Fig. 3

The relationship between neuronal tubulin levels and NFTs. Most neurons with NFT (mean 78 %) did not demonstrate axonal processes stained for tubulins (a, arrowheads left panels). However, some neurons with NFT (blue), had tubulin staining in the cell body and axonal process (A, arrows). Every tubulin modification was retained in the axonal process in some NFT-bearing neurons (A, right panels). Scale bars = 50 μm. Measuring the lengths of the cell bodies and axonal processes in the NFT and normal surrounding neurons in the CA1 region found the NFTs were indeed significantly shorter (b) in all AD cases and one control case. Taken together, the NFT in both AD and control cases, were significantly shorter when compared to the normal neuron population (c), yet no difference in length was found when comparing AD and control NFT, or AD and control normal neurons, suggesting the loss of axonal process length is a result of NFT formation and not disease. *p < 0.001, **p < 0.05