The EADC-ADNI Harmonized Protocol for manual hippocampal segmentation on magnetic resonance: evidence of validity.

BACKGROUND: An international Delphi panel has defined a harmonized protocol (HarP) for the manual segmentation of the hippocampus on MR. The aim of this study is to study the concurrent validity of the HarP toward local protocols, and its major sources of variance. METHODS: Fourteen tracers segmented 10 Alzheimer's Disease Neuroimaging Initiative (ADNI) cases scanned at 1.5 T and 3T following local protocols, qualified for segmentation based on the HarP through a standard web-platform and resegmented following the HarP. The five most accurate tracers followed the HarP to segment 15 ADNI cases acquired at three time points on both 1.5 T and 3T. RESULTS: The agreement among tracers was relatively low with the local protocols (absolute left/right ICC 0.44/0.43) and much higher with the HarP (absolute left/right ICC 0.88/0.89). On the larger set of 15 cases, the HarP agreement within (left/right ICC range: 0.94/0.95 to 0.99/0.99) and among tracers (left/right ICC range: 0.89/0.90) was very high. The volume variance due to different tracers was 0.9% of the total, comparing favorably to variance due to scanner manufacturer (1.2), atrophy rates (3.5), hemispheric asymmetry (3.7), field strength (4.4), and significantly smaller than the variance due to atrophy (33.5%, P < .001), and physiological variability (49.2%, P < .001). CONCLUSIONS: The HarP has high measurement stability compared with local segmentation protocols, and good reproducibility within and among human tracers. Hippocampi segmented with the HarP can be used as a reference for the qualification of human tracers and automated segmentation algorithms.

Comparison of human septal nuclei MRI measurements using automated segmentation and a new manual protocol based on histology.

Septal nuclei, located in basal forebrain, are strongly connected with hippocampi and important in learning and memory, but have received limited research attention in human MRI studies. While probabilistic maps for estimating septal volume on MRI are now available, they have not been independently validated against manual tracing of MRI, typically considered the gold standard for delineating brain structures. We developed a protocol for manual tracing of the human septal region on MRI based on examination of neuroanatomical specimens. We applied this tracing protocol to T1 MRI scans (n=86) from subjects with temporal epilepsy and healthy controls to measure septal volume. To assess the inter-rater reliability of the protocol, a second tracer used the same protocol on 20 scans that were randomly selected from the 72 healthy controls. In addition to measuring septal volume, maximum septal thickness between the ventricles was measured and recorded. The same scans (n=86) were also analyzed using septal probabilistic maps and DARTEL toolbox in SPM. Results show that our manual tracing algorithm is reliable, and that septal volume measurements obtained via manual and automated methods correlate significantly with each other (p<.001). Both manual and automated methods detected significantly enlarged septal nuclei in patients with temporal lobe epilepsy in accord with a proposed compensatory neuroplastic process related to the strong connections between septal nuclei and hippocampi. Septal thickness, which was simple to measure with excellent inter-rater reliability, correlated well with both manual and automated septal volume, suggesting it could serve as an easy-to-measure surrogate for septal volume in future studies. Our results call attention to the important though understudied human septal region, confirm its enlargement in temporal lobe epilepsy, and provide a reliable new manual delineation protocol that will facilitate continued study of this critical region.

The role of overexpressed DYRK1A protein in the early onset of neurofibrillary degeneration in Down syndrome.

