Doxycycline, a matrix metalloprotease inhibitor, reduces vascular remodeling and damage after cerebral ischemia in stroke-prone spontaneously hypertensive rats

Abstract

Matrix metalloproteases (MMPs) are a family of zinc peptidases involved in extracellular matrix turnover. There is evidence that increased MMP activity is involved in remodeling of resistance vessels in chronic hypertension. Thus we hypothesized that inhibition of MMP activity with doxycycline (DOX) would attenuate vascular remodeling. Six-week-old male stroke-prone spontaneously hypertensive rats (SHRSP) were treated with DOX (50 mg·kg(-1)·day(-1) in the drinking water) for 6 wk. Untreated SHRSP were controls. Blood pressure was measured by telemetry during the last week. Middle cerebral artery (MCA) and mesenteric resistance artery (MRA) passive structures were assessed by pressure myography. MMP-2 expression in aortas was measured by Western blot. All results are means ± SE. DOX caused a small increase in mean arterial pressure (SHRSP, 154 ± 1; SHRSP + DOX, 159 ± 3 mmHg; P < 0.001). Active MMP-2 expression was reduced in aorta from SHRSP + DOX (0.21 ± 0.06 vs. 0.49 ± 0.13 arbitrary units; P < 0.05). In the MCA, at 80 mmHg, DOX treatment increased the lumen (273.2 ± 4.7 vs. 238.3 ± 6.3 μm; P < 0.05) and the outer diameter (321 ± 5.3 vs. 290 ± 7.6 μm; P < 0.05) and reduced the wall-to-lumen ratio (0.09 ± 0.002 vs. 0.11 ± 0.003; P < 0.05). Damage after transient cerebral ischemia (transient MCA occlusion) was reduced in SHRSP + DOX (20.7 ± 4 vs. 45.5 ± 5% of hemisphere infarcted; P < 0.05). In the MRA, at 90 mmHg DOX, reduced wall thickness (29 ± 1 vs. 22 ± 1 μm; P < 0.001) and wall-to-lumen ratio (0.08 ± 0.004 vs. 0.11 ± 0.008; P < 0.05) without changing lumen diameter. These results suggest that MMPs are involved in hypertensive vascular remodeling in both the peripheral and cerebral vasculature and that DOX reduced brain damage after cerebral ischemia.

Fig. 1.

Doxycycline (DOX) treatment reduced the expression of active-matrix metalloprotease-2 (MMP-2) in the aorta of SHRSP. Aliquots of aorta supernatants containing 50 μg of protein were resolved by SDS-PAGE and transferred to a PVDF membrane for Western blot analysis. DOX treatment reduced the expression of active-MMP-2, but not pro-MMP-2, in the aorta of stroke-prone spontaneously hypertensive rats (SHRSP; A). Data are shown as a ratio of MMP-2 expression to β-tubulin expression (loading control). B: representative images of blots. *P < 0.05, Student's t-test. Number of animals: SHRSP = 4, SHRSP + DOX = 4.

Fig. 2.

DOX treatment did not alter middle cerebral artery (MCA) reactivity to 5-hydroxytriptamine (5-HT; A) and bradykinin (BK; B) and did not alter MCA myogenic response to increases in intralumenal pressure (C). MCAs were maintained at 80 mmHg of intralumenal pressure, and drugs were added to the bath in a cumulative fashion. The vessel was allowed to equilibrate for 10 min at every dose to reach steady state. To assess myogenic tone (C), intralumenal pressure was increased from 3 to 180 mmHg in 20-mmHg increments, and the vessel was allowed to equilibrate for 5 min at each pressure before measurement was taken. Values are means ± SE. Data were analyzed by two-way repeated-measures ANOVA.

Fig. 3.

DOX treatment attenuated hypertensive remodeling of the MCA in SHRSP. Outer (A) and lumen diameter (B) were increased in SHRSP + DOX and were not different from Wystar Kyoto (WKY) rats ± DOX. Wall thickness was increased in SHRSP ± DOX compared with WKY rats ± DOX (C). The wall-to-lumen ratio was improved by DOX treatment, even without decrease in wall thickness (D). Measurements were obtained from cannulated MCAs using pressure myography under zero flow and calcium-free conditions. Values are means ± SE. Data were analyzed by two-way repeated-measures ANOVA. *P < 0.001.

