Dickson, Dennis W.

TAR DNA binding protein‐43 (TDP‐43)‐immunoreactive neuronal cytoplasmic inclusions (NCIs), characteristically associated with frontotemporal lobar degenerations (FTLD‐U and FTLD‐MND), have been found in the hippocampal dentate granule cells of about 20% of cases with advanced Alzheimer disease (AD). The extent and distribution of TDP‐43 pathology in brain regions other than hippocampus in AD is not known, especially in AD cases without hippocampal NCIs. Sections of hippocampus, amygdala, neocortex, and medulla from patients with clinical dementia and pathologic features of AD were analyzed for the presence of TDP‐43‐immunoreactive NCIs. 22 cases of advanced AD (Braak Stage V‐VI) without hippocampal NCIs and 23 cases of advanced AD with hippocampal NCIs were included. 6 hippocampal‐negative cases (27%) had TDP‐43 positive NCIs in the amygdala. All 23 cases of hippocampal‐positive cases had numerous TDP‐43 NCIs in the amygdala, with decreasing density of NCIs in the other brain regions. In summary, TDP‐43 immunoreactive NCIs are found predominantly in the medial temporal structures in advanced AD, and the amygdala appears to be the area most vulnerable.

Objective: To determine if subcortical, brainstem and cerebellar lesion pattern characteristics of the different types of tau‐negative frontotemporal lobar degeneration (FTLD) would be similar to neocortical and hippocampal patterns [Type 1 has neuronal cytoplasmic inclusions (NCI) and dystrophic neurites (DN), type 2 mainly DN, and type 3 mainly NCI (Mackenzie et al., 2006)]. Design/Methods: We analyzed 40 cases of tau‐negative FTLD from Mayo Clinic, Florida. For each case, 12 standardized sections covering 16 different brain regions or nuclei were sampled including neocortex, hippocampus, subcortical grey, brainstem and cerebellum. TDP‐43 immunohistochemistry was performed and abnormal immunoreactivity semiquantitated. Results: Thirty‐six cases were classified into one of the three types. For subcortical grey and brainstem regions, we did not see predominance of DN in type 2, and the density of NCI and DN were similar in type 3. Type 2 cases had significantly more NCI in basal ganglia compared to the other types (p<0.05). We found a trend that fine DN in hippocampal CA1 region was more frequent in type 1 compared to type 2 and 3 (p=0.07). A novel finding was that NCI were Pick body‐like in type 2 but granular‐like in type 3. Conclusion: This study extends the patterns of pathology previously described in the three variants of tau‐negative FTLD and shows how subcortical areas are affected by TDP‐43.

TAR DNA‐binding protein of 43 kDa (TDP‐43) is a nuclear protein involved in transcriptional repression and alternative splicing. It has been shown to be a component of ubiquitin‐positive, tau‐ and á‐synuclein‐negative inclusions in neuronal soma and nuclei as well as neurites of frontotemporal lobar degeneration (FTLD‐U) and in motor neurons of amyotrophic lateral sclerosis (ALS). Recently, it has been co‐localized with Lewy bodies in Lewy body disease (LBD). In this study we use immunogold electron microscopy to study the ultrastructural features of TDP‐43‐labeled structures. The results showed that in all FTLD‐U, ALS and LBD, TDP‐43 was localized to filaments coated with granular material. Both coated and uncoated regions of the filaments were labeled. In FTLD‐U a subset of neurites filled with densely packed filaments were also labeled. The filaments have a diameter of 10–17 nm. Neurofilament and glial filaments were not labeled by TDP‐43. The data suggests that TDP‐43 is a constituent protein in filamentous inclusions in neurodegenerative diseases. Supported by NIH grant AG17216.

Objective: To determine whether all cases of tau‐negative frontotemporal lobar degeneration (FTLD) show abnormal TDP‐43 immunoreactivity. Design/methods: The neuropathological database at Mayo Clinic, Florida, was queried to identify all cases of tau‐negative FTLD. Forty cases were identified that underwent analysis with TDP‐43 immunohistochemistry. Cases without abnormal TDP‐43 immunoreactivity were further analyzed with H&E, PAS and silver stains, and with immunohistochemistry for phosphorylated tau, alpha‐synuclein, ubiquitin, neurofilament, alpha‐internexin and p62. Results: Of the 40 cases, 3 did not show abnormal TDP‐43 immunoreactivity. Cases 1 and 3 had clinical features of behavioral variant FTD while case 2 had features of the corticobasal syndrome. Case 3 had a positive family history. Case 1 had a pathological diagnosis of dementia lacking distinctive histology (DLDH) and case 2 basophilic inclusion body disease. Case 3 had ubiquitin‐positive but TDP‐43 negative neuronal inclusions and dystrophic neurites. Conclusion: Not all cases of tau‐negative FTLD have abnormal TDP‐43 immunoreactivity; basophilic inclusion body disease and DLDH are rare examples that do not. Given the positive family history and histological findings in case 3, there likely exists a yet to be identified gene and protein associated with tau negative FTLD.

