uPA impairs cerebrovasodilation after hypoxia/ischemia through LRP and ERK MAPK

Abstract

We have previously observed that soluble urokinase plasminogen activator receptor (suPAR) prevents impairment of cerebrovasodilation induced by hypercapnia and hypotension after hypoxia/ischemia (H/I) in the newborn pig. In this study, we investigated the role of low-density lipoprotein-related protein (LRP) receptor and the ERK isoform of mitogen activated protein kinase (MAPK) in uPA-mediated impairment of vasodilation after H/I in piglets equipped with a closed cranial window. CSF uPA increased from 9+/-2 to 52+/-8 and 140+/-21 ng/ml at 1 and 4 h after H/I, respectively. The LRP antagonist receptor associated protein (RAP) and anti-LRP antibody blunted the increase in CSF uPA at 1 h (17+/-2 ng/ml) but not 4 h post insult. uPA detectable in sham-treated cortex by immunohistochemistry was markedly elevated 4 h after H/I. Phosphorylation (activation) of CSF ERK MAPK was detected at 1 and 4 h post H/I and blocked by RAP. Exogenous uPA administered at 4 h post H/I further stimulated ERK MAPK phosphorylation, which was blocked by RAP. Pre-treatment of piglets with RAP, anti-LRP, and suPAR completely prevented, and the ERK MAPK antagonist U 0126 partially prevented, impaired responses to hypotension and hypercapnia post H/I, but none of these antagonists affected the response to isoproterenol. These data indicate that uPA is upregulated after H/I through an LRP-dependent process and that the released uPA impairs hypercapnic and hypotensive dilation through an LRP- and ERK MAPK dependent pathway. These data suggest that modulation of uPA upregulation and/or uPA-mediated signal transduction may preserve cerebrohemodynamic control after hypoxia/ischemia.

Figure 1

Immunohistochemistry and histopathology four hours after cerebral hypoxia/ischemia. Sections of the parietal cortex from piglet brains after ischemic injury (A-E) and from uninjured control animal (I and F), were subjected to antigen retrieval in citrate buffer and stained with anti-uPA monoclonal antibody (5 μg/ml) (Panels, A-D and F) or with non-immune mouse IgG₁ as a negative control (Panel E), secondary biotinylated anti-mouse IgG (1:200), followed by incubation with HRP-conjugated streptavidin. Magnification shown is 100x for Panels A, B, G, and I, 200X for panels, C, E, F, and H, 400x for Panel D. Adjacent sections from the same brains exposed to ischemic injury (Panels G-H) and from uninjured controls (Panel I), were stained by H&E for histological inspection. These data reflect an n of 2 per experimental group.

Figure 2

Influence of hypoxia/ischemia (H/I) on CSF uPA (ng/ml) as a function of time post insult (hours) in vehicle (saline), IgG (10 μg/ml), RAP (10^-7 M), and anti-LRP antibody (anti LRP Ab) (10 μg/ml) pretreated pigs, n=6 for all except n=3 for IgG. IgG, RAP and anti LRP Ab were administered after the 0 time sample, but 30 min prior to H/I. ^*P<0.05 versus corresponding 0 time value ⁺P<0.05 versus corresponding vehicle value.

Figure 3

Influence of hypercapnia (lo, hi; pCO₂ of 50-55 and 70-75 mm Hg), hypotension (mod, sev; 25 and 45% reductions in mean arterial blood pressure), and isoproterenol (10^-8, 10^-6 M) (panels A-C) on pial artery diameter. Conditions are before (control), 1h after hypoxia/ischemia (p_O2 of 35 mm Hg for 10 min followed by global cerebral ischemia for 20 min) (H/I), 1h after H/I pretreated with IgG (10 μg/ml) 30 min prior to H/I, 1h after H/I pretreated with RAP (10^-7 M), 1h after H/I pretreated with anti LRP antibody (anti LRP Ab) (10 μg/ml), and 1h after H/I pretreated with U 0126 (10^-6 M), n=6 for all except n=3 for IgG. Baseline small artery diameter for control, H/I, H/I + IgG, H/I + RAP, H/I + anti LRP Ab, and H/I + U 0126 were 144 ± 11, 119 ± 9, 118 ± 8, 138 ± 6, 136 ± 7, and 132 ± 7 μm, respectively. ^*P<0.05 versus corresponding control value ⁺P<0.05 versus corresponding non pretreated H/I value.

Figure 4

Phosphorylation of ERK MAPK in cortical periarachnoid CSF was determined prior to hypoxia/ischemia (H/I) (0 min), as a function of time (hours, h) after H/I, and 4h after H/I and additional exogenous administration of uPA (10^-7 M) in vehicle, RAP (10^-7 M), anti LRP Ab (10 μg/ml) and U 0126 (10^-6 M) pretreatment animals, n=6. Data expressed as percent of control by ELISA determination of phospho ERK MAPK and total ERK MAPK isoforms and subsequent normalization to total form. ^*P<0.05 compared with corresponding 0 time value ⁺P<0.05 compared with corresponding 4h non uPA treated value

Figure 5

Influence of uPA (10^-9, 10^-7 M) on A: pial artery diameter and B: CSF phosphorylated ERK MAPK (percent of control) in the absence (vehicle) and presence of U 1026 (10^-6 M) and suPAR (10^-7 M), n=6. ^*P<0.05 compared with corresponding vehicle value.

Figure 6

Immunhistochemistry for phosphorylation of ERK MAPK four hours after cerebral hypoxia/ischemia. Sections of the parietal cortex from piglet brains after ischemic injury (A,B, and E) and from uninjured control animal (C and D) were stained with phospho-p44/42 MAPK rabbit monoclonal antibody (Panels A-D) or with rabbit IgG as a negative control (Panel E). Magnification shown is 100X for Panels A and C, 400X for Panels D,B, and E, X1000 for insert in Panel B. These data reflect an n of 2 per experimental group.

Figure 7

Western analysis of CSF for phospho–p44/42 ERK MAPK at 4h and 0h of hypoxia/ischemia in suPAR, vehicle, and U 0126 pretreated animals, n=2 for each experimental group.

Figure 8

Proposed mechanism relating uPA, LRP, and ERK MAPK to cerebral hemodynamic outcome after hypoxia/ischemia.