Blockade of the MEK/ERK Pathway with a Raf Inhibitor Prevents Activation of Pro-Inflammatory Mediators in Cerebral Arteries and Reduction in Cerebral Blood Flow after Subarachnoid Hemorrhage in a Rat Model

Abstract

Cerebral ischemia that develops after subarachnoid hemorrhage (SAH) carries high morbidity and mortality. Inflammatory mediators are involved in the development of cerebral ischemia through activation of the mitogen-activated protein kinase pathway. We hypothesized that blockade of the MAPkinase/ERK (MEK)/extracellular signal-regulated kinase (ERK) pathway upstream with a specific raf inhibitor would prevent SAH-induced activation of the cerebrovascular inflammatory response. The raf inhibitor SB-386023-b was injected intracisternally in our rat model at 0, 6, or 12 hours after the SAH. After 48 hours, cerebral arteries were harvested, and iNOS, interleukin (IL)-6, IL-1β, matrix metalloproteinase (MMP)-9, tissue inhibitors of metalloproteinase (TIMP)-1, and phosphorylated ERK1/2 were investigated by immunofluorescence, real-time polymerase chain reaction (PCR), and Western blot analysis. Cerebral blood flow (CBF) was measured using autoradiography. Protein levels of MMP-9, TIMP-1, iNOS, IL-6, and IL-1β were increased after SAH, as were mRNA levels of IL-6, MMP-9, and TIMP-1. After SAH, pERK1/2 was increased, but CBF was reduced. Treatment with SB-386023-b at 0 or 6 hours after SAH normalized CBF and prevented SAH-induced upregulation of MMPs, pro-inflammatory cytokines, and pERK1/2 proteins. These results suggested that inhibition of MEK/ERK signal transduction by a specific raf inhibitor administered up to 6 hours after SAH normalized the expression of pro-inflammatory mediators and extracellular matrix-related genes.

Keywords

extracellular signal-regulated kinase 1/2 (ERK1/2)matrix metalloproteinase pro-inflammatory cytokines subarachnoid hemorrhage (SAH)tissue inhibitor of metalloproteinase 1 (TIMP-1)

Introduction

Subarachnoid hemorrhage (SAH) is the result of the rupture of a cerebral aneurysm and is frequently associated with intracranial hypertension, rebleeding, and cerebral ischemia. Late cerebral ischemia typically develops at 4 to 15 days after SAH in humans and is a major cause of human morbidity and mortality. Angiographic and cerebral blood flow (CBF) and metabolism studies have suggested that this phenomenon is more rapid in rodents (Delgado et al, 1985; Prunell et al, 2003). Despite intense research, the pathogenesis of late cerebral ischemia after SAH is still not well understood, and no specific pharmacological treatment is available.

Inflammatory reactions may have an important function in the development of late cerebral ischemia after SAH (Cahill et al, 2006; Dumont et al, 2003; Prunell et al, 2005) and correlate with the outcome of SAH (Gallia and Tamargo, 2006). There is a higher frequency of immune complexes with complement activation and increased levels of cytokines and endothelial adhesion molecules in the cerebrospinal fluid from patients experiencing SAH. The main inflammation mediators involved in the inflammatory response after SAH are cytokines (interleukin-1β (IL-1β), IL-6, and tumor necrosis factor-α), matrix metalloproteinases (MMPs), tissue inhibitors of metalloproteinases (TIMPs), and toxic molecules such as nitric oxide and free radicals (Baker et al, 2007; Cunningham et al, 2005; Kaminska, 2005; Vikman et al, 2007).

Several signal transduction pathways have been suggested to explain the activation of inflammatory mediators and late cerebral ischemia after SAH, such as mitogen-activated protein kinase (MAPK) and protein kinase C (Dumont et al, 2003; Kaminska, 2005). We have reported earlier that inflammatory pathways and genes regulating cytokine and metalloproteinase expression are transcribed through early activation of MAPK and MAPkinase/ERK (MEK)/ extracellular signal-regulated kinase (ERK) through specific transcription factors (Vikman et al, 2007). This study was designed to examine the hypothesis that inhibition of the raf-MEK-ERK1/2 pathway will reduce this response. Here, we reveal that administration of a specific raf inhibitor, SB-386023-b, as late as 6 hours after SAH prevents activation of the MEK/ERK pathway and the cerebrovascular inflammation response; this inhibition is in turn associated with a reduction in the development of late cerebral ischemia, as shown with quantitative measurements of CBF.

Materials and methods

Animals

All animal procedures were performed strictly within the national laws and guidelines and were approved by the Danish Animal Experimentation Inspectorate and the Ethics Committee for Laboratory Animal Experiments at the University of Lund.

Rat Subarachnoid Hemorrhage Model

The SAH was induced by a model originally devised by Svendgaard (Prunell et al, 2003). Anesthesia was induced in male Sprague–Dawley rats (350 to 400g) using 5% halothane (Halocarbon Laboratories, River Edge, NJ, USA) in H₂O/O₂ (30:70). The rats were intubated and artificially ventilated with inhalation of 0.5% to 1.5% halothane in N₂O/O₂ (70:30) during the surgical procedure. The depth of anesthesia was carefully monitored and the respiration was checked by regularly withdrawing arterial blood samples for blood gas analysis (Radiometer, Copenhagen, Denmark). A temperature probe was inserted into the rectum of each rat to record the body temperature, which was maintained at 37°C by a heating pad. An arterial catheter to measure blood pressure was placed in the tail artery, and a catheter to monitor intracranial pressure was placed in the subarachnoid space under the suboccipital membrane. At either side of the skull, 3 mm from the midline and 4 mm in front of the bregma, holes were drilled through the skull bone down to the dura mater (without perforation), allowing for the placement of two laser-Doppler flow probes to measure cortical CBF. Finally, a 27-G blunt cannula with a side hole was introduced 6.5 mm anterior to the bregma in the midline at an angle of 30° to the vertical (using a Kopf stereotaxic frame). With the aperture pointing to the right, the needle was lowered until the tip reached the skull base 2 to 3 mm anterior to the optic chiasm. After 30 minutes of equilibration of the animal, 250 μL blood was withdrawn from the tail catheter and injected through this cannula at a pressure equal to the mean arterial blood pressure (80 to 100 mmHg) (Ansar and Edvinsson, 2009).

