Stereological Quantification of Plaques and Tangles in Neocortex from Alzheimer’s Disease Patients

Abstract

We estimated by stereological methods the neocortical volume occupied by plaques and tangles in females dying with severe Alzheimer’s disease (AD), age-matched female subjects with severe vascular dementia (VaD), and normal control brains. Stereological investigations include a uniform sampling of the tissue in the whole of neocortex and its subdivisions. Resultant volume estimates provide information about the overall burden of these two pathological changes and their volume fractions and allow for correlational studies between the pathological changes and factors such as the total neocortical neuronal cell numbers, dementia test scores, and age. We estimated the volume of plaques and tangles in the entire neocortex and frontal-, temporal-, parietal-, and occipital cortex in nine female AD brains, four female patients dying with VaD, and six neurologically normal female control brains using point-counting in uniform samples of neocortex. The volume occupied by plaques comprised approximately 1% of neocortex, while the neocortical tangles made up approximately 0.1% of neocortex of AD patients but were scarcely present in the other study groups. The individual tangle and plaque volumes did not correlate to the ultimate dementia score of the AD subjects, despite correlating with reduced neocortical volume. In neocortex of AD patients, the burden of plaques and tangles is much higher than that in patients with severe vascular dementia or normal older women but only occupy a small fraction of the neocortical volume.

Keywords

Neurofibrillary tangles quantification senile plaques stereology volume fraction

INTRODUCTION

Alzheimer’s disease (AD) is an irreversible, progressive brain disorder that slowly destroys memory and other cognitive abilities, eventually leading to a condition of complete helplessness and debilitation. The incidence of AD increases with age, and it is the most common cause of dementia among older adults [1, 2]. Classical histological findings representing the main pathological features of AD brain include neurofibrillary tangles, representing intraneuronal accumulation of phosphorylated tau protein, and abundant amyloid plaques, which are characterized by extracellular aggregation of insoluble forms of amyloid-β (Aβ) peptide generated from proteolytic cleavage of the Aβ protein precursor (AβPP) [3 –7]. These lesions have a characteristic distribution, with plaques being present throughout the cortical mantle, and tangles primarily restricted to limbic and association cortices [5 , 9]. Diffuse amyloid plaques are commonly present in the brain of cognitively intact elderly people, whereas the dense-core plaques, particularly those with neuritic dystrophies, are most often found in patients with clinical AD dementia. The pathological boundaries between normal aging and AD dementia are thus not clear-cut, such that many cognitively normal elderly people have substantial cerebral amyloid burden [7 , 10–14]. Making a reliable diagnosis based entirely on clinical grounds is still challenging [15], but imaging techniques are increasingly predictive of definite histopathological findings to postmortem investigations [11 , 16]. The neuropathological staging of AD first described by Braak and Braak [5] is based on the regional development and distribution of tangles. According to their staging, tangles first accumulate in transentorhinal cortex (Braak stage I-II), and subsequently extend into limbic structures (Braak stage III-IV), before finally manifesting in isocortical structures and most of the neocortex (Braak stage V-VI) [5 , 18]. According to the National Institute of Aging (NIA-Reagan) diagnostic criteria, the hierarchical pattern of neurofibrillary degeneration across brain regions is so consistent that a staging scheme based on early lesions in the entorhinal/perirhinal cortex, followed by hippocampal Ammon subfields, association cortex, and finally primary neocortex is well-accepted [19]. There is a general consensus that the cognitive impairment in patients with AD are closely associated with progressive degeneration of the limbic system [8, 20], the basal forebrain [21] and neocortical regions [22]. This neurodegenerative process is characterized by early synaptic damage [23 –28], retrograde degeneration of axons and eventual atrophy of the dendritic tree [29 –32], and finally the perikaryon [33, 34]. It is generally accepted that synaptic loss in the neocortex and limbic system is the best correlate of the cognitive impairment in patients with AD [35, 36]. Furthermore, some studies have found lack of correlation between cognitive decline and plaques and/or tangles [37], suggesting that the progression and spread of these pathological markers is not immediately sufficient to account for the severity of dementia, at least in early stages of the disease. In contrast, other studies have reported a significant correlation between the development and duration of dementia and the density of tangles [38 –42]. The latter studies did not show a correlation between antemortem dementia scores and plaque density, although this relationship was evident in another study [43]. Finally, one study found that the burdens of frontal pE(3)-Aβ and entorhinal hyperphosphorylated tau could independently predict Mini-Mental State Examination (MMSE) scores [44].

