70%) treated with 7.5 U C1-inhibitor and 9 out of 10 mice (90%) treated with the higher dose of 15.0 U C1-inhibitor survived until day 7 (P=0.0087 or P=0.0215, respectively; Figure 1D). In line with these findings, mice receiving 15.0 U C1-inhibitor showed significantly better Bederson scores than controls at more advanced stages of infarct development, ie, on day 5 after tMCAO (Bederson score median 2.0 [2.0–3.0] for control vs 0.0 [0.0–1.0] for 15.0 U, respectively; P<0.001; Figure 1E). This observation excludes that C1-inhibitor simply induces faster recovery from stroke but underlines its sustained effect on functional outcome.
In an attempt to extend the therapeutic time window of exogenously applied C1-inhibitor, wild-type mice also received 7.5 U or 15.0 U C1-inhibitor in a delayed setting that is 6 hours after the induction of tMCAO. The higher dose of 15.0 U C1-inhibitor showed a tendency toward smaller infarct volumes, but differences compared with controls or mice receiving 7.5 U C1-inhibitor were not statistically significant (mean 113.9±7.8 mm3 [control] vs 102.4±9.9 mm3 [7.5 U] or 94.7±15.3 mm3 [15.0 U], respectively; P>0.05; Figure 2A). Notably, however, neurological dysfunction was still significantly less in the 15.0 U C1-inhibitor group compared with the 7.5 U C1-inhibitor group or control animals on day 1 (Bederson score median 3.0 [3.0–4.5] for control vs 3.0 [2.0–3.0] for 7.5 U or 1.0 [1.0–3.0] for 15.0 U, respectively; P<0.001 [control vs 15.0 U]; grip test score median 3.0 [0.0–3.0] for control vs 4.0 [3.0–4.0] for 7.5 U or 5.0 [3.0–5.0] for 15.0 U, respectively; P<0.001 [control vs 15.0 U]; Figure 2B).
According to the current experimental stroke guidelines,26,27 any protective effect requires evaluation in models of both transient and permanent ischemia. We therefore subjected C1-inhibitor–treated mice to filament-induced permanent MCAO, a procedure in which no tissue reperfusion occurs. In contrast to the striking effects observed after tMCAO, C1-inhibitor could not influence stroke size (P>0.05) or neurological outcome (P>0.05) 24 hours after permanent MCAO, irrespective of the dose applied (7.5 U or 15.0 U; Figure 2C, D).
To exclude that C1-inhibitor acts as a neuroprotectant exclusively in mice, we also validated our findings in rats. Rats treated with C1-inhibitor (20 U/kg) in a therapeutic setting, ie, 90 minutes after tMCAO, displayed massively reduced infarct size (mean stroked area: 23.8%±1.1% [control] vs 8.9%±1.4% [20 U/kg], respectively; P<0.0001) and improved neurological function (total neuroscore: median 13.0 [10.0–18.0] for control vs 9.0 [9.0–10.0] for 20 U/kg, respectively; P=0.0008; motility score: median 3.0 [2.0–4.0] for control vs 2.0 [2.0–2.0] for 20 U/kg, respectively; P=0.0004) after 24 hours compared with controls (Figure 3A, B). To assure that exogenously applied C1-inhibitor in the chosen concentration of 20 U/kg indeed reaches its target organ, we also investigated the localization of C1-inhibitor within the rat brain after tMCAO. After 30 minutes of reperfusion, a strong staining for human C1-inhibitor could be detected within the small vessels of the ischemic hemisphere but not of the contralateral side (Figure 3C), and this observation suggests an action of exogenous C1-inhibitor at the vascular endothelial compartment.20
Protection From Ischemic Stroke in C1-Inhibitor–Treated Mice Is a Result of Reduced Edema Formation, Inflammation, and Thrombosis