Each rat was kept under anesthesia for another 60 minutes to allow recovery from the cerebral insult; after this period, catheters were removed and the incisions were closed. The rat was then revived and extubated. A subcutaneous injection of carprofen (4.0 mg/kg) (Pfizer, Copenhagen, Denmark) was administered, and the rat was hydrated subcutaneously using 40 mL isotonic sodium chloride at the end of the operation and on day 1. During the recovery period, the rat was monitored regularly, and if it showed severe distress, the animal was euthanized. Sham animals were injected with saline (250 μL) for 15 minutes to avoid any change in intracranial pressure (Ansar and Edvinsson, 2009), and after 2 days they were processed as described below.

Subarachnoid Hemorrhage Model with raf Inhibition

The SAH animals were treated with 20 μL of 10⁻⁶mol/L of SB-386023-b (a kind gift from Dr AA Parsons, GSK, UK) administered at (1) 0, 6, 12, 24, and 36 hours, (2) 6, 12, 24, and 36 hours, or (3) 12, 24, and 36 hours after the induced SAH. The dose selected was based on an earlier in vivo study with SAH (Beg et al, 2006). The results from the animals receiving the raf inhibitor initiated at 0 hours (data not shown) were no different from those for the animals receiving raf inhibitor starting at 6 hours after SAH.

Autoradiography Measurements of Regional Cerebral Blood Flow

Local CBF was measured using a model originally described by Sakurada et al (1978) and modified by Gjedde et al (1980). In brief, after 48 hours of observation, rats were anesthetized using 5% halothane in N₂O/O₂ (30:70). Each animal was intubated and artificially ventilated with inhalation of 0.5% to 1.5% halothane in N₂O/O₂ (70:30) during the surgical procedure. Anesthesia and respiration were monitored by regularly withdrawing arterial blood samples for blood gas analysis. A catheter to measure mean arterial blood pressure was placed in the right femoral artery, and a catheter for blood sampling was placed in the left femoral artery. This catheter was connected to a constant-velocity withdrawal pump (Harvard Apparatus 22, Boston, MA, USA) for mechanical integration of tracer concentration. In addition, a catheter was inserted into one femoral vein for injection of heparin and for infusion of the radioactive tracer. The mean arterial blood pressure was continuously monitored, and a temperature probe was inserted into the rectum to record the temperature, which was regularly maintained at 37°C. The hematocrit was measured by a hematocrit centrifuge (Beckman Microfuge 11, Brea, CA, USA).

After 30 minutes of equilibration, a bolus injection of 50 μCi of ¹⁴C-iodoantipyrine 4[N-methyl-¹⁴C] (Perkin-Elmer, Boston, MA, USA) was administered intravenously. Arterial blood (122 μL) was withdrawn over 20 seconds. Immediately after the blood was withdrawn, the animal was decapitated, and the brain was removed and immersed in isopentane (JT Baker, Deventer, Netherlands) chilled to −50°C.

The β-radioactivity scintillation counting was performed on the samples, with a program that included quench correction (Packard 2000 CA, Hvidovre, Denmark). The ¹⁴C activity in the tissue was determined after sectioning the brain into 20-μm sections at −20°C using a cryostat (Wild Leitz A/S, Glostrup, Denmark). The sections were exposed to X-ray films (Kodak, Copenhagen, Denmark) together with ¹⁴C methylmethacrylate standards (Amersham Life Science, Buckinghamshire, UK), and the films were exposed for 20 to 30 days. Densities of the autoradiograms were measured with a Macintosh computer equipped with an analog CF 4/1 camera (Kaiser, Germany) and a transparency flat viewer (Color-Control 5000, Weilheim, Germany). The ¹⁴C content was determined in several brain regions (Table 1). The CBF was calculated from the brain tissue ¹⁴C activity determined by autoradiography using the equation of Gjedde et al (1980).

Table 1

Regional cerebral blood flow 48 hours after SAH

	Control (n = 5)	SAH (n = 5)	SAH+SB-386023-b (n = 5)
Frontal cortex	144 ± 18^a	74 ± 12^a,b	130 ± 19^b
Parietal cortex	148 ± 14^a	74 ± 11^a,b	133 ± 24^b
Occipital cortex	154 ± 21^a	75 ± 18^a,b	143 ± 20^b
Caudate putamen	160 ± 20^a	74 ± 11^a,b	130 ± 25^b
Septum	109 ± 15^a	51 ± 10^a,b	104 ± 15^b
Hippocampus	116 ± 18^a	50 ± 8^a,b	87 ± 14^b
Thalamus	128 ± 13^a	68 ± 11^a,b	126 ± 18^b
Hypothalamus	85 ± 3	58 ± 11	99 ± 21
Corpus callosum	42 ± 12	22 ± 9	35 ± 15
Sensorimotor cortex	145 ± 27^a	68 ± 15^a,b	132 ± 19^b
Superior colliculus	126 ± 10^a	75 ± 16^a,b	125 ± 16^b
Inferior colliculus	146 ± 19^a	74 ± 12^a,b	136 ± 26^b
Cerebellar cortex	113 ± 12^a	58 ± 9^a,b	114 ± 23^b
Dentate nucleus	137 ± 19^a	63 ± 9^a,b	126 ± 21^b
Facial nucleus	168 ± 25^a	90 ± 16^a,b	156 ± 17^b
Cochlear nucleus	162 ± 27^a	64 ± 13^a,b	152 ± 13^b
Vestibular nucleus	176 ± 21^a	87 ± 16^a,b	164 ± 20^b
Trigeminal nucleus	147 ± 17^a	77 ± 14^a,b	151 ± 19^b

SAH, subarachnoid hemorrhage.

