Bosutinib Attenuates Inflammation via Inhibiting Salt-Inducible Kinases in Experimental Model of Intracerebral Hemorrhage on Mice
Background and Purpose—Intracerebral hemorrhage (ICH) is a subtype of stroke with highest mortality and morbidity. Pronounced inflammation plays a significant role in the development of the secondary brain injury after ICH. Recently, SIK-2 (salt-inducible kinase-2) was identified as an important component controlling inflammatory response. Here we sought to investigate the role of SIK-2 in post-ICH inflammation and potential protective effects of SIK-2 inhibition after ICH.
Methods—Two hundred and ninety-three male CD-1 mice were used. ICH was induced via injection of 30 μL of autologous blood. Recombinant SIK-2 was administrated 1 hour after ICH intracerebroventricularly. SIK-2 small interfering RNA was injected intracerebroventricularly 24 hours before ICH. Bosutinib, a clinically approved tyrosine kinase inhibitor with affinity to SIK-2, was given intranasally 1 hour or 6 hours after ICH. Effects of treatments were evaluated by neurological tests and brain water content calculation. Molecular pathways were investigated by Western blots and immunofluorescence studies.
Results—Endogenous SIK-2 was expressed in microglia and neurons. SIK-2 expression was reduced after ICH. Exogenous SIK-2 aggravated post-ICH inflammation, leading to brain edema and the neurobehavioral deficits. SIK-2 inhibition attenuated post-ICH inflammation, reducing brain edema and ameliorating neurological dysfunctions. Bosutinib inhibited SIK-2–attenuating ICH-induced brain damage. Protective effects of Bosutinib were mediated, at least partly, by CRTC3 (cyclic amp-response element binding protein-regulated transcription coactivator 3)/cyclic amp-response element binding protein/NF-κB (nuclear factor-κB) pathway.
Conclusions—SIK-2 participates in inflammation induction after ICH. SIK-2 inhibition via Bosutinib or small interfering RNA decreased inflammation, attenuating brain injury. SIK-2 effects are, at least partly, mediated by CRTC3-cyclic amp-response element binding protein-NF-κB signaling pathway.
Spontaneous intracerebral hemorrhage (ICH) accounts for 10% to 15% of all strokes and is associated with high mortality and morbidity. Despite intensive research, there is no effective treatment improving survival chance of ICH patients. There is mounting evidence demonstrating that microglia activation and, subsequently, releases of proinflammatory cytokines are one of the critical factors contributing to ICH-induced secondary brain injury. Therefore, examination of molecular pathways leading to post-ICH inflammation might be able to advance the development of new therapies for ICH. SIK (salt-inducible kinase) was originally identified as an enzyme induced in the adrenal glands of rats treated with a high-salt diet.1 SIKs belong to a family of AMP-activated protein kinases and consist of 3 isoforms.2 It has been demonstrated recently that SIK-2, but not other SIK isoforms, contributes to brain damage development after stroke.3 It has been shown that oxygen–glucose deprivation (an in-vitro model of ischemic stroke) induces degradation of SIK-2 without having the effect of the expression of another SIK isoform. Furthermore, the inhibition of SIK-2 during reperfusion was associated with increased neuronal survival. Finally, neurons from SIK-2 knockout animals were less sensitive to the oxygen–glucose deprivation, and SIK-2 knockout animals were more resistant to the focal cerebral ischemia induced by middle cerebral artery occlusion.