These to some degree inconsistent results may at least in part be explained by the technical challenges in quantifying the histopathological alterations with sufficient precision [7]. Stereological cell counting studies, which are designed to avoid bias, have shown reduced total numbers of neurons in the hippocampus and in vulnerable subcortical structures such as the locus coeruleus and nucleus basalis of Meynert of AD patients [26 , 45–48]. A reduction in neocortical volume and neocortical thickness has been reported by many [49 –52], while the total neocortical neuron number estimated by stereology is not reduced in AD brains compared with controls [53, 54]. Furthermore, a stereological study found no difference in total white fiber length, fiber length density, or subcortical white matter volume in patients with AD compared with controls [55]. For an overview of stereological studies of AD brains, see [56]. In this study, we used stereological methods to estimate the neocortical volumes occupied by plaques and tangles in severely affected AD females compared to age-matched female subjects with severe vascular dementia (VaD) and normal control female brains. These volume estimations provide information about the overall burden of these two pathological hallmarks and their volume fractions and allow for simple correlational analyses of the relationships between these pathological changes and factors such as the total neocortical neuronal cell numbers, dementia test scores and age. Our stereological investigations included a uniform sampling of the total burden of plaques and tangles in the whole of neocortex and in its major sub-divisions.

METHODS

Subjects

The brains used in this study were all collected between 1971 and 1993 following the Danish autopsy laws at that time. The brains have previously been used in several stereological studies [50 , 57]. A total of 20 brains were included in the study; 10 brains from severely demented AD patients, four brains from patients with severe VaD, and six non-demented control subjects. One AD brain had to be excluded for technical reasons leaving nine AD brains for further processing.

The diagnosis, age, brain weight, hemisphere, cause of death and dementia score of the demented and control cases are shown in Table 1. The demented patients all originated from a psychogeriatric ward in Copenhagen, Denmark, and all subjects had been evaluated prospectively with a psychometric dementia test, and neurological examinations once a year during their final years. The neuropsychological tests were performed prior to the development of modern rating scales, e.g., the MMSE [58], Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) evaluation [59], the Alzheimer’s Disease Assessment Scale – cognitive section (ADAS-coq) [60], and the Global Deterioration Scale [61]. However, the tests then available evaluated a wide range of cognitive functions such as attention, orientation, long-term memory, naming, word recall, word recognition, comprehension, constructional spoken language, and ideational praxis, word finding, and delayed recall. A set of subtests from different test batteries were used, including subtests from Wais, Wechsler, Rahnsburg, Kimura, Goldstein-Scheerer, the Stroop-test, and others [50]. A progression in dementia severity had been noted in the AD patients from year to year. Performances on the different psychological subtests were scored on a 7-point scale, and a general score was obtained from 1 to 7, where group 0 represented no dementia; group 1 = mild to moderate dementia, score 1–3; group 2 = marked dementia, score 4-5; and group 3 = severe dementia, score 6-7. More details regarding the tests are given in [50]. The nine AD female subjects had a mean score of 5.6 (range 4 to 7) and were thus all severely demented, as were the four VaD subjects, with a mean score of 6.0 (range 5 to 7).

Table 1

Clinical data of subjects

	Age (years)	Brain weight (g)	Hemisphere (left/right)	Cause of death	Dementia score
Patients with AD
	79	1035	Left	Pneumonia	6
	88	1120	Right	AMI	5
	83	1180	Left	Pneumonia	7
	80	1165	Left	AMI	5
	80	1125	Right	Odema pulm.	5
	81	1245	Left	S.A.	6
	85	1175	Right	Pneumonia	6
	81	1165	Right	AMI	4
	81	1240	Left	Pneumonia	6
Mean [range]	82	1166.3	5/4		5.6
	[79 –88]	[1035–1245]			[4 –7]
Patients with VaD
	86	1180	Left	Pneumonia	6
	81	1230	Left	Stasis pulm.	7
	83	915	Right	Pneumonia	6
	81	1280	Left	Pneumonia	5
Mean [range]	82.8	1108.0	3/1		6.0
	[81 –86]	[915–1230]			[5 –7]
Controls
	72	1160	Right	AMI	–
	81	1090	Right	AMI	–
	89	1190	Right	AMI	–
	85	1180	Left	AMI	–
	64	1240	Left	AMI	–
	83	1100	Left	Odema pulm.	–
Mean [range]	80.4 [64 –89]	1173.3 [1100–1240]	3/3
p-value	0.64^¤	0.63^*	0.85^*

AD, Alzheimer’s disease patients; VaD: Vascular dementia patients; AMI, acute myocardial infarction; Pulm., pulmonary; S.A., severe arteriosclerosis; ^*Student’s t-test; ^¤Mann-Whitney U-test; ^#Chi-squared test; p < 0.05 is considered significant. The p-value for dementia scores are for AD patients compared to patients presented with combined group of vascular dementia patients with control subjects.