Values are expressed in mL per min per 100g and are given as mean ± s.e.m.

Significant difference between SAH and control groups.

Significant difference between SAH and SAH treated with SB-386023-b.

Harvest of Cerebral Arteries

After 48 hours of observation (sham, SAH, and SAH treated with SB-386023-b), rats were anesthetized using CO₂ and decapitated. The brains were quickly removed and chilled in ice-cold bicarbonate buffer solution (Ansar et al, 2007). The middle cerebral artery, the basilar artery (BA), and circle of Willis artery were carefully dissected free from the brain, cleared of connective tissue, and snap frozen at −80°C for immunohistochemistry, real-time polymerase chain reaction (PCR), and Western blot examination.

Immunohistochemistry

For immunohistochemistry, the indirect immunofluorescence method was used. The cerebral vessels were dissected out and then placed into Tissue TEK (Gibco, Invitrogen, Taastrup, Denmark) and frozen. They were then sectioned into slices 10-μm thick. The cerebral artery cryostat sections were fixed for 10 minutes in ice-cold acetone and then rehydrated in phosphate-buffered saline (PBS) containing 0.25% Triton X-100 for 15 minutes. The tissue was then permeabilized and blocked for 1 hour in blocking solution containing PBS, 0.25% Triton X-100, 1% BSA, and 5% normal donkey serum. The sections were incubated overnight at 4°C with the following primary antibodies: rabbit anti-rat inducible nitric oxide synthase (iNOS) (Abcam, Cambridge, UK; ab15326), rabbit anti-rat MMP-9 (Abcam, ab7299), rabbit anti-rat IL-6 (Abcam, ab6672), rabbit anti-rat IL-1β (Abcam, ab9787) diluted 1:400, rabbit anti-human TIMP-1 (AB770; Chemicon, Copenhagen, Denmark) diluted 1:200, and rabbit antiphospho-ERK 1/2 MAPK (Cell Signaling, Beverly, MA, USA; #4376) diluted 1:50. All dilutions were performed in PBS containing 0.25% Triton X-100, 1% BSA, and 2% normal donkey serum. Sections were subsequently washed with PBS and incubated with secondary antibody for 1 hour at room temperature.

The secondary antibody used was donkey anti-rabbit CY2 conjugate (Jackson ImmunoResearch, West Grove, PA, USA; 711-165-152) diluted 1:200 in PBS containing 0.25% Triton X-100 and 1% BSA. The sections were subsequently washed with PBS and mounted with PermaFluor mounting medium (Beckman Coulter, Brea, CA, USA). The same procedure was used for the negative controls, but primary antibodies were omitted. The immunoreactivity of the antibodies was visualized and photographed with a Leica confocal microscope (Solms, Germany) at the appropriate wavelengths.

Double Immunostaining

Double immunostaining was performed for IL-6, IL-1β, iNOS, MMP-9, and TIMP-1 proteins versus smooth muscle actin (expressed in the smooth muscle cells). The same antibodies were used as above, but mouse anti-rat smooth muscle actin antibody (Santa Cruz Biotechnologies, Santa Cruz, CA, USA; SC-53015) 1:200 was also used, diluted in PBS containing 0.3% Triton X-100, 1% BSA, and 2% normal donkey serum. The secondary antibodies used were donkey anti-rabbit Cy2- (Jackson Immuno Research, 711-165-152) 1:200, and donkey anti-mouse Texas Red (Jackson ImmunoResearch, 715-076-150) 1:250, diluted in PBS containing 3% Triton X-100 and 1% BSA. The antibodies were then detected at the appropriate wavelengths using a confocal microscope (EZ-cl, Germany).

Image Analysis

Fluorescence intensity was measured using ImageJ software (http://rsb.info.nih.gov/ij/). Measurements were made in four areas (located on the clock at 0, 3, 6, and 9 hours) from four vessel sections of each vessel sample; the investigator was masked to the treatment groups. The fluorescence intensity of each group is given as the percentage fluorescence in the SAH group compared with the sham group, wherein the sham group was set to 100%, and the mean value for each was used for comparisons.

RNA Isolation

Total cellular RNA was extracted from the BA, middle cerebral artery, and circle of Willis arteries using the Trizol RNA isolation kit (Invitrogen, Taastrup, Denmark) according to the supplier's instructions. Briefly, the arteries were homogenized in 1 mL of Trizol (Invitrogen) by using a Tissue Lyser (VWR International, Stockholm, Sweden). Subsequently, 200 μL of chloroform was added, and the samples were incubated at room temperature for 3 minutes, followed by centrifugation at 15,000g for 15 minutes at 4°C. The supernatant was collected and the organic phase was discarded. Then, 200 μL of chloroform was again added to remove all traces of phenol, and the samples were centrifuged at 15,000g for 15 minutes at 4°C. The aqueous supernatant was again collected. To precipitate the RNA, an equal amount of isopropanol was added and the samples were incubated overnight at −20°C.

Subsequently, the RNA was centrifuged at 15,000gfor 20 minutes at 4°C. The supernatant was discarded, and the resulting pellet was washed with 75% ethanol, air dried, and redissolved in diethylpyrocarbonate-treated water. Total RNAwas determined using a GeneQuant Pro spectrophotometer measuring absorbance at 260/280 nm (Amersham Pharmacia Biotech, Uppsala, Sweden).

