Social stress exacerbates stroke outcome by suppressing Bcl-2 expression

Abstract

The relationship between stressful life events and the onset of disease is well documented. However, the role of psychological stress as a risk factor for life-threatening cerebrovascular insults such as stroke remains unspecified, but could explain individual variation in stroke outcome. To discover the mechanisms through which psychological stress may alter stroke outcome, we modeled the effects of chronic social intimidation and stress on ischemia-induced bcl-2 expression and early neuronal cell loss resulting from cerebral artery occlusion in mice (C57BL/6). The bcl-2 protooncogene promotes cell survival and protects against apoptosis and cellular necrosis in numerous neurodegenerative disorders, including stroke. In our study, male mice were chronically exposed to aggressive social stimuli before induction of a controlled, mild ischemic insult. Stressed mice expressed approximately 70% less bcl-2 mRNA than unstressed mice after ischemia. In addition, social stress greatly exacerbated infarct in wild-type mice but not in transgenic mice that constitutively express increased neuronal bcl-2. Despite similar postischemic concentrations of corticosterone, the major stress hormone in mice, high corticosterone concentrations were significantly correlated with larger infarcts in wild-type mice but not bcl-2 transgenic mice. Thus, enhanced bcl-2 expression offsets the potentially deleterious consequences of high postischemic plasma corticosterone concentrations. Taken together, these data demonstrate that stressful prestroke social milieu strongly compromises an endogenous molecular mechanism of neuroprotection in injured brain and offer a new behavioral target for stroke therapy.

Figure 1

Stress alters postischemic expression of bcl-2 mRNA measured by RNase protection assay. Male mice exposed to social stress exhibit bcl-2 mRNA levels that are less than 31% of control levels in the ischemic hemisphere 24 h after transient experimental stroke. Data (n = 8/group) were analyzed by using an unpaired t test and are presented as mean ± SEM. An asterisk indicates statistical significance at P < 0.05.

Figure 2

Postischemic corticosterone concentrations (ng/ml) and bcl-2 mRNA levels are significantly correlated in stressed (n = 8) and control (n = 8) mice. Bcl-2 mRNA is measured by RNase protection assay.

Figure 3

Increased bcl-2 expression protects against stress-induced exacerbation of cerebral ischemia. Social stress increases infarct size in WT mice but not transgenic mice that express increased bcl-2. Data (n = 5–6/group) are presented as mean ± SEM. An asterisk indicates statistical significance at P < 0.05 compared with the other three experimental groups.

Figure 4

(a) Blood corticosterone concentrations (ng/ml) are significantly lower in bcl-2 transgenic mice than WT controls at baseline but not 1 h after social stress. (b) Twenty-four hours after transient focal ischemia, blood corticosterone concentrations are similar between genotypes of stressed and unstressed animals. Data (n = 5–7/group) are presented as mean ± SEM. An asterisk indicates a statistical significance at P < 0.05 between WT and bcl-2++ mice at baseline.

Figure 5

Postischemic corticosterone concentrations (ng/ml) and percent infarct (relative to the contralateral hemisphere) are significantly correlated in WT mice (solid line) but not transgenic mice that constitutively express excess bcl-2 (dotted line). ●, WT data; ○, bcl-2 transgenic mice data. Data include both stressed and unstressed animals in each genotype.