Isoflurane Attenuates Blood–Brain Barrier Disruption in Ipsilateral Hemisphere After Subarachnoid Hemorrhage in Mice
Background and Purpose—We examined effects of isoflurane, volatile anesthetics, on blood–brain barrier disruption in the endovascular perforation model of subarachnoid hemorrhage (SAH) in mice.
Methods—Animals were assigned to sham-operated, SAH+vehicle–air, SAH+1%, or 2% isoflurane groups. Neurobehavioral function, brain water content, Evans blue dye extravasation, and Western blotting for sphingosine kinases, occludin, claudin-5, junctional adhesion molecule, and vascular endothelial cadherin were evaluated at 24 hours post-SAH. Effects of sphingosine kinase (N,N-dimethylsphingosine) or sphingosine-1-phosphate receptor-1/3 (S1P1/3) inhibitors (VPC23019) on isoflurane's action were also examined.
Results—SAH aggravated neurological scores, brain edema, and blood–brain barrier permeability, which were prevented by 2% but not 1% isoflurane posttreatment. Two percent isoflurane increased sphingosine kinase-1 expression and prevented a post-SAH decrease in expressions of the blood–brain barrier-related proteins. Both N,N-dimethylsphingosine and VPC23019 abolished the beneficial effects of isoflurane.
Conclusions—Two percent isoflurane can suppress post-SAH blood–brain barrier disruption, which may be mediated by sphingosine kinase 1 expression and sphingosine-1-phosphate receptor-1/3 activation.
- blood–brain barrier
- early brain injury
- sphingosine kinase-1
- sphingosine-1-phosphate receptor
- subarachnoid hemorrhage
Blood–brain barrier (BBB) disruption has been an important prognostic factor after aneurysmal subarachnoid hemorrhage (SAH).1 The BBB is critical for brain homeostasis and is located at the cerebral microvessel endothelial cells, which maintain their barrier characteristics through cell–cell contacts made up of tight and adherens junctions.2 Stabilization of tight junctions involves a complex network of occludin, claudin-5, and junctional adhesion molecule.2 Adherens junctions consist of vascular endothelial (VE) cadherins.2
Recently, we reported that 2% isoflurane, a volatile anesthetic, prevented post-SAH neuronal apoptosis through sphingosine-related pathway activation.3 Sphingosine-1-phosphate (S1P) is generated from sphingomyelin by sphingosine kinase-1 (SphK1) and SphK24 and was reported to enhance endothelial barrier integrity.5 However, it remains undetermined whether isoflurane prevents BBB disruption. This study is the first to demonstrate that isoflurane posttreatment prevents BBB disruption after SAH in mice and that the mechanism involves SphK1 expression and S1P receptor-1/3 (S1P1/3) activation.
For expanded methods, see the online-only Data Supplement. The Loma Linda University animal care committee approved all protocols.
In Study 1, male CD-1 mice (30–38 g; Charles River, Wilmington, MA) were randomly divided into sham-operated+vehicle–air (n=17), SAH+vehicle–air (n=25), SAH+1% isoflurane (n=9), and SAH+2% isoflurane (n=22) groups. A SAH endovascular perforation model was produced and sham-operated mice underwent identical procedures except that the suture was withdrawn without puncture.3 One hour post-SAH, 1% or 2% isoflurane (Baxter, Deerfield, IL) was continuously administered for 1 hour with vehicle air (30% O2 and 70% medical air). All evaluations were blindly performed at 24 hours postsurgery. Eighteen-point SAH grading and 18-point neurological scores were evaluated in all surviving animals as previously described.3 Brain water content (n=6 per group) and Evans blue dye extravasation (n=5 per group) were measured as previously described.3 Western blot (n=6 per group) was performed on the left cerebral hemisphere (perforation side) using anti-SphK1 (Abgent, San Diego, CA), anti-SphK2 (Lifespan Biosciences, Seattle, CA), antioccludin, anticlaudin-5, antijunctional adhesion molecule-A, and anti-VE-cadherin (Santa Cruz Biotechnology, Santa Cruz, CA) antibodies as previously described.3
In Study 2, animals were randomly divided into dimethyl sulfoxide (a vehicle)+sham-operated+vehicle–air (n=11), dimethyl sulfoxide+SAH+2% isoflurane (n=15), N,N-dimethylsphingosine (a SphK antagonist; Enzo Life Sciences Inc, Plymouth Meeting, PA)+SAH+2% isoflurane (n=18), and VPC23019 (a S1P1/3-receptor antagonist; Avanti Polar Lipids Inc, Alabaster, AL)+SAH+2% isoflurane (n=18) groups. N,N-dimethylsphingosine (0.17 μg/0.5 μL) or VPC23019 (0.26 μg/0.5 μL) was infused into the right lateral ventricle at a rate of 0.1 μL/min 1 hour before surgery.3 The vehicle groups were given the same volume (0.5 μL) of dimethyl sulfoxide (1.1 g/mL/kg) diluted in phosphate-buffered saline. Isoflurane was administered like Study 1. SAH grading, neurological scores (all surviving animals), brain water content (n=6 per group), and Western blotting for SphK1, claudin-5, and VE-cadherin (n=5 per group) were performed at 24 hours postsurgery as described previously.