Real-time Polymerase Chain Reaction

Reverse transcription of total RNA to cDNA was performed using the GeneAmp RNA kit (Perkin-Elmer Applied Biosystems, Foster City, CA, USA) in a Perkin-Elmer 2400 PCR machine at 42°C for 90 minutes and then at 72°C for 10 minutes. The real-time quantitative PCR was performed with the GeneAmp SYBR Green PCR kit (Perkin-Elmer Applied Biosystems) in a Perkin-Elmer real-time PCR machine (GeneAmp, 5700 sequence detection system). The above-synthesized cDNA was used as a template in a 25-μL reaction volume, and a blank control (without template) was included in all experiments. The system automatically monitors the binding of a fluorescent dye to double-stranded DNA by real-time detection of the fluorescence during each cycle of PCR amplification. Specific primers for the rat iNOS, IL-1β, IL-6, MMP-9, and TIMP-1 and the housekeeping genes, elongation factor-1 (EF-1), and β-actin, were designed by using Primer Express 2.0 software (PE Applied Biosystems, Foster City, CA, USA) and synthesized by TAG Copenhagen A/S (Copenhagen, Denmark). For the primer sequence, we refer to our earlier studies (Vikman et al, 2007).

The PCR reaction was performed as follows: 50°C for 2 minutes and 95°C for 10 minutes, followed by 40 PCR cycles with 95°C for 15 seconds and 60°C for 1 minute. Each sample was examined in triplicate. To verify that each primer pair generated only one PCR product of the expected size, a dissociation analysis was performed after each real-time PCR run. A blank control (without template) was used in all experiments. To confirm that the cDNAs of EF-1, β-actin, iNOS, IL-1β, IL-6, MMP-9, and TIMP-1 were amplified with a similar efficacy during real-time PCR, a standard curve was generated (Ansar et al, 2007). Standard curves for EF-1, β-actin, iNOS, IL-1β, IL-6, MMP-9, and TIMP-1 were performed by dilution of cDNA samples (1:10, 1:100, and 1:1000) (data not shown).

Western Blot Examination

Proximal BA, middle cerebral artery, and circle of Willis artery segments from SAH and sham animals (n = 12 rats in each group; vessels from three rats were pooled for each measurement) were harvested, frozen in liquid nitrogen, and homogenized in cell extract denaturing buffer (BioSource, Carlsbad, CA, USA) that contained both phosphatase and protease inhibitor cocktails (Sigma, St Louis, MO, USA). Whole-cell lysates were sonicated on ice for 2 minutes and centrifuged at 15,000g at 4°C for 30 minutes, and the supernatants were collected as protein samples. Protein concentrations were determined using standard protein assay reagents (Bio-Rad, Hercules, CA, USA) and stored at −80°C in preparation for immunoblot analysis.

The protein homogenates were diluted 1:1 (v/v) with 2 × sodium dodecyl sulfate sample buffer (Bio-Rad). Protein samples (25 to 50 μg) were boiled for 10 minutes in sodium dodecyl sulfate sample buffer and separated on 4% to 15% sodium dodecyl sulfate Ready Gel Precast Gels (Bio-Rad) for 120 minutes at 100 V and transferred to nitrocellulose membranes by electroblotting (Bio-Rad) at 100 V for 60 minutes. The membrane was then blocked for 1 hour at room temperature with PBS containing 0.1% Tween-20 (Sigma) and 5% nonfat dried milk and incubated with primary antibodies, as appropriate (rabbit anti-rat MMP-9 (ab7299; Abcam) diluted 1:200; rabbit anti-human TIMP-1 (AB770; Chemicon) diluted 1:200; rabbit anti-rat iNOS (Abcam; ab15326) diluted 1:200; rabbit anti-rat IL-6 (Abcam; ab6672) diluted 1:400; rabbit anti-rat IL-1β (Abcam; ab9787) diluted 1:200; and rabbit anti-β-actin (Cell Signaling Technology, Beverly, MA, USA) diluted 1:500 overnight at 4°C). This was followed by incubation with horseradish peroxidase-conjugated anti-rabbit IgG secondary antibodies (Amersham Biosciences, Piscataway, NJ, USA) diluted 1:5,000 for 1 hour at room temperature. The labeled proteins were developed using the LumiSensor Chemiluminescent Horseradish Peroxidase Substrate kit (GenScript, Piscataway, NJ, USA). The membranes were visualized using a Fujifilm LAS-1000 Luminescent Image Analyzer (Stamford, CT, USA), and band intensity was quantified using Image Gauge Version 4.0 (Fuji Photo Film, Stamford, CT, USA). Three independent experiments were performed in duplicate. The expression of target proteins was presented as relative to the level of β-actin normalized to the percentage of control.

Calculations and Statistics

Data are expressed as mean ± standard error of the mean (s.e.m.), and n refers to the number of rats. For the immunohistochemistry results, statistical analyses were performed with Kruskal–Wallis nonparametric tests with Dunn's post hoc tests, with P <0.05 considered to be significant. Western blot results were evaluated using two-tailed unpaired Student's t-tests, with P < 0.05 considered significant. Real-time PCR data were analyzed with the comparative cycle threshold method (Hansen-Schwartz et al, 2003). The cycle threshold values for EF-1, iNOS, IL-1β, IL-6, MMP-9, and TIMP-1 mRNA were used as a reference to quantify the relative amount of mRNA.