The control patients had died from non-neurologic illnesses at general surgical and medical wards in Copenhagen. The patients were sent for routine necropsy, where their clinical records were reviewed. The files collected from the control donors were carefully examined to exclude individuals with diseases that might affect the central nervous system, such as neurodegenerative diseases, cerebrovascular diseases, metastatic cancer, diabetes, hypertension, or drug and alcohol abuse. Although the controls had not undergone cognitive assessment, all had been living a normal, independent life prior to their death, and were regarded as cognitively normal by their local general practitioner. Based upon our histological examination, none of the control subjects could be characterized AD-positive according to Braak staging [5], but all exhibited some neuropathological alterations commonly seen in normal old age [62, 63]. The pathological diagnosis was based upon analysis of tissue samples from frontal-, parietal-, medial temporal and occipital lobes, insula, gyrus cinguli, and hippocampus, all of which had been processed routinely by embedding in paraffin. The tissue was sectioned and processed for immunohistochemistry with antibodies against alpha-synuclein (sc-12767, 1:10,000, Santa Cruz Biotechnology, USA), and ubiquitin (Z0458, 1:5,000, DAKO, Denmark). Furthermore, we made hematoxylin and eosin staining, and sections were processed for a Klüver-Barrera staining as well as immunohistochemistry with antibodies against Aβ (M0872, 1:1000, DAKO, Denmark) and tau (A0024, 1:50,000, DAKO, Denmark). The pathoanatomical diagnosis of AD was based on previously described presence of numerous widespread plaques and tangles in the neocortex identified after application of the Bielschowsky silver stain [50 , 53–57].

Ethics approval

The Ethics committee for the Copenhagen Regional Area approved this study (H-16030779).

Preparation

As described elsewhere [64 –66] and schematized in Fig. 1, all brains had been stored in 0.1 M sodium phosphate buffered formaldehyde fixative (pH 7.2, 4% formaldehyde) for at least five years. At time for processing, the meninges were removed, and the cerebellum and brainstem detached at the level of the third cranial nerve. Right or left hemispheres were chosen systematically randomly. The frontal-, temporal-, parietal-, and occipital regions were delineated and painted on the pial surface in different colors (C.D.I.’s Tissue Marking Dye System, Cancer Diagnostics, Inc., Troy, MI, USA) as described elsewhere [67]. Each sampled hemisphere was embedded in 6% agar and sliced coronally at 7 mm intervals. The neocortical volume was estimated by Cavalieri’s principle [68, 69]. A total of 250 points was counted on neocortical sectional areas, providing a coefficient of error, CE = SEM/mean = 2 to 3%, for the total volume. From every second slice, starting randomly between slice one or two, transcortical wedges were sampled uniformly and systematically from each neocortical region [67]. Each wedge was cut into 2-mm-wide parallel bars providing from 25–50 bars per region. These were subsampled uniformly so that each region was represented by 8 to 10 bars, which were embedded in LKB-Historesin^®. Subsequently, a 10-μm-thick section was cut from each of the 30 to 40 bars and stained free-floating with HAGA silver method [70] and a modified Giemsa stain. This silver staining allowed for reliable detection of amyloid-related components of plaques and tangles, while the Giemsa stain provided a background stain (Fig. 2).