Next, we sought to elucidate the underlying mechanisms of this C1-inhibitor–specific neuroprotection in transient stroke. C1-inhibitor plays an important role in the regulation of vascular permeability and suppression of inflammation by inactivating key proteases of the contact-kinin system such as factor XIIa (FXIIa) or plasma kallikrein.14 Consequently, the extent of blood–brain barrier damage and edema formation in the ischemic hemispheres was addressed. On day 1 after tMCAO, the integrity of the blood–brain barrier as reflected by the volume of the vascular tracer Evan's Blue leaking into the brain parenchyma was preserved in mice treated with 15.0 U C1-inhibitor 1 hour after stroke and less pronounced after injection of 7.5 U C1-inhibitor in comparison with untreated controls (mean 51.6±10.8 mm3 [control] vs 33.1±7.9 mm3 [7.5 U] or 13.9±4.3 mm3 [15.0 U], respectively; P<0.05 [control vs 15.0 U]; Figure 4A). This finding correlated with dramatically less brain edema formation (wet/dry weight method) after therapeutic C1-inhibitor application (mean 4.3%±0.5% [control] vs 2.9%±0.4% [7.5 U] or 0.2%±0.4% [15.0 U], respectively; P<0.0001 [control vs 15.0 U]; Figure 4B), a result that also could be confirmed in rats (Supplemental Figure III). Importantly, almost no blood–brain barrier disruption was found in the brain regions (basal ganglia) where infarcts were regularly present in C1-inhibitor–treated mice (Figure 1A, 4A, red arrow). This indicates that the lesser edema seen in the C1-inhibitor group was a specific phenomenon and mechanistically relevant but not simply because of smaller infarct volumes in these animals.
We also analyzed the expression of endothelin-1 in the ischemic brains of C1-inhibitor–treated mice and controls. Endothelin-1 has been shown to be critically involved in the regulation of vascular integrity and edema formation under various pathophysiological conditions, including ischemic stroke.30,31 Twenty-four hours after tMCAO, endothelin-1 mRNA levels were significantly elevated in the cortices and basal ganglia of vehicle-treated mice and mice receiving 7.5 U C1-inhibitor compared with sham-operated mice (relative gene expression cortex: 1.0±0.1 [sham] vs 16.0±2.2 [control] or 15.6±2.3 [7.5 U], respectively; P<0.0001; relative gene expression basal ganglia: 1.0±0.1 [sham] vs 4.3±0.4 [control] or 4.2±0.6 [7.5 U], respectively; P<0.0001; Figure 4C). In contrast, no significant induction of endothelin-1 transcripts was observed in either brain region after treatment with 15.0 U C1-inhibitor (P>0.05). Again, endothelin-1 expression remained low in the basal ganglia after high-dose (15.0 U) C1-inhibitor treatment (Figure 4C), although significant parts of the basal ganglia were uniformly included into the infarcted areas in all animals (Figure 1A).
In line with a blood–brain barrier stabilizing effect of C1-inhibitor in stroke, immunoreactivity against the tight junction protein occludin was preserved in vessels of the ischemic basal ganglia from mice treated with 15.0 U C1-inhibitor but was downregulated in control mice or mice receiving 7.5 U C1-inhibitor as demonstrated by immunohistochemistry (Supplemental Figure IV). To quantify occludin protein expression in more detail, we also performed Western blot analysis (Figure 4D). Again, the amount of occludin on day 1 after tMCAO in the ischemic basal ganglia from untreated mice was low (optical density: 0.08±0.10). In contrast, significantly more occludin protein was detectable after treatment with 7.5 U (optical density 0.6±0.2; P<0.05) or 15.0 U (optical density 0.5±0.1; P<0.05) C1-inhibitor, respectively.
