In vivo neuroprotection in ischemic stroke by activated protein C requires β-arrestin 2

Abstract

The protease activated protein C (APC) and its variants provide neuroprotection for murine ischemic stroke and mortality reduction for murine sepsis. For these actions, APC's in vivo mechanism of action, similar to in vitro studies using cultured cells, involves protease activated receptor 1 (PAR1)-mediated biased signaling. APC/PAR1 signaling in vitro requires β-arrestin 2, an intracellular scaffold protein, and β-arrestin 2-initiated signaling can alter diverse intracellular signaling pathways. This study used a proximal transient middle cerebral artery occlusion model to study the neuroprotective actions of the signaling-selective APC variant, 3K3A-APC, in β-arrestin 2-deficient (Arrb2 ^-/-) mice. Based on quantitation of brain injuries, 3K3A-APC significantly limited brain injury in control mice to relatively small, localized areas, whereas 3K3A-APC's protection was lost in Arrb2 ^-/- mice. Thus, the major in vitro mechanism of action that requires β-arrestin 2 for APC/PAR1 biased signaling is central to the in vivo mechanism of action for APC's neuroprotection.

Graphical abstract

Figure 1.

Time course for experimental stroke and 3K3A-APC treatment. Mice were subjected to proximal tMCAo for 45 minutes starting at 0 minutes, as described elsewhere. Mice were treated with either vehicle or recombinant murine 3K3A-APC (0.8 mg/kg) intraperitoneally at 10 minutes and at 4 hours after the initiation of tMCAo. In vivo MR imaging (MRI) was performed 24 hours after tMCAo initiation. MRI was then followed by tissue collection and analysis.

Figure 2.

3K3A-APC protection against ischemic stroke in vivo requires β-arrestin 2. (A) Representative T2-weighted (structural) coronal MR scans showing postischemic injury area (hyperintense signal) in the ipsilateral cortex 24 hours after tMCAo in control and Arrb2^–/– mice on C57BL/6J genetic background treated with vehicle or 3K3A-APC (0.8 mg/kg intraperitoneally 10 minutes and 4 hours after the start of tMCAo). Postischemic injury areas are delineated by yellow dashed lines. (B-D) Injury (B), infarct (C), and edema volumes (D) 24 hours after tMCAo in control and Arrb2^–/– mice treated with vehicle or 3K3A-APC as above in panel A. Control + vehicle (n = 7); Control + 3K3A-APC (n = 6); Arrb2^–/– + vehicle (n = 7); and Arrb2^–/– + 3K3A-APC (n = 6). All volume measurements derived from structural T2-weighted MR images. Data in panels B-D are shown as dot plots representing single points per mouse. Each bar represents mean ± standard error of the mean. Statistical significance was determined by 1-way analyses of variance followed by Tukey multiple comparison test. ∗P < .05; ∗∗P < .01; ∗∗∗P < .001. ns, nonsignificant. The panels B-D in Figure 2 were modified to have the same y-axes, but the data are the same as originally submitted.

Figure 3.

3K3A-APC reduces postischemic injury areas in control mice but not in Arrb2^–/– mice. (A) Representative images of cresyl violet staining of brain sections 24 hours after tMCAo from control and Arrb2^–/– mice on a C57BL/6J genetic background treated with vehicle or 3K3A-APC as in Figure 1. Injury areas are delineated by yellow dashed lines. Mice used for representative images in panel A are the same as in Figure 2A. The tissue sections for cresyl violet staining were taken from the same locations relative to the bregma as for T2-weighted MR scans in Figure 2; scale bar, 5 mm. (B) Incidence and topography of injury area at the level of optic chiasm (ie, image for the bregma +1.0 mm) in control mice and Arrb2^–/– mice treated with vehicle or 3K3A-APC. Control + vehicle (n = 6); control + 3K3A-APC (n = 6); Arrb2^–/– + vehicle (n = 5); Arrb2^–/– + 3K3A-APC (n = 5).