Fig. 1

Stereological workflow for estimation of neuron numbers (using optical disectors) or plaques and tangle volumes using point-counting. (1) The four subdivisions of neocortex are indicated in different colors. (2) The sampled hemisphere is embedded in agar and cut into slabs of equal thickness. (3) Point-counting grid is used to estimate the volume of the neocortex (V_ref) using the Cavaliere’s formula for volume where a(p) = area per point on the point-counting grid and ΣP= the total number of points falling on neocortex; t = the average slab thickness. (4) A numbered metal plate and plastic holders are used for subsampling of cortical biopsies using a stereological sampling procedure. (5 Left) Neuron counting. The sampled tissue biopsies are embedded in LKB-Historesin^® and 40μM thick sections are cut from the block and stained with a modified Giemsa stain. (5 Right) Estimation of plaques and tangles using a point-counting grid. Three set of points was used: one set of points hitting cortical tissue, one set of points hitting plaques and one set of points hitting tangles, with the relationship 1:25:100. Knowing this relationship, the volume fraction between the different compartments can be calculated. The final on-screen magnification for counting points was 2340X. The volume fraction of tangles and plaques was multiplied by the reference space, V_ref, to get the total volume of plaques and tangles. This volume is multiplied by 2 to give bilateral volumes. (6) The sections are counted in an Olympus BX50 microscope equipped with a camera and connected to a computer where the neurons are counted using unbiased counting frames at a final magnification of approx. 3000x. The neuronal density N_V = ∑Q^-/∑vol (disectors) where ∑Q^-= the total numbers of neurons counted and ∑vol (disectors)= the total volume of all sampled disectors. The neuron number N =N_V · V_ref. Volume of tangles (example) $V_{t (c m^{3})} = (\frac{\sum^{} P_{h i t t i n g t a n g l e s \cdot 100}}{\sum^{} P_{h i t t i n g t i s s u e}}) \cdot V_{r e f}$

Fig. 2

The different pathological changes identified by their morphology as plaques and tangles. Arrow = tangles. Circled clear arrow = plaque. Scale = 30μm. For plaques and tangles we used a HAGA staining.

Stereological design

The stereological design consisted of an Olympus BX50 microscope. A camera transmitted the images to a computer screen where a point-counting grid was superimposed using the computer-assisted stereological software (Newcast, Visiopharm, Hørsholm, DK). A motorized automatic stage controlled movements in the x, y-plane via a connected joystick. All 8–10 uniformly sampled bars per region were included in the final estimation. Three set of points was used: one set of points hitting cortical tissue, one set of points hitting plaques and one set of points hitting the whole cell body in cells with tangles, with the relationship 1:25:100. Knowing this relationship, the volume fraction between the different compartments can be calculated. The average number of points hitting cortex was 2372 (range 825–2984) per brain. The average number of points hitting plaques was 545 (range 214–1150) in the AD group and 89 (range 1–145) in the two control groups. Finally, the average number of points hitting tangles was 123 (range 66–291) in the AD group and 28 (range 0–104) in the control group. In some brains, there were no plaques and tangles, or too few for a meaningful point-counting. These brains were recorded as 0 points and all turned out to be control brains. All counting was performed on coded glasses. The final on-screen magnification for counting points was 2340X. Knowing the total neocortical volume for each hemisphere and the volume of each sub-division, the volume fraction of tangles and plaques was multiplied by the reference space (e.g., neocortical volume); for example V_tangles (cm³) = (ΣP hitting tangles×100/ΣP hitting tissue)×V_neocortex (cm³) to get the total volume of neocortical tangles. This volume is multiplied by 2 to give bilateral volumes.

Statistical tests

In all cases, no statistical differences were observed between the VaD and non-demented control (NC) group in plaque and tangle volumes in each region and in the total neocortex, hence all succeeding analyses referring to the control brains are defined as NC and VaD combine. For statistical analyses in tangle and plaque volumes, we performed the Mann-Whitney U test, the Student’s t-test, or the Welch’s t-test between AD patients (n = 9) and controls (n = 10). Statistical tests applied for each analysis are described in table legends in results.

For neuron numbers and neocortical volume estimates we used Kruskall-Wallis one-way ANOVA and Dunn’s post hoc test between all three groups. All data were analyzed using GraphPad Prism 6.01 software. The coefficient of variance (CV), calculated as CV = SD/mean, is shown in parenthesis throughout the text and listed in table 2. The precision of the stereological quantification is expressed as the coefficient of error (CE), which is calculated as previously described [69]. The CE values are listed in table 2. Normal distribution was tested using the D’Agostino & Pearson normality test and Shapiro-Wilk normality test. If the normality test failed, the data was log-transformed and, if still not showing normal distribution, we applied the non-parametric Mann-Whitney U test. For correlation analysis, we used the Pearson’s correlation or Spearman’s rho. p-values below 0.05 were considered statistically significant.