Fig. 4.

DOX treatment attenuated hypertensive remodeling of the MCA. Outer (A) and lumen (B) cross-sectional area (CSA) were significantly increased in the MCA from SHRSP + DOX compared with untreated SHRSP, and not different from WKY rats ± DOX. Intralumenal stress was higher in WKY rats ± DOX than SHRSP + DOX and untreated SHRSP at intralumenal pressures >60 mmHg (C). SHRSP + DOX showed higher intralumenal stress than untreated SHRSP (C). Strain was not different between the experimental groups (D). Measurements were obtained from MCAs using pressure myography under no-flow and zero calcium conditions. Values are means ± SE. *P < 0.001, by two-way repeated-measures ANOVA.

Fig. 5.

MCA distensibility and stiffness were not changed by DOX treatment. MCA from SHRSP had less distensibility than MCA from WKY rats, and DOX treatment did not change distensibility in either strain (A). Elastic modulus of the MCA (β-coefficient) was not increased in hypertensive animals compared with normotensive WKY rats (B; number of animals: SHRSP = 12, SHRSP + DOX = 9, WKY rats = 4, WKY rats + DOX = 7). Measurements were obtained from cannulated MCAs using pressure myography under zero flow and calcium-free conditions. Values are means ± SE.

Fig. 6.

Chronic hypertension induces wall hypertrophy in mesenteric resistance artery (MRA), and this was prevented by DOX treatment. MRA wall thickness (A), wall CSA (B), and wall-to-lumen ratio (C) were increased in untreated SHRSP compared with WKY rats ± DOX. DOX treatment attenuated this increase in wall mass in SHRSP, although wall thickness and wall CSA were higher in SHRSP + DOX than WKY rats ± DOX. Measurements were obtained from cannulated MCAs using pressure myography under zero flow and calcium-free conditions. Values are means ± SE. *P < 0.001, two-way repeated-measures ANOVA.

Fig. 7.

DOX treatment did not reduce cerebral damage after permanent MCA occlusion (pMCAO). 2,3,5-Triphenyltetrazolium chloride (TTC) staining of the brain of SHRSP that underwent pMCAO showed no differences in infarct size after DOX treatment (A; number of animals: SHRSP = 6, SHRSP + DOX = 8). Data are expressed as %hemisphere infarcted (%HI). Pial blood flow, as measured by scanning laser Doppler (B), was also not altered by DOX treatment. Pial blood flow data are shown as %preischemic blood flow. Values are means ± SE. C: representative gelatin zymography of plasma samples 24 h after cerebral ischemia. No difference was observed in activity of pro-MMP-2 and active-MMP-2 between SHRSP and SHRSP + DOX. MMP-9 activity was not observed in the plasma of these animals.

Fig. 8.

DOX treatment reduced cerebral damage after transient MCA occlusion (tMCAO) and improved pial blood flow in SHRSP. TTC staining of the brain of SHRSP that underwent tMCAO showed a reduction in infarct size in SHRSP + DOX (A; number of animals: SHRSP = 5, SHRSP + DOX = 5). Top: representative images of infarcts from untreated SHRSP and SHRSP + DOX. White area is tissue damaged by ischemia, and the dark area is viable tissue. Bottom: cerebral infarct expressed as %hemisphere infarcted (%HI). The reduction in infarct was accompanied by an increase in pial blood flow in the ischemic hemisphere 23 h after reperfusion, as measured by scanning laser Doppler (B). Pial blood flow in the nonischemic hemisphere was not altered by DOX treatment (C). Pial blood flow data are shown as %preischemic pial blood flow. Values are means ± SE. *P < 0.05, Student's t-test. D: representative images of scanning laser Doppler 23 h after reperfusion in untreated SHRSP (left) and SHRSP + DOX (right). In both images, the left hemisphere is the ischemic hemisphere and the drawn line corresponds to the regions of interest (1: nonischemic hemisphere; 2: ischemic hemisphere). Within the ischemic hemisphere, darker regions correspond to lowest blood flow. E: representative gelatin zymography of plasma samples 24 h after cerebral ischemia. No difference was observed in activity of pro-MMP-2 and active-MMP-2 between SHRSP and SHRSP + DOX. MMP-9 activity was not observed in the plasma of these animals.