Whisker Stimulation Enhances Angiogenesis in the Barrel Cortex following Focal Ischemia in Mice

Animals and Focal Ischemia of the Whisker-Barrel Cortex

The focal barrel cortex ischemic stroke was similarly induced as previously described in rats (Wei et al, 1995), with modifications for a few differences in the surgery procedures and anatomy in mice. Adult C57/BL6 mice weighing between 25 and 30 g and in the age range of 2.5 to 3 months were anesthetized with isoflourane. Focal ischemia of the barrel cortex was induced by permanent occlusion of 2 to 3 branches of the MCA supplying the right whisker barrel cortex. This was accompanied by 7-mins occlusion of the common carotid artery. During surgery and recovery periods, body temperature was monitored using a rectal probe and maintained at 37°C with a homoeothermic blanket in a ventilated incubator. The right femoral artery was cannulated in some of the animals for monitoring arterial blood pressure and for obtaining blood samples for arterial blood gases. Blood pressure was monitored with a Digi-Med blood pressure analyzer (Micro-Med Inc., Tustin, CA, USA). Arterial blood gases were determined before ischemia and 1 to 14 days after the onset of ischemia with a blood gas analyzer (Ciba Corning, Ramsey, MN, USA). Mice were killed by decapitation 7 or 14 days after ischemic stroke. The brain was immediately removed and preserved in OCT compound at −80°C until further processing. Before and after the surgery, animals were housed in standard mice cages (27 × 17 × 13 cm), at 3 to 4 mice per cage, having free access to food and water. The animal protocol was approved by the Institutional Animal Care and Use Committee (IACUC) at MUSC. Animal procedures followed institutional guidelines, which meet or exceed NIH standards.

Whisker Stimulation

Three days after ischemia, mice were randomly divided into three groups: (1) ischemic stroke-only group (control), (2) a group in which all right whiskers were trimmed to the skin, so the animals had to rely on their left side whiskers that have neuronal pathways across the brain stem to the ischemic barrel cortex on the right hemisphere, and (3) a group of mice that received manual stimulation of their left whiskers (Figure 1). The left whiskers in the third group were swiped in a rostro-caudal direction using a stick at a frequency of about 140 strokes per minute (∼ 2.3 Hz) for 15 mins, performed three times a day. During periods of stimulation, animals behaved normally and were allowed to move freely in the cage. Other than trimming and manual stimulation, no additional environmental manipulation was applied.

OPEN IN VIEWER

Figure 1

Experimental groups and whisker stimulation. Adult mice were subjected to permanent occlusion of MCA branches supplying the right barrel cortex. Three days after ischemia, all mice received BrdU injection (50 mg/kg, intraperitoneally, once/day). The mice were randomly divided into three groups: (1) ischemic stroke-only group as the control, (2) all right whiskers trimmed to enhance the whisker activity of the left face whiskers (trimmed group), and (3) direct manual stimulation of left whiskers for 15 mins at about 140 strokes per minute, three times per day (manually stimulated group). The left whiskers have corresponding neuronal pathways that, via the interneurons in the brain stem, lead to the right barrel cortex. The white circle represents the ischemic barrel cortex in the right hemisphere.

5-Bromo-2′-Deoxyuridine Administration

For labeling of cell proliferation, 5-bromo-2′-deoxyuridine (BrdU) (Sigma, St Louis, MO, USA) was administered to animals at a dosage of 50 mg/kg body weight by intraperitoneal injection. Injections began on day 3 after ischemia and continued once daily until killing.

TdT-Mediated dUTP Nick End Labeling Staining

A TdT-mediated dUTP nick end labeling (TUNEL) staining kit (DeadEnd™ Fluorometric TUNEL system, Promega, Madison, WI, USA) was used to assess cell death by catalytically incorporating fluorenscein-12-dUTP at 3′-OH DNA ends using the terminal deoxynucleotidyl transferase and recombinant enzyme. After fixing and pretreatment with ethanol/acetic acid and triton, the brain sections were incubated in an equilibration buffer as described in the kit. The TdT enzyme and nucleotide mix were then added at proportions specified in the kit for 75 mins at room temperature. The slides were washed with the provided wash buffer. To identify endothelial cell death, sections were then incubated with the Glut-1 primary antibody overnight. After phosphate-buffered saline (PBS) washes and incubation with Cy3-conjugated anti-rabbit IgG, brain slides were mounted for viewing and cell counting.