There are indications that SIK-2 is inhibited by a tyrosine kinase inhibitor, even though it is a serine/threonine kinase. SIK-2 inhibition can decrease inflammation, result in expression of macrophages with all characteristics of regulatory macrophages, producing anti-inflammatory cytokine and, on the other hand, decreasing proinflammatory cytokine production.4 Furthermore, in vitro screening of potential SIK-2 inhibitors revealed that an Food and Drug Administration (FDA)–approved Bosutinib can inhibit SIK-2 and that Bosutinib-induced SIK2 inhibition resulted in boost of IL-10 production in macrophages and dendritic cells.4,5 The study also revealed that SIK-2 inhibition leads to dephosphorization of CRTC-3 (cyclic amp-response element binding protein [CREB]-regulated transcription coactivator 3), inducing the translocation of CRTC-3 to the nucleus,4 which subsequently activates CREB. The CREB activation, in turn, induces the activation of anti-inflammation pathways. On the other hand, CREB activation increases the expression of NF-κB (nuclear factor-κB)–negative subunit, IκB, which inhibits the NF-κB–induced inflammation. This interaction explains the strong anti-inflammatory response induced by SIK-2 inhibit.6,7
The inflammation is a major pathophysiological factor of ICH, and post-ICH, neuronal death results in unresolved neurological deficits. It was, therefore, possible that the pharmacological manipulations on SIK-2 will provide vital information able to pave the new therapeutic approaches for ICH patient.
In the present study, we sought to investigate the potential role of SIK-2 in the rise of inflammation after ICH and explore the pathway underlying it. Most importantly, we intent to explore the potential therapeutic utility of FDA-approved tyrosine kinase inhibitor, Bosutinib, for treatment of ICH patients.
Materials and Methods
Animals and ICH Model
Two hundred and ninety-three 8-week-old male CD1 mice (weight=30 g; Charles River, Wilmington, MA) were housed in a 12-hour light/dark cycle at a controlled temperature and humidity with unlimited access to food and water. All experiments on animals in this study were approved by the institutional animal care and use committee at Loma Linda University.
ICH was induced by double infusion of autologous whole blood (30 μL) as described before. Briefly, mice were randomly assigned to the experimental groups, anesthetized with ketamine (100 mg/kg) and xylazine (10 mg/kg) (2:1 vol/vol, intraperitoneal injection), and positioned prone in a stereotactic head frame (Kopf Instruments, Tujunga, CA). Arterial blood was collected in a nonheparinized capillary tube, transferred into a Hamilton syringe, and with a microinjection pump 5 μL of blood was infused into the right basal ganglia at bregma anterior–posterior 0.2 mm, mediolateral 1.8 mm, and dorsoventral 3.0 mm. After 5 minutes, needle was advanced and 25 μL of blood was delivered at 3.5 mm dorsoventral, giving a total injection volume of 30 μL. Sham-operated animals were subjected to needle insertion only.
Following experiments were conducted.
Experiment I: To determine the time course of SIK-2 expression after ICH. The expression of SIK-2 was evaluated in ipsilateral hemisphere using Western blot analysis at 3, 6, 12, 24, 48, and 72 hours after ICH and compared with sham-operated animals. Spatial expression of SIK-2 was investigated by double immunostaining with ionized calcium binding adaptor molecule 1 (microglia), glial fibrillary acidic protein (astrocytes), and neuronal nuclei using sham and ICH (24 hours after surgery) groups.
Experiment II: To evaluate the role of SIK-2 and SIK-2 inhibition in the post-ICH brain injury. Mice were randomly divided into sham, ICH+vehicle, ICH+recombinant SIK-2 (rnSIK-2; Mybiosourse, species: mouse), ICH+SIK-2-siRNA (small interfering RNA), ICH+scramble RNA, ICH+Bosutinib group, ICH+Bosutinib+scramble RNA group, and ICH+Bosutinib+SIK-2-siRNA group. Vehicle (sterile saline) or rnSIK-2 (150 ng/5 μL of sterile saline) were administrated intracerebroventricularly (ICV) at 1 hour after ICH as previously described.8 Bosutinib (1, 5, and 25 mg/kg) was administrated intranasally at 1 hour or 6 hours after ICH as previously described.9 Scramble siRNA (500 pmol/μL) or SIK-2-siRNA (500 pmol/μL) was injected ICV 24 hours before ICH induction.9 SIK-2 and tumor necrosis factor-α (TNF-α) level were measured by Western blot 24 hours after ICH according to the vendor’s recommendation and as we did before.10 Neurological score (Garcia test) and brain water content were measured at 24 hours, 72 hours, or 14 days after ICH.10
Experiment III: To evaluate molecular pathway underlying effects observed in Experiment II. Mice were randomly divided into sham, ICH+vehicle, ICH+Bosutinib (the most effective dosage, experiment II) group, ICH+Bosutinib+scramble siRNA, and ICH+Bosutinib+CRTC3-siRNA group. CRTC3 siRNA was injected ICV 24 hours before ICH induction as described earlier.
