Attenuation of acute stroke injury in rat brain by minocycline promotes blood-brain barrier remodeling and alternative microglia/macrophage activation during recovery

Abstract

Background: Minocycline reduces reperfusion injury by inhibiting matrix metalloproteinases (MMPs) and microglia activity after cerebral ischemia. Prior studies of minocycline investigated short-term neuroprotective effects during subacute stage of stroke; however, the late effects of minocycline against early reperfusion injury on neurovascular remodeling are less well studied. We have shown that spontaneous angiogenesis vessels in ischemic brain regions have high blood-brain barrier (BBB) permeability due to lack of major tight junction proteins (TJPs) in endothelial cells at three weeks. In the present study, we longitudinally investigated neurological outcome, neurovascular remodeling and microglia/macrophage alternative activation after spontaneous and minocycline-induced stroke recovery.

Methods: Adult spontaneously hypertensive rats had a 90 minute transient middle cerebral artery occlusion. At the onset of reperfusion they received a single dose of minocycline (3 mg/kg intravenously) or a vehicle. They were studied at multiple time points up to four weeks with magnetic resonance imaging (MRI), immunohistochemistry and biochemistry.

Results: Minocycline significantly reduced the infarct size and prevented tissue loss in the ischemic hemispheres compared to vehicle-treated rats from two to four weeks as measured with MRI. Cerebral blood flow measured with arterial spin labeling (ASL) showed that minocycline improved perfusion. Dynamic contrast-enhanced MRI indicated that minocycline reduced BBB permeability accompanied with higher levels of TJPs measured with Western blot. Increased MMP-2 and -3 were detected at four weeks. Active microglia/macrophage, surrounding and within the peri-infarct areas, expressed YM1, a marker of M2 microglia/macrophage activation, at four weeks. These microglia/macrophage expressed both pro-inflammatory factors tumor necrosis factors-α (TNF-α) and interleukin-1β (IL-1β) and anti-inflammatory factors transforming growth factor-β (TGF-β) and interleukin-10 (IL-10). Treatment with minocycline significantly reduced levels of TNF-α and IL-1β, and increased levels of TGF-β, IL-10 and YM1.

Conclusions: Early minocycline treatment against reperfusion injury significantly promotes neurovascular remodeling during stroke recovery by reducing brain tissue loss, enhancing TJP expression in ischemic brains and facilitating neuroprotective phenotype alternative activation of microglia/macrophages.

Figure 1

Stroke recovery monitored by magnetic resonance imaging. A. Anatomical T2 MR images. Arrows indicate ischemic hemispheres. Line graph demonstrates quantification of infarct volumes in ischemic hemispheres. B. ADC maps. Line graph demonstrates quantification of edema (acute stage) and tissue loss (late stage) in ischemic hemispheres. C. FA maps. Line graph demonstrates quantification of white matter change in ischemic hemispheres. *P <0.05, ***P <0.001 versus vehicle group, n = 8 in each group.

Figure 2

ASL maps monitored by MRI at 48 hours and one, two and four weeks after stroke. Arrows indicate ischemic hemispheres. Line graph demonstrates quantification of blood flow in peri-infarct areas. *P <0.05 versus vehicle group, n = 8 in each group.

Figure 3

Blood–brain barrier permeability monitored by magnetic resonance imaging. A. Parametric image K _i map by DCE-MRI represents BBB transfer rate at four weeks after stroke. Color-coded permeability coefficient maps reconstructed from contrast-enhanced MRI data demonstrate the regions of high (red) and low (blue) permeability. B. Histogram demonstrates the quantification of BBB permeability in peri-infarct areas (light blue areas in the brain slice cartoon). *P = 0.0504 versus vehicle group, n = 8 in each group. C. RECA1 (marker of endothelial cells) immunostaining shows increased density of new vessels in the peri-infarct area (arrows), corresponding to the regions where BBB transfer rate and plasma volume were measured.

Figure 4

Expression of tight junction proteins and matrix metalloproteinases at four weeks after stroke. A. Western blot analysis for TJPs. ZO-1: **P <0.05 versus V-I, V-C and Mino-C. Occludin: *P <0.05 versus V-I, V-C and Mino-C. Claudin-5: *P <0.05 versus V-I, V-C and Mino-C. B. Gel zymography analysis for MMP-2 and −9. **P <0.01 versus V-C and Mino-C. n = 8 in the vehicle group, 9 in the minocycline group. C. Western blot analysis for MMP-3. Level of MMP-3 including preform (57 kDa) and active form (45 kDa): *P <0.05 versus V-C and Mino-C. Mino-C: minocycline contralateral. Mino-I: minocycline ipsilateral. V-C: Vehicle contralateral. V-I: Vehicle ipsilateral. n = 8 in the vehicle group, 9 in the minocycline group.

Figure 5

Microglia/macrophage activation by immunofluorescence staining of Iba-1. A. Morphological changes of microglia/macrophage in ischemic hemispheres over reperfusion courses. Scale bar = 100 μm. The right panels at each time point present a higher magnification of the images shown in squares in the left panels. Scale bar = 50 μm. DAPI was used to show nuclei and vessel at one week. core-i: core infarct area; peri-i: peri-infarct area; V: vessel. B. Quantification of Iba-1 fluorescence (FL) intensity in ischemic hemispheres over reperfusion courses. *P <0.05, ***P <0.001 versus vehicle group, n = 5 in each group. The brain cartoon shows the three images measured by ImageJ that were obtained from the infarct areas, indicated by the squares in red.

Figure 6

Phenotype of microglia/macrophage alternative activation at four weeks after stroke. A. Double immunostaining shows expression of YM1 in microglia/macrophage (OX-42). Scale bars = 50. DAPI was used to show nuclei. B. Western blot analysis for protein level of YM1. **P <0.01 versus V-C and Mino-C. ***P <0.001 versus V-C and Mino-C. ^# P <0.05 versus V-I. n = 8 in the vehicle group, 9 in the minocycline group.

Figure 7

Inflammatory factors expressed by active microglia/macrophage in peri-infarct areas at four weeks after stroke. A. Double immunostaining shows expression of TNF-α, IL-1β, IL-10 and TGF-β in active microglia (Iba-1). B. Analysis and quantification for co-localization of cytokines and Iba-1 in the microglia/macrophages with Fiji-ImageJ. Representative two-dimensional histogram and scatterplots visualize the correlation of the pixel intensities, over all pixels and voxels in the images with different Li’s ICQ values, generated by Fiji-ImageJ. Statistical bar figures demonstrate the quantification of Li’s ICQ values for co-localization of each cytokine with Iba-1 in vehicle- and minocycline-treated ischemic hemispheres. *P <0.05, **P <0.01 versus V-I. n = 5 in each group. C. Double immunostaining shows little expression of TNF-α and IL-1β in active microglia/macrophage, with extending processes in the peri-infarct areas bordering with intact tissues. Scale bars = 50 μm. D. Double immunostaining shows expression of TGF-β in PDGFR-β-positive pericytes that closely surround the vessels (V). Scale bars = 50 and 100 μm.

Figure 8

Western blot analysis for active forms of inflammatory factors in rat brains at four weeks after stroke. TNF-α: levels of TNF-α were decreased in ischemic rat brains and treatment with minocycline facilitated the reduction. IL-1β: *P <0.05 versus V-I. IL-10: *P <0.05 versus V-I. TGF-α: *P <0.05 versus V-I; ^## P <0.01 versus V-C and Mino-C. n = 8 in the vehicle group, 9 in the minocycline group.