Microvessel Staining

Fresh frozen brains were sliced into coronal sections at 14-μm thick using a cryostat vibratome (Ultapro 5000, St Louis, MO, USA). Sections were completely air dried and subsequently fixed in 10% buffered formalin phosphate (Fisher Scientific, Pittsburgh, PA, USA). Brain sections were then processed in an ethanol:acetic acid (2:1) solution at −20°C for 10 mins, washed three times in PBS solution at pH 7.4, and incubated in 0.2% Triton X-100 for 5 mins. Slides were blocked in 1% fish gelatin (Sigma, St Louis, MO, USA) in PBS for 1 h at room temperature. Incubation with a primary antibody diluted in PBS was performed overnight at 4°C. The rabbit anti-Glut-1 (1:1,000, Chemicon, Temecula, CA, USA) and rat anti-CD-31 (1:400, Cymbus Biotechnology, Eastleigh, UK) antibodies were used to label endothelial cells. Primary antibodies were removed by PBS washes and replaced by secondary antibodies Alexa Fluor 488 anti-rabbit IgG and anti-rat IgG (1:200, Molecular Probes, Carlsbad, CA, USA) for 1 h at room temperature before rinsing again in PBS.

For double staining with BrdU, brain slides were postfixed in 10% buffered formalin phosphate and washed in PBS. Slides were then treated with 100% methanol and allowed to air dry. After re-wetting in PBS, sections were incubated in 2 N HCl for 1 h at 37°C. Sections were treated with 0.1 mol/L borate buffer at pH 8.4 for 10 mins, followed by a 45 mins incubation in 0.2% Triton X-100. After blocking in 1% fish gelatin, slides were incubated with mouse anti-BrdU (1:1,000, Roche, Basel, Switzerland) or rat anti-BrdU (1:400, Abcam, Cambridge, UK) antibodies overnight at 4°C followed by rinsing in PBS before being incubated with Cy3-conjugated goat antimouse IgG (1:500; Jackson ImmunoResearch, West Grove, PA, USA) or Texas Red goat anti-rat IgG (1:400, Abcam, Cambridge, UK) antibodies for 1 h at room temperature. After a final PBS rinsing, slides were mounted with ProLong Antifade mounting medium (Molecular Probes, Carlsbad, CA, USA) for photography and imaging analysis.

Cytochrome Oxidase Histology

Barrel pattern on the cortex was examined using the mitochondrial enzyme cytochrome oxidase (Dowling et al, 1996). Briefly, following videomicroscopy, mice were perfused with warm heparinized saline followed by 10% formalin for 15 mins. The brains were removed and immersed for 24 h in periodate-lysine-paraformaldehyde buffered to pH 7.4, then transferred to pariodate-lysine-paraformaldehyde 30% sucrose solution until they sank. The brains were sectioned parallel to the pial surface into 50 mm slices, and were stained for cytochrome oxidase activity. Serial sections were mounted on chromealum subbed slides, cleared in xylene, and coverslipped.

