A white matter stroke model in the mouse: Axonal damage, progenitor responses and MRI correlates

Abstract

Subcortical white matter stroke is a common stroke subtype but has had limited pre-clinical modeling. Recapitulating this disease process in mice has been impeded by the relative inaccessibility of the subcortical white matter arterial supply to induce white matter ischemia in isolation. In this report, we detail a subcortical white matter stroke model developed in the mouse and its characterization with a comprehensive set of MRI, immunohistochemical, neuronal tract tracing and electron microscopic studies. Focal injection of the vasoconstrictor endothelin-1 into the subcortical white matter produces an infarct core that develops a maximal MRI signal by day 2, which is comparable in relative size and location to human subcortical stroke. Immunohistochemical studies indicate that oligodendrocyte apoptosis is maximal at day 1 and apoptotic cells extend away from the stroke core into the peri-infarct white matter. The amount of myelin loss exceeds axonal fiber loss in this peri-infarct region. Activation of microglia/macrophages takes place at 1 day after injection near injured axons. Neuronal tract tracing demonstrates that subcortical white matter stroke disconnects a large region of bilateral sensorimotor cortex. There is a robust glial response after stroke by BrdU pulse-labeling, and oligodendrocyte precursor cells are initiated to proliferate and differentiate within the first week of injury. These results demonstrate the utility of the endothelin-1 mediated subcortical stroke in the mouse to study post-stroke repair mechanisms, as the infarct core extends through the partially damaged peri-infarct white matter and induces an early glial progenitor response.

Introduction

Subcortical white matter stroke constitutes 15–25% of all stroke subtypes (Bamford et al., 1991) and results in white matter lesions. These lesions encompass small infarcts in deep penetrating vessels in brain, as well as ischemic lesions in end-arterial regions of subcortical white matter (Matsusue et al., 2006, Gouw et al., 2008, Wardlaw, 2008). The infarctions in subcortical white matter are closely related to true microvascular stroke and are represented on MRI imaging as lacunar infarctions with a small stroke cavity as well as ischemic white matter hyperintensities (Gouw et al., 2008). Despite an advanced pre-clinical literature of animal modeling in large artery, “gray matter” stroke, there are few animal models of white matter ischemia. A principle problem in modeling white matter stroke in the rodent brain is that it has substantially less white matter than in higher mammals and humans.

A lack of an appropriate model for subcortical stroke presents a critical gap in stroke basic science research, as the mechanisms of cell death and of repair are likely to differ in white matter stroke from large artery gray matter strokes. For instance, oligodendrocytes undergo a different time course of cell death than neurons in large artery stroke models (Pantoni et al., 1996). Also, white matter astrocytes display a differential sensitivity to ischemic injury than those derived from gray matter (Shannon et al., 2007). In addition to this biology of cell death, the biology of the glial progenitor response is also different in cortical versus subcortical regions, and this may contribute to differing degrees of repair. Oligodendrocyte progenitor cells (OPCs) are widely scattered throughout cortical and subcortical regions in the adult brain. Using genetic labeling, it has recently been shown that OPCs in the subcortical white matter of the adult turnover at a higher rate compared to cortical OPCs (Dimou et al., 2008); and that cortical OPCs are fewer in number and experience a block at the progenitor or pre-oligodendrocyte stage (Dimou et al., 2008). Futhermore, this limited progenitor response of cortical OPCs persists after brain injury with few cortical OPCs producing mature oligodendrocytes (Dimou et al., 2008). Because of these apparently different mechanisms of cell death and of repair in cortical vs. white matter ischemia, specific animal models of subcortical white matter stroke are urgently needed. A white matter stroke model developed in the mouse would provide the added advantage that mouse genetic tools can be applied to the labeling and molecular analysis of the cells involved in white matter stroke and repair.

In this report we describe a subcortical stroke model developed in the mouse by using the vasoconstrictor peptide endothelin-1 (ET-1). ET-1 is not directly neurotoxic (Nikolov et al., 1993) and produces local vasoconstriction and loss of blood flow for up to three hours at the injection site (Fuxe et al., 1992, Hughes et al., 2003). The regulation of regional blood flow by ET-1 has been used to produce several different large artery, focal cortical or striatal infarcts. Focal injection of ET-1 into the posterior limb of the internal capsule in the rat produces a small infarction resembling a lacune (Frost et al., 2006, Lecrux et al., 2008). In the present study, ET-1 was microinjected into the subcortical white matter of the frontal lobe of the mouse, and the patterns of cell death, inflammation, gliosis, axon and myelinated fiber loss and glial progenitor responses were determined using immunohistochemical, tract tracing and electron microscopic techniques. This model produces a focal necrotic cavity in subcortical white matter and an adjacent area of axonal damage, myelin loss and oligodendrocyte gliogenesis.