Cell-cycle reentry and cell death in transgenic mice expressing nonmutant human tau isoforms

Abstract

Mutations in the microtubule-associated protein tau gene have been linked to neurofibrillary tangle (NFT) formation in several neurodegenerative diseases known as tauopathies; however, no tau mutations occur in Alzheimer's disease, although this disease is also characterized by NFT formation and cell death. Importantly, the mechanism of tau-mediated neuronal death remains elusive. Aged mice expressing nonmutant human tau in the absence of mouse tau (htau mice) developed NFTs and extensive cell death. The mechanism of neuron death was investigated in htau mice, and surprisingly, the presence of tau filaments did not correlate directly with death within individual cells, suggesting that cell death can occur independently of NFT formation. Our observations show that the mechanism of neurodegeneration involved reexpression of cell-cycle proteins and DNA synthesis, indicating that nonmutant tau pathology and neurodegeneration may be linked via abnormal, incomplete cell-cycle reentry.

Figure 1.

Pathology in aging htau mice was characterized by cell-body accumulation of phosphorylated and conformationally altered tau that aggregated into classic paired helical filaments and neurofibrillary tangles. An antibody to tau phosphorylated at serine 202 (CP13) labeled neurons in 16-month-old htau mouse piriform cortex (A) and hippocampus (B) in a manner that resembled that in neurofibrillary tangles of human brain (C). Neurons were also positive for MC1 (an AD conformation-specific antibody) in the piriform cortex (D) and hippocampus (E) of 16-month-old htau mice. PHF1 (Ps396/404) antibodies labeled distorted cortical neurons in the neocortex (F) and hippocampus (G) of 16-month-old htau mice. Thioflavine-S, a fluorescent histochemical stain commonly used to detect neurofibrillary tangles, did not detect any in control mice (H) but did reveal intensely labeled neurons in htau mice shown here in the piriform cortex (I) and hippocampus (J) of 15-month-old htau mice. Scale bars, 10 μm. Electron microscopy of Sarkosyl-insoluble material isolated from 16-month-old htau mice that were immunogold labeled with antibodies specific for phosphorylations on tau at serine 202 (CP13; K) and serines 396/404 (PHF1; L) confirmed the presence of paired helical filaments. Very similar results are found in mice aged 16-22 months, and representative data are shown.

Figure 2.

Gross changes in old htau brains, including abnormal neuronal morphology, a decrease in cortical thickness, and a dramatic enlargement of ventricle size. Irregularly shaped neurons were detected with CP13 (A-C) and MC1 (D, E). Note the presence of enlarged vacuoles, irregularly shaped membranes, and distorted processes (A-C). Additionally, some neurons appeared ballooned (D), and still others had the character of late-stage tangles (E). Comparison of cortical thickness from matched brain regions of 1-month-old (F), 10-month-old (G), and 14-month-old (H) htau mice indicated a decrease in thickness and an obvious reduction in the number of detectable nuclei with age. A comparison of ventricle size in level-matched coronal sections between htau mice aged 8 months (I) and 18 months (J) revealed a dramatic increase in the older mice. The red oval delineates the piriform cortex. Scale bars: A-E, 10 μm; F-H, 50 μm; I, J, 250 μm.

Figure 3.

Quantification of age-related neuronal loss in the piriform cortex of htau mice was performed using the optical fractionator method of stereology. A, Neuronal counts and region volume estimates in the piriform cortex provided for the individual htau mouse examined, aged 8 months (n = 5) or 17 months (n = 5). CE, Coefficient of error. B, Comparison of estimated total of neurons between the two ages of htau mice revealed a dramatic reduction in the total number of neurons between 8 and 18 months, from an average of 565,000 to 242,000, a decrease of 67.2% (p < 0.009). There was also a 23% reduction (p < 0.002) in the estimated volume of the piriform cortex between these age groups.

Figure 4.

Nuclear abnormalities in neurons in the older htau mice detected by electron microscopy suggest multiple pathways of cell death. A, B, Features of apoptosis, including nuclear breakdown, chromatin condensation, membrane blebbing, and cytoplasmic shrinkage in a 22-month-old htau mouse cortex (A) and striatum (B). Chromatin condensation into dark dispersed bodies in the absence of cell shrinkage or cell fragmentation was also detected in 22-month-old htau hippocampus (C) and cortex (D). Other neurons in the cortex (E) and thalamus (G) of htau mice displayed no prominent chromatin condensation but extensive organelle swelling and cytoplasmic vacuolization suggestive of cell lysis. Local damage, swelling, vacuolization, and demyelination were visible in axonal processes (H, I), and many of these processes contained filamentous aggregates. Neurons were also detected in the 22-month-old htau brain that had little damage (F) and displayed dispersed chromatin and intact organelles but that had accumulated aggregates. Note that the neurons that displayed nuclear breakdown did not have significant accumulations of filamentous tau (A-F). Scale bars: A-G, 2 μm; H, I, 5 μm.

Figure 5.

DNA fragmentation was detected in the htau mice using TUNEL on paraffin sections. No TUNEL reactivity was detected in a 15-month-old control mouse brain (A), whereas TUNEL-positive cells were detected in an age-matched htau mouse hippocampus (B) and piriform cortex (C). Numbers of positive cells were low: two to three positive cells per section in mice aged 15-18 months. Scale bars, 10 μm.

Figure 6.

Cell death in htau mice did not occur through a classic caspase-dependent apoptotic pathway. A Western blot of brain homogenates of a time course of htau mice (ht) (2, 5, 10, 13, 16, and 17 months of age) and an old aged wild-type (wt) mouse (20 months) probed with an antibody that detects both full length and cleaved (active) forms of caspase-3 revealed no cleavage of caspase-3 (A). Active caspase-3 would be detected as a band at 17/19 kDa. Similar negative results were obtained for other members of the caspase family, including caspase-6, -7, -8, -9, and -10 (data not shown). B, Blots were also probed with SP-14 and antibody that recognizes SNAP-25 to control for loading.

Figure 7.

The S-phase associated cell-cycle molecules ki67 and cyclin D₁ are detectable in htau neurons. Brains from age-matched wild-type mice were not reactive with antibodies to ki67 (A), but sections from htau mice showed numerous positive neurons in the cortex (B) and thalamus (C). In wild-type mice, there was no reactivity with cyclin D antibodies (D), and select populations of neurons in htau mice were positive, shown here in the cortex (E) and thalamus (F) of a 22-month-old mouse. Scale bars, 10 μm. Very similar results have been obtained in 12- and 18-month-old htau mice.

Figure 8.

Cell cycling and DNA synthesis in the htau mouse brain was detected with antibodies to PCNA and ki67 and by incorporation of the thiamine analog BrdU. No reactivity with the PCNA antibody was detected in wild-type mice (A) shown here at 14 months, but reactivity was detected in an age-matched htau mouse brain (B). Double labeling with PCNA (C, brown) and NeuN (C, purple) (a neuronal marker) showed that both neurons and glial cells expressed PCNA. Most cells that were positive for PCNA (D, brown staining) were not positive for CP13, reactive to phospho-serine 202 on tau (D, purple staining), with some rare exceptions. Incorporation of the thiamine analog BrdU (E, F, H, I), indicating DNA synthesis, was detected in some postmitotic brain regions of 12- and 18-month-old htau mice, including the somatosensory cortex (H, brown). Double labeling with antibodies to BrdU and ki67 (H, purple) revealed some cells that were positive for either marker and other cells that were positive for both (I).