RESULTS

Plaques

In the AD brains, the total mean volume of neocortical plaques was 3.28 cm³ (1.39) compared to the 0.92 cm³ (0.97) in aged control brains, which was a highly statistically significant difference (p = 0.004). In all areas, the plaque volume was significantly higher in AD brains compared to control brains (Fig. 3A-E). The plaque volume increase was found to be highest in the frontal cortex, AD = 1.28 cm³ (0.97), compared to 0.21 cm³ (1.13) in controls, p = 0.01; followed by the parietal cortex, 0.77 cm³ (0.50) in AD compared to 0.31 cm³ (1.41) in controls, p = 0.008; the temporal cortex, 0.65 cm³(0.75) in AD compared to 0.22 cm³ (1.22) in controls, p = 0.025; and lastly the occipital cortex, AD = 0.41 cm³ (1.00) compared to 0.11 cm³ (2.10) in controls, p = 0.01.

Fig. 3

Total volume fractions in percent (%) of the cortical area volume of plaques (A-E), tangles (F-J) and total burdens (K-O) in Alzheimer’s disease (AD) patients, n = 9, vascular demented (VaD) subjects, n = 4 and normal control subjects (NC), n = 6. The p-values represent the difference between the AD group (n = 9) and the combined group of VaD subjects and NCs (n = 10). The groups were analyzed using the parametric Students t-test or Welch’s t-test when normally distributed, if not the non-parametric Mann-Whitney U test was applied. p-values below 0.05 were considered significant.

Table 2

Estimates of volumes of tangles, plaques and total burden in controls and AD

	Controls (n = 10)			AD (n = 9)
Brain region	Mean volumes (cm³)	CV	CE	Mean volumes (cm³)	CV	CE	p-value	Changes in percent %
Frontal cortex
Tangles	0.035	0.99	0.31	0.11	1.18	0.39	0.090^*	↑	325
Plaques	0.21	1.13	0.36	1.28	0.97	0.32	0.01 ^#	↑	617
Total Burden	0.24	1.02	0.32	1.40	0.96	0.32	0.008^*	↑	575
Parietal cortex
Tangles	0.011	1.29	0.41	0.047	0.80	0.27	0.02^*	↑	427
Plaques	0.31	1.41	0.45	0.77	0.50	0.17	0.008 ^#	↑	249
Total Burden	0.32	1.39	0.44	0.82	0.48	0.16	< 0.001^*	↑	255
Temporal cortex
Tangles	0.015	1.97	0.62	0.078	1.00	0.33	0.003 ^#	↑	520
Plaques	0.25	1.22	0.38	0.65	0.75	0.25	0.03 ^¤	↑	301
Total Burden	0.23	1.16	0.37	0.73	0.71	0.24	0.02 ^¤	↑	316
Occipital cortex
Tangles	0.002	2.36	0.75	0.017	0.74	0.25	0.003 ^#	↑	850
Plaques	0.11	2.10	0.66	0.41	1.00	0.33	0.02 ^#	↑	376
Total Burden	0.11	2.06	0.65	0.43	0.96	0.32	0.01 ^#	↑	384
Total neocortex
Tangles	0.058	1.06	0.34	0.25	0.85	0.28	0.03^*	↑	422
Plaques	0.92	1.32	0.42	3.28	0.61	0.20	0.004 ^#	↑	356
Total Burden	0.98	1.27	0.40	3.53	0.59	0.20	0.004 ^#	↑	359

Controls, Combined control group including VaD and NC; AD, Alzheimer’s disease; CV, coefficient of variation = SD/mean. ^*Welch’s t-test. ^#Mann-Whitney U test. ^¤Student’s t-test. p-values below 0.05 were considered significant.

Tangles

The total neocortical mean volume of tangles in AD brains was 0.25 cm³ (0.85), which was statistically significant different compared to the 0.058 cm³ (1.06) in the control brains, p = 0.029. In all regions but the frontal lobe, AD = 0.11 cm³ (1.18); 0.035 cm³ (0.99) in controls, p = 0.09, the AD brains showed a significant higher burden of tangles compared to control subjects (Fig. 3F-J). The occipital cortex showed the highest difference, AD = 0.017 cm³ (0.74), compared to 0.002 cm³ (2.36) in controls, p = 0.003; followed by the temporal lobe, 0.078 cm³ (1.00) in AD compared to 0.015 cm³ (1.97) in controls, p = 0.003; and the parietal lobe, 0.047 cm³ (0.80) in AD compared to 0.011 cm³ (1.29) in controls, p = 0.024. The volumes for tangles, plaques, and the total burden are summarized in Table 2.