Nuclear and cytoplasmic CRTC3 and NF-κB protein level was measured by Western blot and double immunostaining 24 hours after ICH. CREB/DNA binding was detected in a colorimetric assay. Neurological deficits and brain water content were measured at 24 hours after ICH.
Neurological scores were assessed by 2 trained investigators blinded to the animal groups, using a modified Garcia test as previously described.10,11 The Garcia scoring system consists of 6 subtests (spontaneous activity, spontaneous movements of all limbs, forelimbs outstretching, climbing ability, body perception, and response to vibrissae stimulation), with a maximum of 21 points being scored by animals without neurological deficits.
Measurement of Brain Water Content
The brain water content was measured using wet/dry method.12 Briefly, mice were decapitated under deep anesthesia. Brains were immediately removed and cut into 4 mm sections around the needle track. Each section was divided into 4 parts: ipsilateral and contralateral basal ganglia, ipsilateral and contralateral cortex. The cerebellum was collected as an internal control. Each part was weighed (wet weight [WW]) on an electronic analytic balance (APX-60; Denver Instrument) and then dried at 100°C for 24 hours to determine the dry weight (DW). Brain water content (%) was calculated as ([WW−DW]/WW)×100.
Twenty-four hours after ICH, mice were perfused under deep anesthesia with 40 mL of ice-cold PBS followed by perfusion with 40 mL formalin (10%). The brains were removed and fixed in formalin at 4°C for a minimum of 3 days. Samples were then dehydrated with 30% sucrose in PBS and sectioned with cryostat (CM3050S; Leica Microsystems) in 10 μm coronal slices. Immunofluorescence was performed as previously described. Coronal brain sections (10 μm) were permeabilized with 0.3% Triton X-100 in PBS for 30 minutes. The sections then were blocked with 5% donkey serum for 1 hour before being incubated at 4°C overnight with primary antibodies: anti-Iba-1 (1:100; Abcam, Cambridge, MA), anti-glial fibrillary acidic protein (1:100; Santa Cruz Biotechnology, Santa Cruz, CA), anti-neuronal nuclei (1:100; Abcam), anti-SIK-2 (1:50, Santa Cruz Biotechnology), and anti-CRTC3 (1:100, Abcam). The sections then were incubated with appropriate fluorescence-conjugated secondary antibodies (Jackson Immunoresearch, West Grove, PA) for 2 hours at room temperature followed by being visualized using a fluorescence microscope (Olympus OX51, Japan).
Mice were euthanized and perfused with 40 mL of ice-cold PBS. The brains were removed, separated into ipsilateral and contralateral hemisphere, and stored at –80°C until analysis. Western blotting was performed as previously described.12 Nuclear proteins were extracted by NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce Biotechnology, Rockford, IL) following the instruction manual. Primary antibodies used are listed as follow: anti-SIK-2 (1:100; Santa Cruz Biotechnology), anti-CRTC3 (1:200; Abcam), anti-CRTC3 (1:100; Abcam), anti-NF-κB p65 (1:200; Cell signaling, Danvers, MA), and anti-TNF-α (1:200; Santa Cruz Biotechnology).
