Intraneuronal Aβ immunoreactivity is not a predictor of brain amyloidosis-β or neurofibrillary degeneration

Amyloid beta (A beta) immunoreactivity in neurons was examined in brains of 32 control subjects, 31 people with Down syndrome, and 36 patients with sporadic Alzheimer's disease to determine if intraneuronal A beta immunoreactivity is an early manifestation of Alzheimer-type pathology leading to fibrillar plaque formation and/or neurofibrillary degeneration. The appearance of A beta immunoreactivity in neurons in infants and stable neuron-type specific A beta immunoreactivity in a majority of brain structures during late childhood, adulthood, and normal aging does not support this hypothesis. The absence or detection of only traces of reaction with antibodies against 4-13 aa and 8-17 aa of A beta in neurons indicated that intraneuronal A beta was mainly a product of alpha- and gamma-secretases (A beta(17-40/42)). The presence of N-terminally truncated A beta(17-40) and A beta(17-42) in the control brains was confirmed by Western blotting and the identity of A beta(17-40) was confirmed by mass spectrometry. The prevalence of products of alpha- and gamma -secretases in neurons and beta- and gamma-secretases in plaques argues against major contribution of A beta-immunopositive material detected in neuronal soma to amyloid deposit in plaques. The strongest intraneuronal A beta(17-42) immunoreactivity was observed in structures with low susceptibility to fibrillar A beta deposition, neurofibrillary degeneration, and neuronal loss compared to areas more vulnerable to Alzheimer-type pathology. These observations indicate that the intraneuronal A beta immunoreactivity detected in this study is not a predictor of brain amyloidosis or neurofibrillary degeneration. The constant level of A beta immunoreactivity in structures free from neuronal pathology during essentially the entire life span suggests that intraneuronal amino-terminally truncated A beta represents a product of normal neuronal metabolism.