Western Blot Analysis of Angiogenic Factors

Tissue samples were taken from the penumbra region of brain and proteins were extracted using the NE-PER Nuclear and Cytoplasmic Extraction Kit (Pierce Biotechnology, Rockford, IL, USA). Protein concentration of each sample was determined using the Bicinchoninic Acid Assay (Sigma, St Louis, MO, USA). Proteins from each sample (50 μg) were separated by SDS-polyacrylamide gel electrophoresis in a Hoefer Mini-Gel system (Amersham Biosciences, Piscataway, NJ, USA) and transferred in the Hoefer Transfer Tank (Amersham Biosciences, Piscataway, NJ, USA) to a PVDF membrane (BioRad, Hercules, CA, USA). Membranes were blocked in 7% evaporated milk diluted in Tris-buffered saline containing 0.1% Tween 20 (TBS-T) at room temperature for at least 2 h, and then incubated with one of the following antibodies: VEGF (Santa Cruz, Santa Cruz, CA, USA), Tie-1 (Santa Cruz, Santa Cruz, CA, USA), Tie-2 (Santa Cruz, Santa Cruz, CA, USA), basic fibroblast growth factor (bFGF) (Santa Cruz, Santa Cruz, CA, USA), Ang-1 (Chemicon, Temecula, CA, USA), and Ang-2 (Chemicon, Temecula, CA, USA). A mouse β-actin antibody (Sigma, St Louis, MO, USA) was used for protein loading control. After primary antibody incubation, membranes were washed with TBS-T and incubated with alkaline phosphatase-conjugated anti-mouse or anti-rabbit IgG (Promega, Madison, WI, USA) antibodies for 2 h at room temperature. Finally, membranes were washed with TBS-T followed by three washes with TBS and signal was detected by the addition of 5-bromo-4-chloro-3-indolylphosphate/nitroblue tetrazolium (BCIP/NBT) solution (Sigma, St Louis, MO, USA).

Stereological Cell and Vessel Counting

For systematic random sampling in design-based stereological cell counting, every fifth brain section (70 mm apart) across the entire region of interest were counted. For multistage random sampling, six fields per brain section were randomly chosen under × 40 magnification of a light microscope or in confocal images (see figure legends for animal numbers of each experiment).

Laser Doppler Measurement of Local Cerebral Blood Flow

Cerebral blood flow was measured using laser-Doppler flowmetry (PeriFlux System 5000-PF5010 LDPM unit, Perimed, Stockholm, Sweden). The tip of a flexible 0.5 fiber optic probe was pointed to a selected MCA branch visualized through a cranial window with the dura intact over the right somatosensory cortex (see Figure 7). Visualization of the arteries was achieved by means of angiography and videomicroscopy (Wei et al, 2001). Briefly, under isoflourane anesthesia and epifluorescent illumination, 0.05 mL boluses of 1.3 mmol/L fluorescein isothiocyanate were injected into a branch of the right jugular vein. Dye transits through the arteries, capillaries, and veins of the cortex were observed and recorded through the microscope under a × 4 objective with a Quantex ICCD videocamera set to manual black and fixed gain settings on VHS tape. Vascular images were used for identifying blood vessels and processed with NIH Image 1.61.

Local cerebral blood flow (LCBF) was measured before, during, and after ischemic surgery. The cranial windows could be reopened after ischemia for fluorescein isothiocyanate angiographic videomicroscopy and LCBF measurements. Changes in LCBF after MCA occlusion were expressed as percentages of baseline values before ischemia.

Statistical Analysis

All analyses were performed using SAS, version 9.1. Analyzed data were graphed using GraphPad Prism 4.0. As a result of the hierarchical nested experimental design, the data were clustered. To properly account for this correlation, a Poisson regression using generalized estimating equations was used with the natural log of the count data for each slide, clustering on mouse and slide within mouse; counts within a slide were considered approximately independent of one another. For the time-series data (examining the counts at days 7 and 14), a profile analysis approach was used, which included time in the analysis. To assess whether the change in counts from day 7 to 14 was significantly different between treatment group and control, an interaction term was created as the product of the time variable and treatment group; the significance of these interaction effects in the final analysis was then assessed. Assessing the total number of vessels per field, proliferating vessels per field and cell death were considered as separate outcomes, so an a level of 0.05 was used for each. To account for multiple comparisons, an a level of 0.025 was used within each of these outcomes (Dunnett's procedure for all pairwise comparisons with control). Bar figures show the mean value and standard deviation (s.d.).