All statistical analyses were performed using GraphPad Prism 5 (GraphPad software). Data are represented as a mean±standard deviation. One-way analysis of variance followed by Tukey multiple comparisons test was used for different groups’ comparison. Chi-square test was used for mortality analyses. P values <0.05 were considered statistically significant.
The total animal mortality rate of the study was 5.46% (16/293). No sham-operated animal died in this study. Neither autologous blood injection–induced ICH nor administration of different agents caused significant mortality.
Endogenous SIK-2 Was Downregulated After ICH Injury
ICH resulted in significant inhibition of SIK-2 expression as early as 3 hours after ICH. SIK-2 expression remained downregulated 24 hours after ICH. The recovery of the SIK-2 expression was observed at 72 hours after ICH (Figure 1A).
Double immunostaining revealed that SIK-2 was expressed in neurons and microglia of the perihemorrhagic area 6 hours after ICH. No astrocyte SIK-2 expression was observed (Figure 1B through 1D).
Exogenous SIK-2 Aggravated Inflammation and Brain Injury at 24 Hours After ICH
ICV administration of rnSIK-2 (150 ng/5 μL) resulted in increased concentration of the SIK-2 in the ipsilateral hemisphere evaluated by Western blot at 24 hours after ICH (Figure 2A). The increase of SIK-2 was accompanied with aggravation of post-ICH inflammation (TNF-α expression; Figure 2B). Increased inflammation consequently resulted in exacerbation of brain edema and neurological dysfunctions (Figure 2C and 2D).
Administration of SIK-2 siRNA Reduced Brain Injury at 24 Hours After ICH
ICV injection of SIK-2 siRNA significantly decreased SIK-2 expression in the ipsilateral hemisphere 24 hours after ICH (Figure 3A). Furthermore, SIK-2 in vivo knockdown (siRNA) was accompanied with a decrease of TNF production (Figure 3B), consequently resulting in reduced perihematoma brain edema and improved neurological functions (Figure 3C and 3D).
Bosutinib Attenuated Brain Injury and Inhibit SIK-2 Expression After ICH
Compared with sham operated, all ICH animals revealed a significant increase of brain water content in ipsilateral cortex and basal ganglia, as well as sever neurological deficits (Figure 4A and 4B). FDA-approved drug, Bosutinib, attenuated dose-dependently development of brain edema, improving post-ICH neurological functions at 24 hours after ICH (Figure 4A and 4B). The effective dose of Bosutinib significantly inhibited SIK-2 expression (Figure 4C), resulting in less post-ICH inflammation (Figure 4D). To investigate the optimal time for SIK inhibition after ICH, we have performed additional experiment and administrated Bosutinib intranasally 6 hours after ICH. The results revealed that intranasal Bosutinib administration 1 or 6 hours after ICV injection resulted in similar protection and attenuated brain edema and neurological deficits (Figure I in the online-only Data Supplement). We also performed experiment showing effects of Bosutinib given in conjunction with SIK-2 knockdown on brain edema and neurological functions. The results showed that Bosutinib given in conjunction with SIK-2 knockdown did not show any additional protection compared with Bosutinib administration only (Figure II in the online-only Data Supplement). This result indicated that Bosutinib had a protective effect mainly through inhibition of SIK-2 rather than multiple kinases.
Moreover, effective concentrations of Bosutinib attenuated development of ICH-induced brain edema evaluated 72 hours after ICH. Administration of 5 mg/kg of Bosutinib significantly improved neurological functions at this time point (Figure III in the online-only Data Supplement). Furthermore, we conducted the long-term study 14 days after ICH. At this time point, we observed a tendency for increased brain water content after ICH. The treatment with Bosutinib showed the tendency to decrease the post-ICH increase of brain water content and significantly improve the neurological functions (Figure IV in the online-only Data Supplement). These indicate that SIK inhibition has therapeutic value on both short and long term.
