Alzheimer's disease (AD) is a devastating dementia of late life that is correlated with a region-specific neuronal cell loss. Despite progress in uncovering many of the factors that contribute to the etiology of the disease, the cause of the nerve cell death remains unknown. One promising theory is that the neurons degenerate because they reenter a lethal cell cycle. This theory receives support from immunocytochemical evidence for the reexpression of several cell cycle-related proteins. Direct proof for DNA replication, however, has been lacking. We report here the use of fluorescent in situ hybridization to examine the chromosomal complement of interphase neuronal nuclei in the adult human brain. We demonstrate that a significant fraction of the hippocampal pyramidal and basal forebrain neurons in AD have fully or partially replicated four separate genetic loci on three different chromosomes. Cells in unaffected regions of the AD brain or in the hippocampus of nondemented age-matched controls show no such anomalies. We conclude that the AD neurons complete a nearly full S phase, but because mitosis is not initiated, the cells remain tetraploid. Quantitative analysis indicates that the genetic imbalance persists for many months before the cells die, and we propose that this imbalance is the direct cause of the neuronal loss in Alzheimer's disease.
Key words: cell cycle; PCNA; cyclin B; hippocampus; nucleus basalis; FISH (fluorescent in situ hybridization); neurodegenerationThe death of nerve cells during early development is a constructive part of pruning and shaping the emerging nervous system. The loss of neurons during adult life, however, is a pathological process that produces behavioral disorders and potentially the death of the organism. Recently, several laboratories have offered evidence that an attempt by the cell to reenter a mitotic cycle precedes many instances of neuronal death (Smith and Lippa, 1995;Arendt et al., 1996Arendt et al., , 1998aVincent et al., 1996Vincent et al., , 1997Vincent et al., , 1998McShea et al., 1997McShea et al., , 1999Nagy et al., 1997aNagy et al., , 1998Busser et al., 1998;Smith et al., 1999). Together, these studies suggest that the paradoxical association of the generative process of cell division with the degenerative process of cell death is found at all stages of the existence of a neuron. The prohibition against cell division begins at the earliest stages of maturation of a neuron. Hindbrain neurons in mice lacking the retinoblastoma tumor suppressor gene are unable to avoid reentering the cell cycle and die by apoptosis immediately after their emigration from the ventricular zone (Lee et al., 1994). Neuronal cell death later in development has also been linked to ectopic cell cycling. Using mutant models of target-related cell death, Herrup and Busser (1995) showed that target-deprived neurons initiated the synthesis of cell cycle enzymes [cyclin D and proliferating cell nuclear antigen (PCNA)] and incorporated bromodeoxyuridine (BrdU) into high molecular weight DN...