Barrel Ischemic Stroke and Stimulation of the Whisker-Barrel Pathway

The sensory barrel cortex on the right hemisphere was subjected to permanent ligation of the MCA branches supplying the region (Wei et al, 2001, 1995). The focal ischemia resulted in restricted infarct and distinctive cell death in the targeted barrel cortex 24 h later. To test the effects of increased whisker activity on angiogenesis in the ischemic cortex region, animals were divided into three groups 3 days after ischemia. In addition to a stroke-only control group, mice in two other groups received additional manipulation to enhance whisker activity on the left face, sending afferent signals to the right side (ischemic) barrel cortex via the thalamocortical pathway. In one experimental group, all right side whiskers were trimmed away, so the animals had to solely rely on their left whiskers for searching. Previous investigations showed that whisker trimming increased activity in spared cortex within hours of trimming (McCasland and Woolsey, 1988), and single-unit recordings showed increased synaptic activity up to 30 days after trimming (Armstrong-James et al, 1994). In the third group, no trimming was performed but the left whiskers were directly stimulated by rhythmic sweeps at a frequency of 140 strokes per minute for 15 mins, performed three times a day until killing (Figure 1).

Barrel cortex ischemia induced a restrictive infarct in the right side of the brain cortex 24 h after ischemia (Figures 2A and 2B). Cytochrome oxidase staining revealed the disappearance of most barrel structures in the ischemic barrel cortex (Figures 2C and 2D). Three to seven days after ischemia, many TUNEL-positive cells were observed in the ischemic core and penumbra regions (Figure 2B). Double labeling with Glut-1, a specific marker for glucose transporter in brain vascular endothelial cells (Wei et al, 2001), indicated injured endothelial cells (Figures 3A–3F). In the penumbra region, TUNEL-positive endothelial cells were decreased by enhanced whisker activity induced by manual stimulation, although no reduction was seen with the whisker-trimmed group (Figure 3G).

OPEN IN VIEWER

Figure 2

Ischemia-induced barrel cortex infarct formation and cell death. Middle cerebral artery branch ligations induced a focal ischemic infarct and cell death in the barrel cortex. (A) Selective damage occurred to the right barrel cortex region shown as a negative (white) area in triphenyl tetrazolium staining (red) 24 h after ischemia. The pink area between the normal cortex and ischemic core represents the bordering penumbra area. (B) A brain section at lower magnification shows the massive cell death (TUNEL-positive cells of green color) in the ischemic core and scattered cell death in penumbra 3 days after ischemia. Glut-1 staining (red) shows vascular endothelial cells in the penumbra and normal brain tissues. Blue color is NeuN staining of neurons, which illustrates that the focal ischemia did not cause injury in other brain structures such as the hippocampus. (C and D) Cytochrome oxidase staining revealed the barrel surface structure on the cortex (C) and the ischemic damage (D). The lowercase letters (a–e) mark the characteristic pattern of barrel distribution in the cortex. Fourteen days after ischemia, most of the barrel column structures disappeared except a few barrel-like clumps (arrowhead) preserved in the peripheral region of the barrel cortex.

OPEN IN VIEWER

Figure 3

Ischemia-induced endothelial cell death. Using confocal microscopy, endothelial cells were identified using Glut-1 staining and cell death was assessed using TUNEL staining of DNA damage 7 days after ischemia. (A–C) Glut-1-positive (red in (A and C)) and TUNEL-positive (green in (B and C)) cells in the penumbra; the merged image (E) shows double staining of Glut-1 and TUNEL, illustrating endothelial cell death. (D–F) Enlarged images of the frame in (C). The three-dimensional confocal image confirms the colocalization of Glut-1 and TUNEL staining. (G) Average endothelial cell death in a single field calculated from cell counts in six random fields in the penumbra region per brain section. The whisker-trimmed group showed no significant trend in reducing endothelial cell death; a significant decrease in endothelial cell death was induced by manual whisker stimulation. N = 7 mice for each group, *P < 0.05 compared with stroke-only controls. Bar = 20μm in (A–C), and 10μm in (D–F).