Bosutinib-Induced Protection Was Mediated by CRTC3/CREB Pathway
While ICH had no effect on the total CRTC3 expression (data not shown), ICH induced translocation of CRTC3 into the nucleus, resulting in the increased ratio of nuclear/cytoplasmic CRTC3 (Figure 5A). Bosutinib escalated the nuclear translocation of CRTC3, resulting in the further increase of the nuclear/cytoplasmic ratio in Bosutinib, compared with vehicle-treated animals evaluated by western blot (24 hours after ICH; Figure 5A). Immunostaining study confirmed results of Western blot and clearly demonstrated that Bosutinib treatment induced nucleus translocation of CRTC3 in microglia (Iba-1-positive cells; Figure 5B).
The nuclear translocation of CRTC3 resulted in CREB activation monitored by CREB ability to bind to the DNA via colorimetric assay. While ICH alone did not activate CREB, Bosutinib significantly activated CREB 24 hours after ICH (Figure 5C).
Proinflammatory effects were, furthermore, evaluated by monitoring of NF-κB p65 spatial expression. Compared with sham, ICH induced significant increase of nuclear portion of NF-κB-p65 24 hours after ICH. Bosutinib treatment attenuated this effect (Figure 5D).
CRTC3 In Vivo Knockdown Attenuated Protective Effects of Bosutinib
SiRNA-induced knockdown of CRTC3 resulted in a significant decrease of CRTC3 expression (Figure V in the online-only Data Supplement). In vivo knockdown of CRTC3 attenuated the protective effects of Bosutinib. Compared with scramble RNA, in vivo knockdown resulted in less CREB activation in Bosutinib-treated animals (Figure 6A), increasing nuclear accumulation of NF-κB-p65 (Figure 6B). Additionally, animals treated with a combination of Bosutinib/(CRTC3 siRNA) had higher brain water content of the ipsilateral hemisphere, consequently resulting in aggravation of neurological dysfunction (Figure 6C and 6D).
Inflammation is a major factor contributing to both high mortality and morbidity after ICH. In the current study, we investigated for the first time the role of SIK-2 in the rise of inflammation after ICH. We were able to demonstrate that SIK-2 was expressed in microglia and neurons and that the expression was transiently inhibited in ICH animals. SIK-2 inhibition via Bosutinib or in vivo knockdown of SIK-2 suppressed ICH-induced inflammation, resulting in fewer brain injuries and, consequently, in better neurological functions. Finally, we established that anti-inflammatory effects of SIK-2 inhibition were, at least partly, mediated via CRTC3-CREB-NF-κB pathway.
SIK was originally identified as a kinase induced in the adrenal glands of rats fed with a high-salt diet and shortly after it as a kinase induced by membrane depolarization of neurons.1,13 SIK belongs to an AMP-activated protein kinase family and plays a crucial role in the regulation of stress response.14 SIK consists of 3 isoforms, and potential participation of different isoforms in stress response is controversially discussed in the literature. There are indications that overexpression of SIK-3 and SIK-1 but not SIK-2 inhibits NF-κB activation in response to TLR4 stimulation and decreases production of proinflammatory cytokines.15 SIK-3, but not SIK-1 and SIK-2, deficiency is responsible for the profound inflammatory response on lipopolysaccharide.16 Although SIK-1 and SIK-3 inhibition seems to be proinflammatory, the inhibition of SIK-2, on the contrary, is beneficial and increases neuronal survival both in cell culture and in whole animals after ischemic stroke.3 Ischemic stroke induced SIK-2 degradation, while no effects of ischemia on another SIK isoforms were established in in vivo or in vitro part of the study.3 Furthermore, it has been demonstrated that inhibition of SIK-2 induced a switch of macrophages toward a regulatory phenotype, resulting in the decrease of proinflammatory cytokines release and increase of anti-inflammatory cytokines.4 Interesting enough, the anti-inflammatory effect of SIK-2 inhibition was lost on macrophages expressing mutated, drug-resistant form of SIK-2 isoform, indicating that the SIK-2 is a key molecular switch between pro- and anti-inflammatory macrophage phenotype.4 SIK-2 appeared to be a promising target, able to decrease inflammation and increase neuronal survival after ICH.