85 year old man with progressive dementia and history of stroke clinically diagnosed with Alzheimer's disease (AD) came to autopsy. There was diffuse bilateral atrophy of the frontal/temporal lobes and focal cavitations grossly consistent with remote infarcts (confirmed microscopically) were noted in parietal lobe, basal ganglia, and cerebellar hemispheres. Microscopically, numerous tau inclusions including neurofibrillary tangles, globular inclusions, grains and threads with tau immunohistochemical stain were seen. Tau deposits were present throughout the brain with increased numbers in the limbic lobe, basal telencephalon and diencephalon. Leukoencephalopathy was due to atherosclerosis and deposition of amyloid β protein predominantly in the cortical capillaries, with diffuse cortical involvement. Only rare senile plaques were noted with amyloid immunostains. The distribution of tauopathy was unusual and similar to argyrophilic grain disease (AGD), however, the morphology of tau deposits seen are inconsistent with AGD. The presence of the cerebral amyloid angiopathy is unusual due to involvement of parenchymal capillaries outside the primary visual cortex and suggests a genetic component. A low Braak stage and scattered cortical plaques were insufficient for diagnosis of AD. To our knowledge the constellation of findings seen are unique and not previously reported.

Hypertension is highly prevalent in the elderly population and microinfarcts are the most common vascular brain pathology identified in older adults at autopsy. We investigated the associations between systolic and diastolic blood pressures measured antemortem and the presence of microinfarcts at autopsy. Study subjects (n=302; age range=71-95) were participants in the population-based Mayo Clinic Study of Aging autopsy study, who had blood pressure measurements recorded during life. We investigated both cross-sectional systolic and diastolic blood pressure measurements at the baseline visit and the change in blood pressure (slope). Presence and location (subcortical or cortical) of chronic microinfarcts was abstracted from the autopsy reports. Of the 302 study subjects, 47 (16%) had cerebral microinfarcts, and, of those, 18 (38%) had subcortical microinfarcts and 29 (62%) had only cortical microinfarcts. The baseline blood pressures were not different between subjects with no microinfarcts, subcortical microinfarcts, and only cortical microinfarcts. In a logistic regression model including time between last blood pressure measurement and death, a greater decline in systolic [OR= 1.06 (1.01, 1.11); p=0.02]) and greater decline in diastolic [OR= 1.11 (1.02, 1.20); p=0.01] blood pressures were predictors of the presence of subcortical microinfarcts at autopsy. However, these variables were not associated with the presence of cortical microinfarcts. In conclusion, microinfarcts are common in the older adult population, and most of them are located in the cortex. A greater decline in both systolic and diastolic blood pressures and their association with subcortical microinfarcts, but not with cortical microinfarcts, may have implications for aggressive lowering of blood pressure in the elderly population.

ABSTRACT Background and Purpose Neurodegeneration of the substantia nigra in Lewy body disease is associated with iron deposition, which increases the magnetic susceptibility of the substantia nigra on MRI. Our objective was to measure iron deposition in the substantia nigra in patients with probable dementia with Lewy bodies (pDLB) and patients who are at risk for pDLB by quantitative susceptibility mapping (QSM). Methods Participants included pDLB ( n = 36), mild cognitive impairment with at least one core feature of DLB (MCI‐LB; n = 15), idiopathic rapid eye movement sleep behavior disorder (iRBD; n = 11), and an age‐and gender‐matched clinically unimpaired control group ( n = 102). QSM was derived from multi‐echo 3D gradient recalled echo MRI at 3T, and groups were compared on mean susceptibility values of the substantia nigra and its relation to parkinsonism severity. Results Patients with pDLB had higher susceptibility in the substantia nigra compared to controls ( p < 0.001) and MCI‐LB ( p = 0.043). The susceptibility of substantia nigra showed an increasing trend from controls to iRBD and MCI‐LB, and to pDLB ( p < 0.001). Parkinsonism severity was not associated with the mean susceptibility in the substantia nigra in the patient groups. Conclusions Our data suggested that QSM is sensitive to the increased magnetic susceptibility due to higher iron content in the substantia nigra in pDLB. The trend of increasing susceptibility from controls to iRBD and MCI‐LB, and to pDLB suggests that iron deposition in the substantia nigra starts to increase as early as the prodromal stage in DLB and continues to increase as the disease progresses, independent of parkinsonism severity.

Abstract Aims To assess the burden of transactive response DNA‐binding protein of 43 kDa (TDP‐43) inclusions in a unique cohort of old‐age patients with genetic frontotemporal lobar degeneration (gFTLD‐TDP) and compare these patients with sporadic old‐age individuals with TDP‐43, either in the presence of Alzheimer's disease (AD‐TDP) or in isolation (pure‐TDP). Methods The brain bank at Mayo Clinic‐Jacksonville was searched for cases ≥75 years old at death with TDP‐43 extending into middle frontal cortex. Cases were split into the following groups: (1) gFTLD‐TDP ( n = 15) with progranulin ( GRN )/ C9ORF72 mutations; (2) AD‐TDP ( n = 10)—cases with median Braak neurofibrillary tangle (NFT) stage VI, Thal phase V; (3) pure‐TDP ( n = 10)—cases with median Braak NFT stage I, Thal phase I. Clinical data were abstracted; TDP‐43 burden was calculated using digital pathology. Results Amnestic Alzheimer's dementia was the clinical diagnosis in ≥50% patients in each group. The distribution of TDP‐43 burden in gFTLD‐TDP and AD‐TDP, but not pure‐TDP, was limbic‐predominant targeting CA1 and subiculum. Patients with gFTLD‐TDP had higher burden in entorhinal cortex compared to AD‐TDP. TDP‐43 burden in middle frontal cortex did not differ between the three groups. Conclusions In old age it is challenging to clinically and pathologically differentiate gFTLD‐TDP from AD‐TDP and pure‐TDP‐43 based on burden. Like AD‐TDP, old age gFTLD‐TDP have a limbic predominant TDP‐43 distribution. The finding that amnestic Alzheimer's dementia was the most common clinical diagnosis regardless of group suggests that TDP‐43 directly and indirectly targets limbic regions.