Whisker Stimulation and Angiogenic Factor Expression in the Ischemic Barrel Cortex

Expression of several angiogenic factors, VEGF, bFGF, Ang-1, Ang-2, Tie-1, and Tie-2, in the ischemic core/penumbra region was analyzed 7 days after ischemia by Western blot (Figure 4). Manual stimulation of whiskers led to significant increases of expression of all angiogenic factors tested except Tie-2 and Ang-1 compared with the stroke-only controls (Figure 4). Conversely, whisker trimming increased the expression of Tie-1 and bFGF, but not other factors (Figure 4). In the case of Ang-1 expression, both stimuli showed a noticeable trend of increasing this angiogenic factor, although the changes were not statistically significant (Figure 4).

OPEN IN VIEWER

Figure 4

Expression of angiogenic factors in the penumbra region and effect of whisker stimulation. The protein levels of Tie-1, Tie-2, Ang-1, Ang-2, VEGF, and bFGF were detected using Western blot analysis. (A) Electrophoresis gels show the levels of different angiogenic factors 7 days after ischemia. (B-G) Densitometry analysis for comparisons of each angiogenic factor. Gray intensity was normalized against β-actin (not shown) and quantified using Adobe Photoshop. Whisker hyperactivity induced by manual stimulation enhanced the expression of Tie-1, Ang-2, VEGF, and bFGF. Mice that received right side whisker trimming showed increases in the expression of Tie-1 and bFGF. Both stimuli showed a trend of increasing Ang-1 expression although the changes were not significant in statistical analysis. N = 3 to 5 animals for each test. *P < 0.05 compared with stroke-only controls.

In immunohistochemical assessments 7 days after ischemia, Tie-1 immunoreactivity was barely detectable in the contralateral cortex, whereas positive staining was seen in the ipsilateral cortex including the ischemic core and penumbra regions. Tie-1 and VEGF double labeling almost exclusively stained microvessels but not large vessels (Figures 5A–5C).

OPEN IN VIEWER

Figure 5

Endothelial cell proliferation and microvessel formation in the ischemic penumbra region. Immunohistochemical staining for angiogenesis and vessel formation 7 days after ischemia. (A–C) VEGF (red (A)) and Tie-1 (green (B)) expression in the penumbra region. The expression of VEGF expression largely overlapped with Tie-1 expression (yellow (C)) except in presumably larger vessels where only VEGF was seen. (D–F) Microvessels with Glut-1 staining (green (D)). The blue color is NeuN staining. Proliferating cells were revealed by BrdU staining (red (E)). Overlapping images in (F) show the relationship between vessels and neuronal cells and colocalization of a few Glut-1/BrdU-positive cells (arrows), suggesting endothelial cell proliferation. (G) An example of co-labeling of BrdU and Glut-1 under a fluorescence microscope. (H) Confocal imaging shows BrdU-labeled (red) microvessels (Glut-1, green); the narrow images on the top and right show the three-dimensional location of the BrdU staining. (I–K) Confocal imaging shows colocalization of immunostaining for CD-31 and Glut-1. Both CD-31 (red) and Glut-1 (green) labeled the same cells, supporting the use of Glut-1 as a microvessel marker. The narrow images on the top and right of (K) show the three-dimensional colocalization of the two markers. Bar = 50 μm (A–C) and 20 μm (D–K).

Endothelial Cell Proliferation and Vessel Growth After Focal Cerebral Ischemia

Proliferating endothelial cells were detected by double labeling of Glut-1 and the proliferation marker BrdU. Little BrdU labeling was observed in blood vessels in the contralateral cortex or in the normal cortex adjacent to the ischemic penumbra. Some surface vessels and intraparenchymal microvessels in the penumbra expressed BrdU-labeled nuclei 7 and 14 days after ischemia (Figures 5D–5H).

CD-31 is a well-known marker specific for endothelial cells. We performed double labeling with anti-CD-31 and anti-Glut-1 to confirm the specificity of our Glut-1 labeling. Confocal microscopy showed that CD-31 and Glut-1 antibodies indeed labeled identical microvessels of the mouse brain tissue (Figures 5I–5K).