Results

Subarachnoid Hemorrhage Model

The mortality rate was 8% in this study of SAH, and there was no difference in the mortality rates among the various SAH groups. The rats showed no distressed behavior; they moved around, ate, and drank, and their fur was not disturbed. In all operated rats, mean arterial blood pressure (103 ± 3 mmHg), partial pCO₂ (39 ± 3 mm Hg), partial pO₂ (106 ± 4 mmHg), hematocrit (40 ± 1 mmHg) values, and body temperature were within acceptable limits during the operation. There was no statistical difference in physiologic parameters among the groups (sham, SAH, and SAH treated with SB-386023-b). As a result of injecting the blood, the cortical blood flow dropped over both hemispheres to 14% ± 5% of the resting flow and the intracranial pressure increased from 12 ± 2 to 121 ± 9 mmHg. The laser Doppler blood flow and the elevated intracranial pressure returned to basal values within 1 hour of postoperative monitoring (Ansar and Edvinsson, 2009). There was no difference in this response between the SAH groups; thus, SB-386023-b had no acute effect by itself.

Regional Cerebral Blood Flow to Evaluate the Overall Consequences of Subarachnoid Hemorrhage

There was a significant decrease in CBF as measured at 48 hours in the SAH group (63 ± 2 mL per 100 g per minute as compared with the saline control group (140 ± 6 mL per 100 g per minute; P < 0.05). Treatment with SB-386023-b with a start at 6 hours after the SAH (128 ± 4 mL per 100 g per minute) prevented the reduction in CBF seen after SAH (Figure 1). Animals from the SAH group showed a reduction in the regional CBF in 16 of the 18 brain regions examined as compared with the control (sham) group (Table 1). The same degree of effect was seen when the raf inhibitor was given at 0 hour, immediately after the induced SAH (data not shown, and in part similar data published before: Beg et al, 2006). Initiation of treatment with SB-386023-b at 12 hours after induction of SAH did not prevent this reduction in regional CBF (data not shown). There was no difference as compared with the control group for any of the regions studied.

Figure 1

Effect of treatment with a specific raf inhibitor SB-386023-b on the cerebral blood flow (CBF) given 6 hours after induced subarachnoid hemorrhage (SAH), compared with sham (control) and vehicle-treated SAH. Data are expressed as mean ± s.e.m. (n = 5 to 6) values, ***P≤0.001.

Western Blot Analysis

Protein levels of iNOS (174% ± 23%), IL-6 (295% ± 24%), IL-1β (285% ± 52%), MMP-9 (393% ± 70%), and TIMP-1 (199% ± 17%) were significantly increased after SAH as compared with the sham group (Figure 2).

Figure 2

Western blot showing inducible nitric oxide synthase (iNOS), interleukin (IL)-6, IL-1β, matrix metalloproteinase (MMP)-9, and tissue inhibitors of metalloproteinase (TIMP)-1 protein expression levels in cerebral arteries 48 hours after subarachnoid hemorrhage (SAH) as compared with the sham group. β-Actin was used as a loading control. Data are expressed as mean ± s.e.m., n = 4. *P < 0.05, **P < 0.01.

Immunohistochemistry

The iNOS, IL-6, IL-1β, MMP-9, TIMP-1, and p-ERK1/2 immunoreactivities were seen in the SMCs, and the expression of each was more abundant after SAH. Notably, they were observed to be equally increased in large cerebral arteries and in microvessels within the brain itself (Figure 3).

Figure 3

Sections from the basilar artery showing inducible nitric oxide synthase (iNOS), interleukin (IL)-6, IL-1β, matrix metalloproteinase (MMP)-9, and tissue inhibitors of metalloproteinase (TIMP)-1 immunoreactivity in the smooth muscle cell layer in sham, subarachnoid hemorrhage (SAH), SAH treated with SB-386023-b initiated at 6 hours, and SAH treated with SB-386023-b initiated at 12 hours. There were increased levels of iNOS, IL-6, IL-1β, MMP-9, and TIMP-1 proteins in the smooth muscle cell layer in SAH group as compared with sham group. Treatment with SB-386023 starting at 6 hours but not 12 hours prevented the increase in these proteins. iNOS, IL-1β, MMP-9, and TIMP-1 immunoreactivity are located in the SMC, whereas IL-6 immunoreactivity is seen both in the SMC and in the endothelial cells (white arrows). Data were obtained with confocal microscopy. Scale bar = 50 μm.

The effect of the raf inhibitor (SB-386023-b) on the protein expression of iNOS, IL-6, IL-1β, MMP-9, and TIMP-1 in the cerebral arteries was investigated at 48 hours after experimental SAH. The iNOS protein was expressed in the SMCs, and this signal was significantly increased in SAH (331% ± 39%, P < 0.05) as compared with the sham group (100% ± 3%). Similarly, IL-6 (231% ± 25%, P < 0.05) and IL-1β (193% ± 6%, P < 0.05) protein expressions were enhanced in SAH rats as compared with the sham group (100% ± 3% and 100% ± 14%, respectively). Treatment with the raf inhibitor SB-386023-b with initiation at 6 hours after SAH prevented the increase in iNOS (134% ± 15%, P < 0.05), IL-6 (117% ± 13%, P < 0.05), and IL-1β (104% ± 9% P < 0.05) protein expression in the SMCs as compared with the SAH, but not when treatment was initiated at 12 hours (iNOS (267% ± 11%), IL-6 (218% ± 8%), and IL-1β (165% ± 8%)) (Figure 3).

MMP-9 immunoreactivity was significantly increased in SAH (249% ± 39%, P < 0.05) as compared with the sham group (100% ± 2%). This outcome was seen both in large cerebral arteries and in microvessels. Treatment with the raf inhibitor (SB-386023-b) initiated at 6 hours prevented this upregulation (100% ± 12%, P < 0.05), but treatment initiated at 12 hours post-SAH did not (227% ± 9.1% as compared with SAH, P>0.05). The TIMP-1 was significantly increased in SAH (232% ± 12%) as compared with the sham group (100% ± 6%, P < 0.05). Treatment with the SB-386023-b given at 6 hours prevented this upregulation of TIMP-1 (135% ± 4.2%, P < 0.05) in the SMCs as compared with the SAH group, but treatment at 12 hours did not (196% ± 7% compared with the SAH group) (Figure 3). In addition, there was enhanced expression of pro-inflammatory and MMP proteins in the microvessels within the brain itself; this response was also prevented by the raf inhibitor SB-386023-b. As noted, administration of the inhibitor immediately after the SAH (0 hour) yielded results similar to those for administration at 6 hours (data not shown).