The gene encoding the minibrain kinase/dual-specificity tyrosine phosphorylated and regulated kinase 1A (DYRK1A) is located in the Down syndrome (DS) critical region of chromosome 21. The third copy of DYRK1A is believed to contribute to abnormal brain development in patients with DS. In vitro studies showing that DYRK1A phosphorylates tau protein suggest that this kinase is also involved in tau protein phosphorylation in the human brain and contributes to neurofibrillary degeneration, and that this contribution might be enhanced in patients with DS. To explore this hypothesis, the brain tissue from 57 subjects including 16 control subjects, 21 patients with DS, and 20 patients with sporadic Alzheimer's disease (AD) was examined with two antibodies to the amino-terminus of DYRK1A (7F3 and G-19), as well as two polyclonal antibodies to its carboxy-terminus (X1079 and 324446). Western blots demonstrated higher levels of full-length DYRK1A in the brains of patients with DS when compared to control brains. Immunocytochemistry revealed that DYRK1A accumulates in neurofibrillary tangles (NFTs) in subjects with sporadic AD and in subjects with DS/AD. Overexpression of DYRK1A in patients with DS was associated with an increase in DYRK1A-positive NFTs in a gene dosage-dependent manner. Results support the hypothesis that overexpressed DYRK1A contributes to neurofibrillary degeneration in DS more significantly than in subjects with two copies of the DYRK1A gene and sporadic AD. Immunoreactivity with antibodies against DYRK1A not only in NFTs but also in granules in granulovacuolar degeneration and in corpora amylacea suggests that DYRK1A is involved in all three forms of degeneration and that overexpression of this kinase may contribute to the early onset of these pathologies in DS.

Cell type- and brain structure-specific patterns of distribution of minibrain kinase in human brain.

The minibrain kinase (Mnb/Dyrk1A) gene is localized in the Down syndrome (DS) critical region of chromosome 21. This gene encodes a proline-directed serine/threonine protein kinase (minibrain kinase-Mnb/Dyrk1A), which is required for the proliferation of distinct neuronal cell types during postembryonic neurogenesis. To study the distribution of Mnb/Dyrk1A during human brain development and aging, we raised Mnb/Dyrk1A-specific antibody (mAb 7F3) and examined 22 brains of normal subjects from 8 months to 90 years of age. We found that neurons were the only cells showing the presence of 7F3-positive product in both cell nucleus and cytoplasm. Nuclear localization supports the concept that Mnb/Dyrk1A may be involved in control of gene expression. Synaptic localization of Mnb/Dyrk1A also supports our previous studies suggesting that Mnb/Dyrk1A is a regulator of assembly of endocytic apparatus and appears to be involved in synaptic vesicle recycling and synaptic signal transmission. Accumulation of numerous 7F3-positive corpora amylacea in the memory and motor system subdivisions in subjects older than 33 years of age indicates that Mnb/Dyrk1A is colocalized with markers of astrocyte and neuron degeneration. Differences in the topography and the amount of Mnb/Dyrk1A in neurons, astrocytes, and ependymal and endothelial cells appear to reflect cell type- and brain structure-specific patterns in trafficking and utilization of Mnb/Dyrk1A.

Intraneuronal Abeta immunoreactivity is not a predictor of brain amyloidosis-beta or neurofibrillary degeneration.

Amyloid beta (Abeta) 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 Abeta immunoreactivity is an early manifestation of Alzheimer-type pathology leading to fibrillar plaque formation and/or neurofibrillary degeneration. The appearance of Abeta immunoreactivity in neurons in infants and stable neuron-type specific Abeta 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 Abeta in neurons indicated that intraneuronal Abeta was mainly a product of alpha- and gamma-secretases (Abeta(17-40/42)). The presence of N-terminally truncated Abeta(17-40) and Abeta(17-42) in the control brains was confirmed by Western blotting and the identity of Abeta(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 Abeta-immunopositive material detected in neuronal soma to amyloid deposit in plaques. The strongest intraneuronal Abeta(17-42) immunoreactivity was observed in structures with low susceptibility to fibrillar Abeta deposition, neurofibrillary degeneration, and neuronal loss compared to areas more vulnerable to Alzheimer-type pathology. These observations indicate that the intraneuronal Abeta immunoreactivity detected in this study is not a predictor of brain amyloidosis or neurofibrillary degeneration. The constant level of Abeta immunoreactivity in structures free from neuronal pathology during essentially the entire life span suggests that intraneuronal amino-terminally truncated Abeta represents a product of normal neuronal metabolism.