The C1-inhibitor has been shown to inhibit cell migration from the vasculature to inflammation sites by binding of cell adhesion molecules.32 We therefore quantified the numbers of immune cells invading the ischemic brain by immunohistochemistry (Figure 5A). Twenty-four hours after the induction of tMCAO, significantly more neutrophilic granulocytes (mean 299.1±52.2 [control] vs 107.2±44.7 [15.0 U]; P<0.05) as well as macrophages/microglia cells (mean 676.3±75.2 [control] v 117.1±32.5 [15.0 U]; P<0.0001) had entered the ischemic basal ganglia of untreated control mice than of mice that had been treated with 15.0 U C1-inhibitor 1 hour after stroke. In contrast, the lower dose of 7.5 U of C1-inhibitor was unable to reduce cell trafficking after focal cerebral ischemia (P>0.05; Figure 5A).
As a next step, we analyzed the gene expression profiles of the prototypic proinflammatory cytokines interleukin-1β and tumor necrosis factor-α in the brains of C1-inhibitor–treated mice and controls 24 hours after tMCAO (Figure 5B). Both cytokines are able to promote ischemic brain damage.1 The amount of interleukin-1β mRNA in the infarcted cortices was strongly elevated both in the untreated group as well as in the group receiving 7.5 U C1-inhibitor compared with sham-operated mice (P<0.0001). In contrast, no significant increase of interleukin-1β could be observed in the cortex after high-dose (15.0 U) C1-inhibitor treatment (P>0.05), and this dose also led to significantly lower interleukin-1β levels compared with untreated mice (P<0.001; Figure 5b). Similar results were obtained for tumor necrosis factor-α. Again, application of 15.0 U C1-inhibitor attenuated the increase of tumor necrosis factor-α expression observed in control mice or after treatment with 7.5 U C1-inhibitor (P<0.001, control vs 15.0 U).
The C1-inhibitor also acts on FXIIa, the prime activator of the intrinsic pathway of blood coagulation.14 Therefore, we analyzed the impact of C1-inhibitor on thrombotic activity after brain ischemia-reperfusion injury. The amount of fibrin(ogen) detected by immunoblot in the ischemic cortex (mean optical density: 2.8±0.5 [control] vs 1.7±0.4 [7.5 U] or 0.03±0.01 [15.0 U], respectively; P<0.0001 [control vs 15.0 U]) and basal ganglia (mean optical density: 2.8±0.5 [control] vs 1.2±0.3 [7.5 U] or 0.3±0.1 [15.0 U], respectively; P<0.001 [control vs 15.0 U]) was significantly reduced on day 1 after stroke after high-dose (15.0 U) C1-inhibitor application 1 hour after the induction of tMCAO (Figure 6A). The results from immunoblots also could be validated by immunohistochemistry showing less fibrin(ogen) deposits in the brain capillaries of C1-inhibitor–treated (15.0 U) mice (Supplemental Figure V). Accordingly, the microvascular patency was significantly increased compared with naïve controls (thrombosis index: 15.8±1.4 [control] vs 12.2±1.2 [7.5 U] or 9.8±1.4 [15.0 U], respectively; P<0.05 [control vs 15.0 U]; Figure 6B).
We show that plasma-derived C1-inhibitor protects from brain ischemia-reperfusion injury in rodents in different “clinically” meaningful settings. C1-inhibitor halved the infarct size in male mice when applied 1 hour after the onset of stroke and was still effective up to 6 hours. Its beneficial activity was preserved in female mice, at later stages of infarction, as well as in older cohorts. Moreover, C1-inhibitor also was effective in another species, ie, rats. Combined antiinflammatory, antiedematous, and antithrombotic modes of C1-inhibitor action could be identified as underlying mechanisms.