Colocalization with Actin Smooth Muscle Cells

The IL-6 and IL-1β proteins were localized in the cytoplasm of SMCs in the medial layer of the cerebral artery (colocalization with actin) (Figure 4). A weak IL-6 staining was seen in endothelial cells, but this did not change with SAH. The expression of iNOS was localized in both the cytoplasm and the nucleus of the SMCs (Figures 3 and 4).

Figure 4

Double-immunofluorescence staining for inducible nitric oxide synthase (iNOS), interleukin (IL)-6, IL-1β, matrix metalloproteinase (MMP)-9, or tissue inhibitors of metalloproteinase (TIMP)-1 (green) and actin in smooth muscle cells (red) of the basilar artery after subarachnoid hemorrhage (SAH). The third column shows their colocalization in smooth muscle cells as yellow fluorescence in the merged pictures. The enhanced expression of iNOS, IL-1β, and MMP-9 was localized in the smooth muscle cells (white arrows), whereas IL-6 was localized in both smooth muscle cells (white arrows) and endothelium cells (pink arrows). The TIMP-1 immunoreactivity was located in both SMC and adventitial layer (white arrows). Data were obtained with confocal microscopy. Scale bar = 50 μm. For color Figure please see the full text version.

The MMP-9 was expressed and located in the cytoplasm of the SMCs as verified by double immunostaining for MMP-9 and actin. The TIMP-1 expression was observed in the media layer but was mostly located in the adventitia layer of the cerebral vessel walls (Figure 4).

Extracellular Signal-Regulated Kinase

The pERK1/2 immunoreactivity was enhanced in the SMCs after SAH (188% ± 5%, P < 0.05) as compared with the sham group (100% ± 2%). Treatment with the specific raf inhibitor SB-386023-b with initiation at 6 hours after SAH prevented the pERK1/2 immunoreactivity (102% ± 4%, P < 0.05) but had no effect when the initial dose was given 12 hours after SAH (165% ± 3%) (Figure 5). The activation of ERK1/2 was observed both in large cerebral arteries and in microvessels within the brain parenchyma. We observed no significant activation of pERK1/2 in the brain tissue.

Figure 5

Sections from the basilar artery showing phosphorylated extracellular signal-regulated kinase 1/2 (ERK1/2), immunoreactivity in the smooth muscle cell layer in sham, subarachnoid hemorrhage (SAH), SAH treated with SB-386023-b starting at 6 hours, and SAH treated with SB-386023-b starting at 12 hours. There was an increased activation of phosphorylated ERK1/2 after SAH as compared with sham-operated rats. Treatment with SB-386023-b given at 6 hours after SAH prevented the increased protein expression. Data were obtained with confocal microscopy.

Real-time Polymerase Chain Reaction

The mRNA levels for middle cerebral artery, BA, and circle of Willis arteries extracted at 48 hours after SAH and after treatment were compared by using two endogenous standards, EF-1 and β-actin. The expression of EF-1 and β-actin was used to normalize the gene expression, and both genes gave the same results after normalization of the real-time PCR data (Stenman and Edvinsson, 2004).

The standard curves for each primer pair had similar slopes, demonstrating that EF-1, β-actin, iNOS, IL-1β, IL-6, MMP-9, and TIMP-1 cDNAs were amplified with the same efficiency. In each PCR experiment, a no-template control was included, and there were no signs of contaminating nucleic acids. As the results from the three types of brain vessels were identical, the groups of vessels (middle cerebral artery, BA, and circle of Willis arteries, from 5 to 7 rats per group) were merged for the final statistical analysis. There was significant upregulation of MMP-9 and TIMP-1 mRNA (P < 0.05) relative to the amount of EF-1 in vessels from the SAH group as compared with the sham group. Treatment with SB-386023-b starting at 6 hours after SAH abolished the upregulation of these mRNA levels (Figure 6), but an initial administration at 12 hours did not reduce it (data not shown). The IL-6 mRNA was increased in the SAH group, but this increase was not significant. The mRNA levels of both iNOS and IL-1β were lower in the SAH group as compared with the sham group, and these levels remained unchanged in the treated groups as compared with the sham group (Figure 6).

Figure 6

Bar graphs showing the mRNA levels of inducible nitric oxide synthase (iNOS) (A), interleukin (IL)-1β (B), IL-6 (C), matrix metalloproteinase (MMP)-9 (D), and tissue inhibitors of metalloproteinase (TIMP)-1 (E) genes after subarachnoid hemorrhage (SAH) and treatment with SB-386023-b given at 6 hours. Data were obtained by real-time polymerase chain reaction (PCR) and are expressed as means ± s.e.m. Values are expressed relative to elongation factor-1 (EF-1) mRNA levels for 7 to 9 animals per group. *P≤0.05.

Discussion

We have for the first time identified a transcriptional upregulation of inflammatory pathways (genes for cytokines and metalloproteinases) in cerebral arteries and microvessels after SAH associated with activation of the MEK/ERK1/2 pathway. Specific blockade of phosphorylation of this pathway by an upstream raf inhibitor abolished the upregulation of the expression of the cytokines, MMP-9, and TIMP-1 in large cerebral arteries and in microvessels even if the inhibitor was given as late as 6 hours after the induced SAH. Several inflammatory processes have been proposed to be involved in cerebral ischemia after SAH; here, a specific raf inhibition aborted or decreased this inflammatory process after cerebral ischemia.