The total burden of plaques and tangles combined was increased in all regions of the AD brains compared to the control brains (Fig. 3K-O).

No significant difference was found in pathological changes between the two hemispheres.

Using a Kruskall-Wallis One-way ANOVA prior to a Dunn’s post hoc test, tangles were significantly different between regions in AD patients (F(4,36) = 8.054; p = 0.045). However, the post hoc test could not identify which region/regions that was significantly different. No difference was observed in the distribution of plaques in AD and NC or in the distribution of tangles in NC.

In AD, the volume of tangles was significantly correlated to the total neocortical volume (95% CI –0.78 to –0.04; r = –0.49; p = 0.032) and to the tangle burden in the parietal (95% CI –0.76 to 0.006; r = –0.46; p = 0.047), the temporal (95% CI –0.84 to –0.20; r = –0.61; p = 0.006), and occipital cortex (95% CI –0.77 to –0.051; r = –0.51; p = 0.028). Results are summarized in Table 3. The plaque burden proved to have a significant correlation only with the total neocortical volume (95% CI –0.77 to –0.04; r = –0.48; p = 0.036). The combined volume of plaques and tangles also correlated with the total neocortex volume (95% CI –0.77 to –0.05; r = –0.49; p = 0.03) (Table 3).

Table 3

Estimated volume correlations between tangles and plaques, and the cortical volume

	Correlation	p-value
Frontal cortex
Tangles	–0.08	0.74
Plaques	–0.32	0.18
Total Burden	–0.33	0.17
Parietal cortex
Tangles	–0.46	0.047^*
Plaques	–0.41^#	0.08
Total Burden	–0.43^#	0.07
Temporal cortex
Tangles	–0.61	0.006^*
Plaques	–0.26^#	0.29
Total Burden	–0.31^#	0.20
Occipital cortex
Tangles	–0.51	0.03^*
Plaques	–0.41	0.09
Total Burden	–0.42	0.08
Total Neocortex
Tangles	–0.49	0.03 ^*
Plaques	–0.48^#	0.046 ^*
Total Burden	–0.49^#	0.03 ^*

Values are represented as correlations defined by either Spearmans rho or Pearsons r-value^#. *: identifies significant outcomes. p-values < 0.05 were considered significant.

The mean total neocortical volume was significantly different (p = 0.026) between groups, and AD brains (343 cm³ (0.12)) were significantly decreased from those in the NC group (424 cm³ (0.10); p < 0.05; Fig. 4A). Further, the occipital cortical volume was similar significantly different (p = 0.026) and decreased in AD brains (34 cm³ (0.17)) compared to that of the NC group (45 cm³ (0.16); p < 0.05). No other differences were observed. The volumes of the nine AD brains compared to the four VaD subjects and six NC subjects are shown in Fig. 4A-E.

Fig. 4

The total volume of the total neocortex (A), the frontal cortex (B), the parietal cortex (C), the temporal cortex (D) and the occipital cortex (E) in Alzheimer’s disease (AD) patients, n = 9, vascular demented (VaD) subjects, n = 4 and normal control (NC) subjects, n = 6. Group differences were analyzed using Kruskall-Wallis one-way ANOVA and Dunn’s post hoc test. p-values below 0.05 were considered significant.

Estimates of total neocortical neuron numbers are shown in Fig. 5. The total number of neocortical neurons was 17.1×10⁹ (0.12) in the AD brains, 17.2×10⁹ (0.19) in VaD subjects and 17.1×10⁹ (0.15) in NC subjects, which was not statistically significant different between groups (p = 0.90).

Fig. 5

Total numbers of neurons in the neocortex in Alzheimer’s disease (AD) patients, n = 9, vascular demented (VaD) subjects, n = 4 and normal control (NC) subjects, n = 6. Horizontal bars represent the group mean. Group differences were analyzed using Kruskall-Wallis one-way ANOVA and Dunn’s post hoc test. p-values below 0.05 were considered significant.

Only a few significant correlations were found between the clinical data and the pathological burdens in the AD group. The dementia scores correlated significantly only with the total pathological burden in the temporal cortex (95% CI 0.006 to 0.76; r = 0.46; p = 0.048) and the increased tangle burden in the occipital cortex (95% CI 0.097 to 0.79; r = 0.53; p = 0.02). No correlations were found between plaques and tangles and neuron numbers or plaques and tangles, and age.