We and others recently could demonstrate that the contact-kinin system is critically involved in the pathology of ischemic stroke at different levels.33 Genetic disruption7 or pharmacological blocking34 of FXIIa, the common origin of the contact-kinin system and activator of the intrinsic coagulation cascade, led to near-resistance against ischemic neurodegeneration by preventing microvascular thrombosis. However, inhibition of another component of the contact-kinin system, the bradykinin receptor B1, which acts downstream of FXIIa and mediates the biological effects of kinins, also protected from ischemic stroke in this case by dramatically reducing edema formation and recruitment of inflammatory cells.12 This underpins the critical relevance of both thrombotic and inflammatory circuits for stroke development. The important crosstalk between thrombosis and inflammation in the context of brain ischemia, termed “thromboinflammation,” has only recently been recognized.10,35 With this respect, C1-inhibitor is of particular therapeutic attraction because it is able to mitigate clot formation by counteracting FXIIa activity as well as inflammation, eg, by inhibiting plasma kallikrein and, hence, the generation of kinins.36 Moreover, C1-inhibitor can bind selectins on endothelial cells that, in turn, are of great importance for the regulation of cell trafficking.32 The C1-inhibitor profoundly reduced intracerebral fibrin formation and kept inflammatory cells from entering the ischemic brain after tMCAO.
Whereas the anti-inflammatory properties of C1-inhibitor are well-established in models of ischemia-reperfusion injury of different organ systems, including the central nervous system,15,16,19,20,37,38 the present description of C1-inhibitor as powerful antithrombotic and antiedematous compound is novel and further adds to our understanding of this multifaceted molecule. Interestingly, C1-inhibitor could improve stroke outcome only after tMCAO but not permanent MCAO. Given the profound antithrombotic effect of C1-inhibitor after tMCAO, this observation is coherent because clot formation is unlikely to be influenced in the absence of tissue reperfusion, ie, in a situation when the occlusion of a large proximal brain vessel persists. Accordingly, C1-inhibitor treatment had no effect on fibrin(ogen) formation after permanent MCAO (not shown). Moreover, the finding in the permanent MCAO model mirrors our observations in FXII-deficient mice39 and indicates that blocking of the contact-kinin system after stroke is only promising if tissue reperfusion can be achieved in parallel.
It is noteworthy from a translational perspective that C1-inhibitor already is used in humans for many years in the therapy of C1-inhibitor deficiency (hereditary or acquired angioedema), so far without any major safety concerns,40,41 although it is obvious that compensation of naturally lacking C1-inhibitor in individuals with angioedema and raising of C1-inhibitor levels above the normal range in stroke patients reflect different conditions. Moreover, measuring C1-inhibitor plasma levels in mice and rats revealed that the terminal half-life is between 9.0 and 9.5 hours,42 whereas in healthy humans the C1-inhibitor half-life is ≈28 hours,36 indicating that C1-inhibitor doses used here cannot be directly compared with the human situation. Nevertheless, by using blood-sensitive magnetic resonance imaging we could at least rule out at an increased risk of infarct-related or reperfusion-related intracranial bleeding associated with the use of plasma C1-inhibitor, which corresponds to our previous experience after blocking FXIIa.7,34
We cannot exclude from our findings that suppression of the complement pathway contributed to the profound antiischemic effects of C1-inhibitor.14 The C1q−/− mice showed a trend toward smaller brain infarctions after tMCAO.43 However, in the same study, plasma C1-inhibitor was still able to significantly reduce infarct volumes in these animals, indicating that C1-inhibitor acts independently from the activation of the classical complement pathway.43
The C1-inhibitor formulation used in our experiments could positively influence neurological outcome, even when administered 6 hours after stroke, although beneficial effects on stroke size were no longer observable under these conditions. This mismatch is probably attributable to the relatively poor correlation between stroke volume and functional deficits in rodent models of ischemic stroke.8,44 However, because we could not observe reduced thrombosis or edema formation under this delayed treatment regime (not shown), additional mechanisms apart from those reported here probably underlie the neuroprotective effects of C1-inhibitor in focal cerebral ischemia that await clarification.