As shown here, the present model of SAH results in enhanced expression of MMP-9 and TIMP-1 in the smooth muscle cells both in large cerebral arteries and in microvessels. The MMP-9 has been reported to be upregulated in blood vessels affected by intracranial aneurysms (Pannu et al, 2006). The MMP-9 is upregulated early in injured tissue, and activation of MMP-9 has a pivotal function in contributing to brain damage after ischemia with opening of the blood–brain barrier (BBB) and contributing to edema formation (Asahi et al, 2000; Baker et al, 2007; Napoli, 2002). Disruption of the basal lamina that surrounds cerebral vessels has been postulated to be the primary cause of microvascular hemorrhage after an ischemic event, and there is a close relationship between BBB leakage and MMP-9 expression in cerebral ischemia (Cunningham et al, 2005; Mun-Bryce and Rosenberg, 1998; Rosenberg, 2002) and in human hemorrhagic stroke (Rosell et al, 2006). In addition, the MMP-9 knockout mouse has reduced infarct size and less BBB damage after experimental stroke (Asahi et al, 2000). Tight junctions form an essential part of the BBB and are disrupted by activation of MMP-9, which degrades claudin-5 and occludin; these two proteins form an essential part of the tight junctions (Yang et al, 2007). Administration of an MMP-9 inhibitor prevented the BBB degradation.

The MMP-9 activity is tightly controlled by TIMP-1, and an imbalance of MMP/TIMP regulation is likely to be involved in SAH. We observed elevated levels of MMP-9 and TIMP-1 in cerebral vessel walls after SAH, and this expression is regulated by transcription through the MEK/ERK1/2 pathway because both their protein and mRNA levels were increased after SAH and abolished by MEK/ERK inhibition. In situations of enhanced expression of MMP-9 and TIMP-1, this expression can be prevented by transcription inhibition at the raf level, providing a novel way of treating brain edema after SAH. The maximum reduction in regional and global CBF appears at 48 hours after SAH (Ansar et al, 2007), and this reduction in CBF was prevented by treatment with SB-386023-b given either acutely or initiated 6 hours after induction of SAH.

Vascular endothelial and smooth muscle cells as well as inflammatory cells and macrophages produce cytokines such as IL-1β and IL-6, and these may further upregulate local MMP levels in the blood vessel walls (Pannu et al, 2006). We observed elevated levels of the pro-inflammatory mediators IL-1β, IL-6, and iNOS in the smooth muscle cells in both small and large cerebral vessels after SAH. According to Iadecola et al (1997), iNOS is not normally present in the CNS, but its expression can be triggered by cytokines and during inflammatory processes. In experimental ischemia, it has been shown that an iNOS inhibitor reduced the infarct volume and improved neurologic outcome (Iadecola et al, 1997). Here, we observed that the iNOS protein is weakly present in cerebral vessels, but this expression is enhanced in cerebral vessels after induced SAH and during inflammatory processes. The raf inhibition given at 6 hours but not at 12 hours after the start of SAH abolished the enhanced expression of the cytokines (IL-6 and IL-1β) and of iNOS in brain vessels.

The lack of a significant beneficial effect of anti-inflammatory drugs and corticosteroids on patients suggests that late cerebral ischemia is a multifactorial disease and that several mechanisms are involved at different stages of the disease with significant interplay. Many studies have shown involvement of the MAPK signaling pathway in cerebral ischemia (Beg et al, 2006; Henriksson et al, 2007; Tibbs et al, 2000). Interestingly, it is only ERK1/2 and not p38 or JNK that is active during the first 24 hours after experimental SAH, as shown with quantitative Western blot of the proteins in large cerebral arteries and microvessels within the brain tissue proper (Ansar and Edvinsson, 2008). JNK and p38 are activated only at 48 hours, and this timing may relate to later effects involving further inflammatory mediators and apoptosis. In support, we observed that SAH results in enhanced phosphorylation of ERK1/2 in the vascular smooth muscle cells and that SB-386023-b treatment normalized this expression. This finding also confirms that intracerebroventricular administration of the raf inhibitor results in specific blockade of the MEK/ ERK1/2 signaling pathway within the cerebrovascular system. We revealed that the specific raf inhibitor SB-386023-b prevented elevated expression in vessel walls of MMP-9, TIMP-1, iNOS, IL-6, IL-1β, and pERK1/2 proteins.

Interestingly, detailed morphologic analysis using confocal microscopy revealed that upregulation of the pro-inflammatory mediators, the extracellular matrix-related genes, and the activation of the MAPK pathway was localized to cerebral vessels with only weak or negligible enhanced expression in the adjacent brain tissue. Both the brain microvessels and the large cerebral arteries are involved to some extent in the inflammatory process after SAH, and the main activation occurs in the cytoplasm of the smooth muscle cells. The mRNA results suggest that MMP-9 and TIMP-1 activation occurs through transcription, whereas cytokine expression occurs through translation.

In conclusion, this study has shown that treatment with a raf inhibitor even as late as 6 hours after SAH onset prevents enhanced expression of the extracellular matrix-related genes, pro-inflammatory mediators, and cytokines typically seen after SAH. This finding suggests a novel approach to modification of the inflammatory processes that are observed in conjunction with cerebral ischemia after SAH.

Footnotes

The authors declare no conflict of interest.

