Fig. 2

Representative EM images highlight damaged (dM) and intact mitochondria (iM) in AD (A, B) and non-AD axon terminals (C, D). Presynaptic terminals are colored light blue and the dendritic spines are colored light yellow if present. The control cases display a trend toward more presynapses with mitochondria (E, p = 0.15), however, the percentage of presynaptic axons with damaged mitochondria is significantly greater in the AD group (F, p < 0.01). The number of mitochondria per presynaptic axon is lower in AD cases (G, p = 0.07), and further analysis finds that the mean % of preysnapses lacking mitochondria trends higher in the AD group (p = 0.13), while the control cases have a higher % with more than one mitochondria (p = 0.10) (H). Mitochondria aspect ratio is not different between the AD and control cases (I, p > 0.05). Damaged mitochondria have lower aspect ratios than intact mito in both AD and non-AD groups (J, p < 0.05 AD, p < 0.01control group). Mitochondria size only trends higher in the AD cases (K, p = 0.08). Yet, intact mitochondria in AD presynaptic boutons are larger than those in non-AD cases (L, p < 0.01). While the mean size of AD mitochondria is unchanged between intact and damaged groups (L, p > 0.05), the damaged mitochondria in control cases are larger than their intact counterparts (L, p < 0.0001). Data are Means ± S.E.M. of six AD and six control cases. Student t test (E–I, K) or one-way analysis of variance (J, L). *p < 0.05, **p < 0.01, ****p < 0.0001