In our study, we demonstrated that ICH induced transient reduction of SIK-2 production as early as 3 hours after ICH. The production remained downregulated 24 hour and returned to the normal level at 72 hours after ICH. It was in agreement with previous reports, demonstrating in in vitro ischemic stroke model that SIK-2 was downregulated in early stage of reoxygenation without recovery during 24 hours of the reoxygenation.3 We have hypothesized the spontaneous downregulation of SIK-2 as a negative feedback to inflammatory factors increase in the early stage of ICH. However, the endogenous inhibition of SIK-2 is not sufficient to attenuate the neuroinflammation. Further inhibition of SIK-2 with exogenous inhibitors is, therefore, beneficial. Furthermore, we demonstrated that SIK-2 was expressed on neurons and microglia of both ICH and sham-operated animals. Neuronal SIK-2 expression was demonstrated before.3 We showed for the first time that SIK-2 is expressed on microglia as well. However, given the numerous demonstrations of SIK-2 expression on macrophage and the fact that microglia is resident macrophages of the central nervous system, this finding does not contradict previous publications.4,6 Although previous publication showed expression of SIK-2 on astrocytes, no SIK-2 expression on astrocytes was observed in our study. It is, however, worth to mention that the authors of previous study used cortical cell culture and Western blot as a detection method. Under these conditions, they could see a signal of SIK-2 on astrocytes, which was only a marginal compared with neuronal signal of SIK-2. We think that lack of SIK-2–positive astrocytes in the basal ganglia is not in direct conflict with previous study and is to explain with different experimental conditions.
To investigate the role of SIK-2 in the rise of inflammation after ICH, we administrated rnSIK-2 using ICV delivery. As expected, administration of rnSIK-2 significantly increased the concentration of the kinase in the brain. Giving that dramatic increase of TNF-α expression in post-ICH brain is well established event, the change of TNF-α expression was used to investigate the effect of SIK-2 on post-ICH inflammation.8 We demonstrated that the increase of SIK-2 was associated with significant upregulation of TNF-α production, clearly demonstrating the involvement of SIK-2 in an inflammatory response. TNF-α is able to directly disrupt blood–brain barrier, contributing to development of brain edema.17 Anti-TNF-α therapy improved recovery in murine model of ICH, indicating that TNF-α is a key factor for the development of brain injury after ICH.18,19 It is, therefore, not surprising that SIK-2–induced increase of TNF-α expression had an opposite effect and led to advanced brain edema in ICH animals treated with SIK-2 compared with the ICH alone and to more severe neurological deficits. We confirmed the finding by conducting in vitro knocking down of SIK-2 via siRNA.20 Well in agreement with previous experiment, knockdown of SIK-2 significantly decreases post-ICH production of TNF-α, resulting in attenuation of brain edema and improving neurological functions.
Targeting of specific protein by siRNA represents unquestionably promising treatment strategy, which, unfortunately, has only limited clinical relevance up to now.21 To increase translatability of our results into clinical practice, we used Bosutinib (an FDA-approved small molecular tyrosine kinase inhibitor) to inhibit SIK-2. Bosutinib was originally approved for treatment of chronic myelogenous leukemia.22 Recent publication demonstrated that Bosutinib can inhibit SIK by depriving tyrosine kinases of access to the Cdc37-Hsp90 molecular chaperone system on which they depend for their cellular stability, leading to their ubiquitylation and degradation.5 Furthermore, Bosutinib had beneficial effects in the model of neurodegenerative diseases, such as AD23. In our study, we first investigated the dose-dependent response of Bosutinib. Authors of previous publication demonstrated effectiveness of intraperitoneal delivery of the drug. However, they had applied Bosutinib for 6 weeks and used a model of chronic inflammation (Alzheimer’s disease).23 Because Bosutinib has a high molecular weight, for treatment of acute post-ICH inflammation, we choose an intranasal delivery of the drug. Intranasal delivery is a reliable way, allowing a quick delivery of the high molecular weight drug into brain.9,24 We demonstrated that Bosutinib dose-dependently decreased brain edema and improved neurological functions. According to our long-term study, Bosutinib can improve neurological functions by the time of 14 days after ICH, when brain edema and neuroinflammation secondary to ICH already become less predominant. It indicated that besides neuroinflammation, Bosutinib may also affect neuronal survival directly. The direct investigation into the treatment effects on the neuronal survival is a goal of our future study.
