Matrix metalloproteinases contribute to endotoxin and interleukin-1beta induced vascular dysfunction

Abstract

Background and purpose: The acute vascular inflammatory dysfunction associated with endotoxaemia may reflect an imbalance between matrix metalloproteinases (MMPs) and their natural inhibitors (TIMPs), induced by the endotoxin. This possibility was tested in rat aortic tissue.

Experimental approaches: Tone induced by phenylephrine in aortic rings was measured after exposure in vitro to ambient lipopolysaccharide (LPS) or the proinflammatory cytokine interleukin-1beta (IL-1beta) for 6h, with or without MMP inhibitors (doxycycline or GM6001). Gelatinase and MMP activities, TIMP proteins and contractility were measured in aortae taken from rats 6h after receiving LPS in vivo.

Key results: Inhibition of MMP prevented the loss of phenylephrine-induced tone in aortic rings after LPS or IL-1beta. IL-1beta also increased release of MMP-2 activity from aortic tissue. In aortae exposed in vivo to LPS, net gelatinase, MMP-9 activities and TIMP-1 protein levels were increased, whereas TIMP-4 was reduced. These aortae were hypocontractile to both phenylephrine and KCl. Hypocontractility was partially reversed by doxycycline ex vivo.

Conclusions and implications: MMP inhibitors ameliorate vascular hyporeactivity induced by either LPS or IL-1beta in vitro. LPS in vivo alters the balance between MMPs and TIMPs, contributing to vascular dysfunction which is partially reversed by MMP inhibitors. Vascular MMPs are activated as a result of LPS or IL-1beta-induced stress and contribute to the hyporeactivity of blood vessels to vasoconstrictors.

Figure 1

Structure of MMP inhibitors (a) doxycycline and (b) GM6001.

Figure 2

Response of aortic rings taken from normal rats to a time-dependent loss of phenylephrine (PE)-induced tone. PE was added as indicated by the arrow after 1.5 h equilibration and then rings were treated with either (a) polymyxin B (10 μg ml⁻¹) (b) doxycycline (30 μM) or ddH₂O vehicle or (c) 10 μM GM6001 or 0.1% ethanol vehicle (^*P<0.05, two-way repeated measures ANOVA, n=4–5 aortic rings taken from distinct animals/group).

Figure 3

(a) Response of aortic rings taken from normal rats to an IL-1β (10 ng ml⁻¹) spontaneous loss of PE-induced tone. Rings were treated with either doxycycline (30 or 100 μM) or ddH₂O vehicle (^*P<0.05 vs IL-1β, one-way ANOVA, n=3 rings taken from distinct animals/group). (b) A representative zymogram of MMP-2 activity in incubation media of aortic rings taken from normal rats and incubated for 6 h at 37°C in the presence or absence of IL-1β (10 ng ml⁻¹) (^*P<0.05, independent samples t-test, n=6 rings taken from distinct animals/group). (c) Left: contractile response of aortic rings taken from normal rats and incubated for 6 h at 37°C in the presence or absence of IL-1β (10 ng ml⁻¹)±GM6001 (10 or 30 μM) and then placed in organ baths and exposed to increasing concentrations of PE. Right: summary data of maximal contractile response to PE (^*P<0.05 vs Control, one-way ANOVA followed by Fisher's LSD test, n=8–13 aortic rings taken from distinct animals/group). (d) Contractile response of aortic rings as prepared in (b) to 75 mM KCl (^*P<0.05 vs Control, one-way ANOVA followed by Fisher's LSD test, n=8–13 aortic rings taken from distinct animals/group).

Figure 4

(a) Gelatinolytic and (b) collagenolytic activies of homogenates prepared from aortae removed 6 h after i.p. injection of either LPS-(LPS, 4 mg kg⁻¹) or pyrogen-free water vehicle (Con) (^*P<0.05, independent samples t-test, n=5 rats/group).

Figure 5

(a) Left: A representative zymogram of vascular homogenate MMP-2 activities. Aortae from two control rats (Con) and two LPS-treated rats (LPS) show primarily 72 kDa activity. ‘Std' represents culture media from HT-1080 cells. Gel incubation time: 20 h. Right: a representative immunoblot showing 72 kDa MMP-2 protein content. Position of molecular weight markers is shown on the left of the immunoblot. (b) Left: a representative zymogram of vascular homogenate MMP-9 activity. 92 kDa activity appears in aortae from two LPS-treated rats but not from Con rats. Gel incubation time: 30 h. Right: a representative immunoblot showing 92 kDa MMP-9 protein content (^*P<0.05, independent samples t-test, n=9–13 rats per group).

Figure 6

(a) TIMP-1, (b) TIMP-2 and (c) TIMP-4 protein content in aortae excised from lipopolyssaccharide (LPS) or vehicle (Con) – treated rats. Right panels show representative immunoblots taken from aortae from two control rats (Con) and two LPS-treated rats (LPS). TIMP-3 was not detectable (data not shown). Position of molecular weight markers is depicted on the left (^*P<0.05, independent samples t-test, n=12–13 rats/group for TIMP-1 and TIMP-4, n=6 rats/group for TIMP-2).

Figure 7

Left: contractile response of aortic rings taken from rats 6 h after i.p. injection of either lipopolysaccharide (LPS, 4 mg kg⁻¹) or pyrogen-free water vehicle (Con). Rings were treated ex vivo with either doxycyline (100 μM) or ddH₂O vehicle. Right: summary data of maximal contractile response to PE (^*P<0.05 vs Control, ^#P<0.05 vs LPS, one-way ANOVA followed by Fisher's LSD test, n=3–4 aortic rings/group).