DISCUSSION

This study reports a significant difference between AD patients and control subjects in the volume and volume fractions of plaques and tangles in neocortex. The volume of the plaques was 10-20-fold higher than the volume of the tangles, and the plaque volume in all cortical areas was significantly higher in AD brains compared to control brains. Further, in all regions of the neocortex apart from the frontal lobe, the AD brains showed a significantly higher burden of tangles compared to controls. No difference was observed in the volume of tangles and plaques between the VaD and the control brains in all regions, indicating that VaD patients, as expected, had few plaques or tangles and could for the present purposes properly be pooled with the control subjects. Also, the combined burden of plaques together with tangles was increased in all regions of AD brains compared to the control brains. Since the plaque burden is an order of magnitude higher than the tangle burden, it is the dominating factor in terms of volume of the overall plaque and tangle pathology in AD brains. However, it could be noted that when sufficiently developed, tangles impair severely the function of the whole neuron affected, which represents indeed a larger fraction of tissue volume. Notably, the total pathological volume of all plaques was only ∼1% in AD, while the volumetric burden of tangles was much lower 0.1%. So, although plaques and tangles are characteristic features in the AD neocortex, they only represent a very small fraction of the neocortical volume. This is in contrast to transgenic mouse models in which the plaque burden may be as high as 10–20% of neocortex volume, indicating that present animal studies fail to match human AD pathology with respect to the abundance of cortical amyloid deposits [71 –75].

The dementia scores of our patient group correlated to the volumes of plaques and tangles in the temporal cortex. This is perhaps unsurprising, since it has repeatedly been found that the temporal lobe is severely affected in AD patients [76 –79], whereas the occipital lobe is relatively spared. However, neuropathological hallmarks in AD have previously been found in the occipital cortex, such as increased tangle burden and decreased mitochondrial enzyme activity [80, 81]. Furthermore, occipital involvement is clinically correlated to constructional apraxia in AD [82, 83]. Unfortunately, the medical charts of our patients do not contain information for testing for specific clinical correlates of the pathology distributions.

The total number of neocortical neurons was almost the same in all three groups, as we have previously seen with earlier stereological analysis of this same material [50 , 69]. Thus, we assert that dementia does not arise from a loss of neocortical neurons, but rather from declining synaptic density and axonal projections.

Our results must be considered in the light of several limitations. Due to a low number of plaques and tangles in some control brains, a small number of points were sampled in neocortex in these subjects, resulting in high CE values (table 2). However, the biological variance was also high, with CVs between 0.48 and 2.36 (table 2). Thus, despite some estimations being of low precisions, the mean volumes presented here for plaques and tangles are still a good indicator of the true volumes. Since this preliminary study is limited by its small sample size, negative outcomes must be interpreted with special caution as potential false negative findings. Future studies including larger sample sizes are needed to confirm our results. Further, the study includes only female subjects. Although there is no indication that the burden of plaques and tangles differs in human male and female AD patients [41], studies of transgenic AD animal models have shown sex differences [84, 85], suggesting that gender differences may yet emerge in human studies. Finally, differential shrinkage caused by fixation and histologically processing in different diseases can never be ruled our but generally we have not encountered statistically significant changes in volumetric parameters caused by histological handling between brains of control subjects, AD brains or brains from patients with vascular dementia using Historesin embedding media. However, volumetric data on biological tissue should always be evaluated with this possible limitation. Strengths of the study includes stereological quantification of the volume of plaques and tangles in brain of AD subjects using strict sampling strategies, thus providing volume estimates that are based on unbiased principles. This gives us great confidence to the estimates for total volumes or plaque and tangle volumes for entire lobes or for entire neocortex. To the best of our knowledge, this is the first stereological study to estimate these parameters with a precision pre-selected by the investigators and using uniform sampling over the different sub-regions of cerebral cortex. The present results can inform future studies of quantitative pathological measures in other neurodegenerative brain diseases.

Conclusions

In neocortex of patients dying with AD patients, the burden of plaques and tangles is much higher than that in control subjects and in VaD cases. The volume fraction of neocortical plaques was approximately 1%, which falls below findings in transgenic mice overexpressing Aβ, and the volume occupied by tangles was only some 0.1%. The burden of plaques and tangles correlated with decreased neocortical volume in AD patients, with the strongest association observed in the temporal cortex, followed by the occipital and parietal lobes. Furthermore, the dementia score correlated significantly with the composite burden of plaques and tangles in temporal cortex and with the tangle burden in the occipital lobe.