Data were expressed as median±25th to 75th percentiles or mean±SD and were analyzed using Kruskal-Wallis test followed by Steel-Dwass multiple comparisons, one-way analysis of variance with Tukey-Kramer post hoc tests, Fisher exact, or χ2 tests as appropriate. P<0.05 was considered statistically significant.
Isoflurane Prevents Post-SAH BBB Disruption (Study 1)
The mortality was not different among the SAH groups (vehicle–air, 32.0% [8 of 25 mice]; 1% isoflurane, 33.3% [3 of 9]; and 2% isoflurane, 22.7% [5 of 22]) at 24 hours. No sham-operated mice died. SAH grade was similar among the groups (Figure 1A).
Although 1% isoflurane had no significant effects, 2% isoflurane improved post-SAH neurological impairments (P=0.010), brain water content (P=0.003), and Evans blue dye extravasation (P=0.005) in the left cerebral hemisphere compared with the vehicle group (Figure 1B–D). Western blots showed that 2% isoflurane significantly increased SphK1, but not SphK2, in the left cerebral hemisphere compared with the sham (P=0.002) and vehicle (P<0.001) groups (online-only Data Supplement Figure I). In addition, 2% isoflurane increased expressions of occludin, junctional adhesion molecule-A, and VE-cadherin compared with the sham (P=0.001, respectively) and vehicle (P<0.001, respectively) groups and claudin-5 expression compared with the vehicle group (P=0.010; Figure 2).
SphK and S1P1/3-Receptor Antagonists Inhibit Isoflurane's Effects (Study 2)
No sham-operated mice died. The mortality was not different among the dimethyl sulfoxide+SAH+2% isoflurane (26.7%, 4 of 15 mice), N,N-dimethylsphingosine+SAH+2% isoflurane (38.9%, 7 of 18), and VPC23019+SAH+2% isoflurane (38.9%, 7 of 18) groups. SAH grade was similar among the groups (online-only Data Supplement Figure IIA).
Both N,N-dimethylsphingosine and VPC23019 significantly aggravated neurological scores (P<0.001 and P=0.003, respectively; online-only Data Supplement Figure IIB) and brain edema in the left cerebral hemisphere (P<0.002 and P=0.05, respectively; Figure 3A) and decreased expressions of SphK1, claudin-5, and VE-cadherin in 2% isoflurane-treated SAH mice (Figure 3B–D).
A key pathological manifestation of post-SAH early brain injury is BBB disruption.1 BBB dysfunction may allow greater influx of bloodborne cells and substances into brain parenchyma, thus amplifying inflammation, leading to further parenchymal damage and edema formation. In this study, 1-hour 2% isoflurane administration at 1 hour post-SAH improved neurological score, brain edema, and BBB permeability associated with increased SphK1 expression and S1P1/3 activation. Isoflurane also prevented a post-SAH decrease in expressions of tight junction (occludin, junctional adhesion molecule-A, and claudin-5) and adherens junction (VE-cadherin) proteins.
S1P is well known to decrease endothelial permeability.5–7 Vascular endothelial cells produce S1P, whereas they express S1P1, S1P2, and S1P3 with S1P1>S1P2 to S1P3.7 S1P was reported to induce the formation of tight junctions through S1P1.5 FTY720, a S1P-receptor modulator, accompanied by an increase in S1P1/5 and a decrease in S1P3/4, reversed BBB disruption in a rat model of encephalomyelitis.6 This study suggested that isoflurane induced SphK1 to synthesize S1P and that the activation of S1P1/3 was required for maintenance of post-SAH BBB function in mice. However, nothing is known on effects of S1P3 on endothelial permeability.7 Taken together, SphK1 and S1P1 activation may be key factors for isoflurane to induce S1P-mediated protection of post-SAH BBB.
Isoflurane anesthesia dose-dependently increases cerebral blood flow at the same time as decreasing metabolism8 and significantly increased BBB permeability associated with capillary dilatation at 3% in normal animals.9 Isoflurane inhibited neuronal injury dose-dependently, which was maximal at 2%.10 In this study, we tested 2 concentrations (1% and 2%) of isoflurane treatment because it is clinically relevant and demonstrated that only 2% isoflurane is protective against post-SAH BBB disruption in mice. Our data suggest that 2% isoflurane may work not only for anesthesia, but also for prevention of post-SAH BBB disruption through the sphingosine-related pathway.
This study has some limitations, including no studies of effects of isoflurane or the inhibitors on cerebral blood flow and sham-operated animals, comparisons of isoflurane's neuroprotective effects with other anesthetics as well as the detailed mechanisms how isoflurane protects or enhances BBB-related proteins. Thus, further studies are needed.
Sources of Funding
This study is partially supported by National Institutes of Health NS060936 to J.T. and NS053407 to J.H.Z.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112.661728/-/DC1.
- Received April 19, 2012.
- Revision received May 14, 2012.
- Accepted May 17, 2012.
- © 2012 American Heart Association, Inc.
- Brown RC,
- Davis TP
- Lee JF,
- Zeng Q,
- Ozaki H,
- Hand AR,
- Hla T,
- Wang E,
- et al
- Foster CA,
- Mechtcheriakova D,
- Storch MK,
- Balatoni B,
- Howard LM,
- Bornancin F,
- et al