References

Ansar

Edvinsson

(2008) Subtype activation and interaction of protein kinase C and mitogen-activated protein kinase controlling receptor expression in cerebral arteries and microvessels after subarachnoid hemorrhage. Stroke 39:185–90

Ansar

Edvinsson

(2009) Equal contribution of increased intracranial pressure and subarachnoid blood to cerebral blood flow reduction and receptor upregulation after subarachnoid hemorrhage. Laboratory investigation. J Neurosurg 111:978–87

Ansar

Vikman

Nielsen

Edvinsson

(2007) Cerebrovascular ETB, 5-HT1B, and AT1 receptor upregulation correlates with reduction in regional CBF after subarachnoid hemorrhage. Am J Physiol Heart Circ Physiol 293:H3750–8

Asahi

Jung

del Zoppo

Fini

(2000) Role for matrix metalloproteinase 9 after focal cerebral ischemia: effects of gene knockout and enzyme inhibition with BB-94. J Cereb Blood Flow Metab 20:1681–9

Baker

Sica

Work

Williams-Ignarro

de Nigris

Lerman

Casamassimi

Lanza

Schiano

Rienzo

Ignarro

Napoli

(2007) Brain protection using autologous bone marrow cell, metalloproteinase inhibitors, and metabolic treatment in cerebral ischemia. Proc Natl Acad Sci USA 104:3597–602

Beg

Hansen-Schwartz

Vikman

Edvinsson

(2006) ERK1/2 inhibition attenuates cerebral blood flow reduction and abolishes ET(B) and 5-HT(1B) receptor upregulation after subarachnoid hemorrhage in rat. J Cereb Blood Flow Metab 26:846–56

Cahill

Calvert

Zhang

(2006) Mechanisms of early brain injury after subarachnoid hemorrhage. J Cereb Blood Flow Metab 26:1341–53

Cunningham

Wetzel

Rosenberg

(2005) Multiple roles for MMPs and TIMPs in cerebral ischemia. Glia 50:329–39

Delgado

Brismar

Svendgaard

(1985) Subarachnoid haemorrhage in the rat: angiography and fluorescence microscopy of the major cerebral arteries. Stroke 16:595–602

10.

Dumont

Chow

Lin

Calisaneller

Ley

Kassell

Lee

(2003) Cerebral vasospasm after subarachnoid hemorrhage: putative role of inflammation. Neurosurgery 53:123–33; discussion 133–135

11.

Gallia

Tamargo

(2006) Leukocyte-endothelial cell interactions in chronic vasospasm after subarachnoid hemorrhage. Neurol Res 28:750–8

12.

Gjedde

Hansen

Siemkowicz

(1980) Rapid simultaneous determination of regional blood flow and blood-brain glucose transfer in brain of rat. Acta Physiol Scand 108:321–30

13.

Hansen-Schwartz

Hoel

Zhou

Svendgaard

Edvinsson

(2003) Subarachnoid hemorrhage enhances endothelin receptor expression and function in rat cerebral arteries. Neurosurgery 52:1188–94; 1194–1195

14.

Henriksson

Stenman

Vikman

Edvinsson

(2007) MEK1/2 inhibition attenuates vascular ETA and ETB receptor alterations after cerebral ischaemia. Exp Brain Res 178:470–6

15.

Iadecola

Zhang

Casey

Nagayama

Ross

(1997) Delayed reduction of ischemic brain injury and neurological deficits in mice lacking the inducible nitric oxide synthase gene. J Neurosci 17:9157–64

16.

Kaminska

(2005) MAPK signalling pathways as molecular targets for anti-inflammatory therapy—from molecular mechanisms to therapeutic benefits. Biochim Biophys Acta 1754:253–62

17.

Mun-Bryce

Rosenberg

(1998) Matrix metalloproteinases in cerebrovascular disease. J Cereb Blood Flow Metab 18:1163–72

18.

Napoli

(2002) MMP inhibition and the development of cerebrovascular atherosclerosis: the road ahead. Stroke 33:2864–5

19.

Pannu

Kim

Guo

King

Van Ginhoven

Chin

Chang

Shete

Milewicz

(2006) The role of MMP-2 and MMP-9 polymorphisms in sporadic intracranial aneurysms. J Neurosurg 105: 418–23

20.

Prunell

Mathiesen

Diemer

Svendgaard

(2003) Experimental subarachnoid hemorrhage: subarachnoid blood volume, mortality rate, neuronal death, cerebral blood flow, and perfusion pressure in three different rat models. Neurosurgery 52:165–75; discussion 175–176

21.

Prunell

Svendgaard

Alkass

Mathiesen

(2005) Inflammation in the brain after experimental subarachnoid hemorrhage. Neurosurgery 56:1082–92; discussion 1082–1092

22.

Rosell

Ortega-Aznar

Alvarez-Sabin

Fernandez-Cadenas

Ribo

Molina

Montaner

(2006) Increased brain expression of matrix metalloproteinase-9 after ischemic and hemorrhagic human stroke. Stroke 37:1399–406

23.

Rosenberg

(2002) Matrix metalloproteinases in neuroinflammation. Glia 39:279–91

24.

Sakurada

Kennedy

Jehle

Brown

Carbin

Sokoloff

(1978) Measurement of local cerebral blood flow with iodo [14C] antipyrine. Am J Physiol 234:H59–66

25.

Stenman

Edvinsson

(2004) Cerebral ischemia enhances vascular angiotensin AT1 receptor-mediated contraction in rats. Stroke 35:970–4

26.

Tibbs

Zubkov

Aoki

Meguro

Badr

Parent

Zhang

(2000) Effects of mitogen-activated protein kinase inhibitors on cerebral vasospasm in a double-hemorrhage model in dogs. J Neurosurg 93:1041–7

27.

Vikman

Ansar

Edvinsson

(2007) Transcriptional regulation of inflammatory and extracellular matrix-regulating genes in cerebral arteries following experimental subarachnoid hemorrhage in rats. Laboratory investigation. J Neurosurg 107:1015–22

28.

Yang

Estrada

Thompson

Liu

Rosenberg

(2007) Matrix metalloproteinase-mediated disruption of tight junction proteins in cerebral vessels is reversed by synthetic matrix metalloproteinase inhibitor in focal ischemia in rat. J Cereb Blood Flow Metab 27:697–709