In contrast to our findings, another plasma-derived C1-inhibitor only showed a rather narrow therapeutic window after focal cerebral ischemia, ie, 30 minutes, but was ineffective already after 60 minutes.43 The exact reasons for this discrepancy are unclear at present but differences in the duration of brain ischemia (30 minutes vs 60 minutes) or the C1-inhibitor preparations used might play a role. This is further underlined by the fact that in our study 7.5 U of C1-inhibitor had only small effects on stroke outcome, whereas the minimal effective dose in mice was 5.0 U in the study by De Simoni et al.43
In summary, C1-inhibitor improves stroke outcome by interfering with key mechanisms of ischemic brain damage, namely thrombosis, edema formation, and inflammation. The fact that stroke protection could be achieved in different scenarios mimicking the clinical situation together with the multifaceted modes of C1-inhibitor action is promising and should be taken as a basis for further translational studies in relevant disease models.
Sources of Funding
This work was supported by the Deutsche Forschungsgemeinschaft (DFG), SFB 688 (TP A13 to Dr Kleinschnitz, TP B1 to Dr Stoll and Dr Nieswandt), the Wilhelm-Sander Stiftung (209.017.1 to Dr Kleinschnitz), and CSL Behring GmbH, Marburg, Germany.
Dr Nolte, Dr Hofmeister, and Dr Dickneite are employees of CSL Behring GmbH Marburg, Germany, and Dr Nolte and Dr Dickneite hold stocks in CSL Limited, Parkville, Australia. Dr Kleinschnitz received financial support for conducting this research project from CSL Behring GmbH Marburg, Germany. The other authors have nothing to disclose.
The authors are grateful to Daniela Urlaub, Andrea Sauer, Marianne Babl, and Virgil Michels (Würzburg), as well as Franz Kaspereit, Wilfried Krege, Gisela Birkenstock, Patrick Letmade, and Jennifer Krupka (CSL Behring GmbH, Marburg) for expert technical assistance.
The online-only Data Supplement is available at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112.660340/-/DC1.
- Received April 9, 2012.
- Accepted May 8, 2012.
- © 2012 American Heart Association, Inc.
- Bardutzky J,
- Schwab S
- Stoll G,
- Kleinschnitz C,
- Nieswandt B
- Kleinschnitz C,
- Stoll G,
- Bendszus M,
- Schuh K,
- Pauer HU,
- Burfeind P,
- et al
- Kleinschnitz C,
- Pozgajova M,
- Pham M,
- Bendszus M,
- Nieswandt B,
- Stoll G
- Austinat M,
- Braeuninger S,
- Pesquero JB,
- Brede M,
- Bader M,
- Stoll G,
- et al
- Horstick G,
- Heimann A,
- Götze O,
- Hafner G,
- Berg O,
- Böhmer P,
- et al
- Longa EZ,
- Weinstein PR,
- Carlson S,
- Cummins R
- Kleinschnitz C,
- Blecharz K,
- Kahles T,
- Schwarz T,
- Kraft P,
- Göbel K,
- et al
- Bederson JB,
- Pitts LH,
- Tsuji M,
- Nishimura MC,
- Davis RL,
- Bartkowski H
- Moran PM,
- Higgins LS,
- Cordell B,
- Moser PC
- Fisher M,
- Feuerstein G,
- Howells DW,
- Hurn PD,
- Kent TA,
- Savitz SI,
- et al
- Macleod MR,
- Fisher M,
- O'Collins V,
- Sena ES,
- Dirnagl U,
- Bath PM,
- et al
- Matsuo Y,
- Mihara Si,
- Ninomiya M,
- Fujimoto M
- Cai S,
- Davis AE III.
- Storini C,
- Bergamaschini L,
- Gesuete R,
- Rossi E,
- Maiocchi D,
- De Simoni MG
- Hagedorn I,
- Schmidbauer S,
- Pleines I,
- Kleinschnitz C,
- Kronthaler U,
- Stoll G,
- et al
- Caliezi C,
- Wuillemin WA,
- Zeerleder S,
- Redondo M,
- Eisele B,
- Hack CE
- Wahl F,
- Allix M,
- Plotkine M,
- Boulu RG