Age-dependent impairment of somatosensory response in the amyloid precursor protein 23 transgenic mouse model of Alzheimer's disease

Abstract

Quantitative functional magnetic resonance imaging was applied to characterize brain function in amyloid precursor protein 23 (APP23) transgenic mice, which reproduce the neuropathological alterations associated with Alzheimer's disease. Electrical stimulation of the paw led to cerebral blood volume increases in the contralateral somatosensory cortex. In APP23 mice this hemodynamic response decreased with increasing age of the animal and with increasing stimulus amplitude as compared with wild-type animals. The age-dependent dysfunction in APP23 mice may be attributed in part to a compromised cerebrovascular reactivity. Quantitative functional brain mapping that uses standardized sensory inputs should allow for assessment of disease progression and therapy response (e.g., passive immunization against beta-amyloid) in patients also.

Figure 1.

a, Statistically derived ROIs with a correlation coefficient threshold of 0.3 and a minimum cluster size of 10 pixels, superimposed on a horizontal high-resolution image. Intracerebral ROI (R5) is confined to cortical representation of the S1 region, and minor activations (R6, R7) also are observed at the level of hippocampus (approximately -3 mm posterior relative to Bregma). b, ROIs used for quantitative signal analysis in normal mice overlaid on a high-resolution image: the activated area (ROI) comprising the S1HL region contralateral to the stimulated hindpaw as obtained from the statistical analysis, the corresponding ROI on the ipsilateral side, and a third area comprising ipsilateral hippocampal structures serving as a signal reference (control ROI). c, Temporal profiles of ΔCBV_% for the three ROIs defined in b. ΔCBV_% responses for the different stimulation periods (indicated by bars) increased with progressive current amplitude. In-between stimulation blocks (off-periods) ΔCBV_% reached values of the order of 2-4% over baseline. Data are given as the mean ± SEM.

Figure 2.

Statistically derived ROIs (correlation coefficient threshold of 0.3 and a minimum cluster size of 10 voxels) overlaid on coronal spin-echo images (0.96 mm posterior relative to the Bregma) for a 25-month-old control (a) and an age-matched APP23 mouse (b). The ROIs are consistent with the known topographic location of the murine hindlimb region within the primary somatosensory cortex S1.

Figure 3.

Time course of ΔCBV_% calculated from ROIs (defined in Fig. 2) for 6-month-old (a), 13-month-old (b), and 25-month-old (c) APP23 mice and age-matched control littermates. The bars at the bottom of the panels represent the stimulation design (repetitive stimulation periods of 105 sec with progressive current amplitudes of 0.5, 1, and 2 mA). Data are given as the mean ± SEM. All control littermates show an increased ΔCBV_% response with increasing current amplitude. An analogous response pattern is observed in 6- and 13-month-old APP23 mice, with a tendentiously reduced response for the highest current amplitude (2 mA). In 25-month-old APP23 mice, however, no CBV dependence on current amplitude is found; in fact, values at 2 mA were decreased as compared with lower current strengths. d, On-line monitoring of blood carbon dioxide of 25-month-old control and age-matched APP23 mice. Averaged PtcCO₂ values (in mmHg) during the fMRI experiments for control (39.1 ± 3) and APP23 mice (38.4 ± 2) did not show any alterations during electrical stimulation of the hindpaw.

Figure 4.

Maximal ΔCBV_% response of APP23 mice, relative to the corresponding wild-type values. Data are obtained from statistically derived ROI on contralateral side to the stimulated hindpaw of 6-, 13-, and 25-month-old control and APP23 mice for current amplitudes of 0.5, 1, and 2 mA, respectively. The response decreases with both increasing age of the animals and increase in the amplitude of the stimulation current. Individual comparison between the groups revealed significant differences between control and APP23 mice only at an age of 25 months for the 2 mA current strength (p < 0.029) and between 6- and 25-month-old APP23 mice (p < 0.003; ANOVA, Tukey's test).