It has been revealed previously that SIK-2 inhibition leads to dephosphorization of CRTCs, including CRTC1, CRTC2, and CRTC3. While CRTC1 and CRTC2 are mainly expressed on neurons, CRTC3 is primarily expressed on microglia. In our current study, we focused on relationship between SIK-2 and inflammatory response, such as TNF-α production. Giving the factor that the inflammatory cytokines are mostly produced by microglia, we focused on SIK-2–induced CRTC3 regulation in the mechanism study. SIK-2 inhibition leads to dephosphorization of CRTC-3, inducing the translocation of CRTC-3 to the nucleus.4 The translocation activates CREB, promoting CREB-dependent gene transcription and inhibiting NF-κB.4,7,25,26 Well in agreement with these findings, we demonstrated that Bosutinib significantly increases the nuclear fraction of CRTC3, resulting in CREB activation. The activated CREB inhibited NF-κB and consequently decreased inflammation in brain of mice after ICH.
Taken together, our results indicated that SIK-2 contributed to inflammation after ICH, and this effect was mediated, at least partly, through CRTC3–CREB pathway. Most importantly, we demonstrated that Bosutinib inhibited SIK-2, decreasing inflammation and attenuating brain injury after ICH. Considering Bosutinib is an FDA-approved well-tolerated drug, our results might provide a clue to a new strategy for treatment of ICH patients.
Sources of Funding
This study was supported by Natural Science Foundation of China: 81601029 to Dr Ma, and National Institutes of Health NS078755, NS082124, and NS091042 to Dr Zhang.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.117.017681/-/DC1.
- Received December 8, 2016.
- Revision received August 21, 2017.
- Accepted August 23, 2017.
- © 2017 American Heart Association, Inc.
- Clark K,
- MacKenzie KF,
- Petkevicius K,
- Kristariyanto Y,
- Zhang J,
- Choi HG,
- et al
- Sundberg TB,
- Choi HG,
- Song JH,
- Russell CN,
- Hussain MM,
- Graham DB,
- et al
- MacKenzie KF,
- Clark K,
- Naqvi S,
- McGuire VA,
- Nöehren G,
- Kristariyanto Y,
- et al
- Henriksson E,
- Säll J,
- Gormand A,
- Wasserstrom S,
- Morrice NA,
- Fritzen AM,
- et al
- Yang P,
- Manaenko A,
- Xu F,
- Miao L,
- Wang G,
- Hu X,
- et al
- Zhang Y,
- Chen Y,
- Wu J,
- Manaenko A,
- Yang P,
- Tang J,
- et al
- Sanosaka M,
- Fujimoto M,
- Ohkawara T,
- Nagatake T,
- Itoh Y,
- Kagawa M,
- et al
- Zhou QH,
- Boado RJ,
- Hui EK,
- Lu JZ,
- Pardridge WM
- Zheng Y,
- Hu Q,
- Manaenko A,
- Zhang Y,
- Peng Y,
- Xu L,
- et al
- Fukuda AM,
- Badaut J
- Wu J,
- Zhang Y,
- Yang P,
- Enkhjargal B,
- Manaenko A,
- Tang J,
- et al
- Datusalia AK,
- Sharma SS