Impact of Acute Myocardial Infarction on Neuroinflammation and Alzheimer's Disease Pathology in a Transgenic Mouse Model

Introduction: Myocardial infarction (MI) and progressive heart failure are associated with a higher incidence of cognitive impairment and Alzheimer's dementia. Acute and recurrent neuroinflammation secondary to MI and chronic heart failure may contribute to neurodegeneration. Transgenic mice overexpressing amyloid precursor protein (APP) and presenilin-1 (PS1) develop amyloid-β plaque in the brain early in adulthood. We hypothesized that MI and resultant neuroinflammation would increase susceptibility to amyloid deposition and cognitive impairment. 

Methods: Transgenic APP/PS1 and non-transgenic (nTg) littermate male mice underwent permanent ligation of the left coronary artery (n= 15 APP/PS1, n= 14 nTg) or sham surgery (n= 10 APP/PS1, n= 10 nTg) at 9wk of age. Whole body positron emission tomography (PET) imaging of the mitochondrial translocator protein (TSPO) using ¹⁸F-GE180 determined cardiac and neuroinflammation at 7d, 3wk and 6wk after surgery. Cardiac magnetic resonance imaging (MRI) assessed contractile function at 6wk and myocardial perfusion imaging determined infarct size. Imaging findings in the heart and brain were verified by immunohistology. At the terminal endpoint, brain amyloid- β content was measured by commercial immunoassay and brain cytokines were quantified.

Results: Coronary artery ligation resulted in comparable infarct sizes in both APP/PS1 and nTg MI mice (%LV: 38±7 vs 33±12, p=0.299). At 6wk, left ventricle ejection fraction (LVEF) was consistently impaired in both genotypes compared to matched sham operated animals (LVEF: 23±8 vs 27±13 vs 53±9%, p<0.001). At 7d after MI, infarct territory TSPO PET signal was elevated compared to sham-operated animals in APP/PS1 mice (%ID/g: 13.9±3.2 vs 5.2±0.9, p<0.001) and nTg mice (14.7±3.1 vs 6.5±2.8, p<0.001). There was no difference between genotypes (p=0.46). A modest increase in TSPO PET signal was observed in the brain at 7d after MI in nTg mice (2.0±0.2 vs 1.8±0.3, p=0.050). Brain TSPO signal post MI was similar in APP/PS1 mice, but did not show a significant elevation relative to genotype-matched sham animals at 7d (2.1±0.4 vs 1.9±0.2, p=0.61).&nbsp; At 6wk after MI, TSPO signal from remote non-infarct myocardium was consistently higher in both genotypes compared to sham animals (APP/PS1 7.7±1.9 vs 5.8±1.0, p=0.004; nTg 7.9±2.2 vs 5.1±0.9, p<0.001). While the global brain TSPO signal tended to be elevated in both genotypes compared to sham, there was no difference between APP/PS1 and nTg mice at 6wk after MI (%ID/g: 2.0±0.3 vs 2.1±0.3, p=NS). Notably, APP/PS1 at 6wk after MI exhibited higher cerebral inflammatory cytokine levels compared to nTg MI mice including interleukin-1&beta; (+100±27%, p=0.002) and tumour necrosis factor-&alpha; (+35±6%, p=0.025). However, there was no difference in amyloid-&beta; content after MI compared to sham in transgenic mice (pg/ml: 6.4x10⁵±3.8x10⁴ vs 6.7x10⁵±7.2x10³, p=NS).

Conclusions: MI evokes acute neuroinflammation in both non-transgenic and transgenic mice prone to early onset Alzheimer's disease. Higher levels of cerebral proinflammatory cytokines in APP/PS1 mice post-MI suggest a more profound inflammatory response despite unchanged TSPO signal. Whether this variance in neuroinflammation will affect downstream amyloid pathology warrants further investigation including longer-term follow-up and behavioural studies.