The hyperactivity induced by whisker trimming and manual stimulation increased Glut-1-positive endothelial cells/vessels (Figure 6). To quantify endothelial cell proliferation, Glut-1-positive microvessels co-labeled with BrdU were counted in the penumbra region. At 7 and 14 days after ischemia, both trimmed and manually stimulated groups showed significant increases in endothelial proliferation compared with stroke-only controls (Figure 7A). The total vessel number, indicated by Glut-1-positive staining, was also increased by whisker manipulations (Figure 7B). Interestingly, the vessel number in the penumbra of stroke-only mice increased from 7 to 14 days after ischemia, although the increase was significantly smaller than that in stimulated groups (Figure 7). Consistently, better recovery of LCBF in the stroke region of whisker-stimulated mice was detected 14 days after ischemia compared with that of mice without whisker stimulation (Figure 8). Meanwhile, physiologic variables of mice were not changed during 14-day experiments (Table 1).

OPEN IN VIEWER

Table 1

Physiologic variables before and after ischemia

	MABP (mm Hg)	pH	pCO2 (mm Hg)	pO2(mm Hg)
Before MCAO	60 ± 2	7.3 ± 0.02	40 ± 6	80 ± 5
1 day after MCAO	59 ± 2	7.2 ± 0.02	39 ± 7	81 ± 7
14 days after MCAO	60 ± 1	7.4 ± 0.01	40 ± 4	80 ± 5

Values are mean ± s.e.m.; n = 5–7. MCAO, middle cerebral artery occlusion; MABP mean arterial blood pressure.

OPEN IN VIEWER

Figure 6

Vessel staining in ischemic penumbra and effect of whisker stimulation. Fluorescence microscopy showed Glut-1 staining (green) of microvessels in the penumbra region 7 days after ischemia. (A and B) Glut-1 staining in a brain section from a stroke-only mouse. (B) is an enlarged image from the frame in (A). (C and D) A higher density of vessels was labeled by Glut-1 in the brain section from a whisker-trimmed mouse. (D) is an enlarged image from the frame in (C). (E and F) Significantly more Glut-1 staining was seen in mice that received manual whisker stimulation. (F) is an enlarged image from the frame in (E). Bar in (A–F) = 100 μm.

OPEN IN VIEWER

Figure 7

Vessel formation increased by whisker stimulation in the penumbra region. At 7 and 14 days after ischemia, blood vessel proliferation and the number of vessels were increased by whisker stimulation. (A) Summary of the number of vessels per survey field positively labeled by both Glut-1 and BrdU, representing vessel proliferation. The proliferating activity was significantly higher in the whisker-trimmed group and in the manually stimulated group. (B) The number of total vessels positive for Glut-1 staining. Whisker stimulation, either by the trimming method or the manual stroking method, increased the total vessel number in the penumbra region. N = 7 to 8 animals for each group. *P < 0.05 compared with corresponding stroke-only controls.

OPEN IN VIEWER

Figure 8

Enhanced LCBF promoted by whisker hyperactivity. Vascular alterations and LCBF in the ischemic and perifocal region of the ipsilateral hemisphere were observed and measured using videomicroscopy and laser Doppler flowmetry. (A) Fluorescent fluorescein isothiocyanate angiography through the cranial window shows the MCA artery (a) branches, and the asterisk indicates the spot where the laser Doppler probe was placed. (B) Videomicroscopy from the same location after MCA ligation shows the same MCA artery branches and a ligation site (arrow). At the asterisk site and surrounding area, the darkening of the image indicates reduced distal blood flow in the ischemic region. Local cerebral blood flow was measured at the same artery branch and same location (*). (C) Fourteen days after ischemia, a partial but significant reappearance of the vascular network was observed through the re-opened cranial window. Local cerebral blood flow was measured again at the same spot as shown in (A) and (B). (D) The average surface blood flow in this region decreased to ∼25% of the baseline 5 mins after MCA branch occlusion plus common carotid artery ligation. Local cerebral blood flow remained at the low level for many minutes and gradually recovered in the following days. By 14 days after ischemia, both stroke-only mice and manually stimulated mice showed significant restoration of LCBF, but the LCBF was significantly higher in mice that received whisker stimulation. Local cerebral blood flow in sham control animals was unchanged from day 0 to 14 (▪). N = 10 mice per group; *P < 0.05 compared with stroke-only controls.