Footnotes

ACKNOWLEDGMENTS

The authors are grateful to Susanne Sørensen for technical assistance and acknowledge manuscript revisions by Inglewood Biomedical Editing. The work was supported by the Dagmar Marshalls Foundation, Familien Hede Nielsens Foundation and The Alzheimer-Research Foundation.

Authors’ disclosures available online ().

References

Stelzmann

, Schnitzlein

, Murtagh

(1995) An English translation of Alzheimer’s 1907 paper, “über eine eigenartige erkankung der hirnrinde.” Clin Anat 8, 429-431.

Maurer

, Volk

, Gerbaldo

(1997) Auguste D and Alzheimer’s disease. Lancet 349, 1546-1549.

Anand

, Gill

, Mahdi

(2014) Therapeutics of Alzheimer’s disease: Past, present and future. Neuropharmacology 76 Pt A, 27-50.

Ball

(1977) Neuronal loss, neurofibrillary tangles and granulovacuolar degeneration in the hippocampus with ageing and dementia. A quantitative study. Acta Neuropathol 37, 111-118.

Braak

, Braak

(1991) Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82, 239-259.

Kumar

, Singh

, Ekavali (2015) A review on Alzheimer’s disease pathophysiology and its management: An update. Pharmacol Rep 67, 195-203.

Serrano-Pozo

, Frosch

, Masliah

, Hyman

(2011) Neuropathological alterations in Alzheimer disease. Cold Spring Harb Perspect Med 1, 1-23.

Arnold

, Hyman

, Flory

, Damasio

, Van Hoesen

(1991) The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer’s disease. Cereb Cortex 1, 103-116.

Thal

, Rub

, Orantes

, Braak

(2002) Phases of Aβ-deposition in the human brain and its relevance for the development of AD. Neurology 58, 1791-1800.

10.

Knopman

, Parisi

, Salviati

, Floriach-Robert

, Boeve

, Ivnik

, Smith

, Dickson

, Johnson

, Petersen

, McDonald

, Braak

, Petersen

(2003) Neuropathology of cognitively normal elderly. J Neuropathol Exp Neurol 62, 1087-1095.

11.

Tomlinson

, Blessed

, Roth

(1968) Observations on the brains of non-demented old people. J Neurol Sci 7, 331-356.

12.

Tomlinson

, Blessed

, Roth

(1970) Observations on the brains of demented old people. J Neurol Sci 11, 205-242.

13.

Arriagada

, Growdon

, Hedley-Whyte

, Hyman

(1992) Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology 42, 631-639.

14.

Green

, Kaye

, Ball

(2000) The Oregon brain aging study: Neuropathology accompanying healthy aging in the oldest old. Neurology 54, 105-113.

15.

Masters

, Bateman

, Blennow

, Rowe

, Sperling

, Cummings

(2015) Alzheimer’s disease. Nat Rev Dis Prim 1, 15056.

16.

Alafuzoff

, Arzberger

, Al-Sarraj

, Bodi

, Bogdanovic

, Braak

, Bugiani

, Del-Tredici

, Ferrer

, Gelpi

, Giaccone

, Graeber

, Ince

, Kamphorst

, King

, Korkolopoulou

, Kovács

, Larionov

, Meyronet

, Monoranu

, Parchi

, Patsouris

, Roggendorf

, Seilhean

, Tagliavini

, Stadelmann

, Streichenberger

, Thal

, Wharton

, Kretzschmar

(2008) Staging of neurofibrillary pathology in Alzheimer’s disease: A study of the BrainNet Europe consortium. Brain Pathol 18, 484-496.

17.

Braak

, Braak

, Mandelkow

(1994) A sequence of cytoskeleton changes related to the formation of neurofibrillary tangles and neuropil threads. Acta Neuropathol 87, 554-567.

18.

Braak

, Braak

(1995) Staging of Alzheimer’s disease-related neurofibrillary changes. Neurobiol Aging 16, 271-284.

19.

Consensus

recommendations for the postmortem diagnosis of Alzheimer’s disease

(1997) The national institute on aging, and reagan institute working group on diagnostic criteria for the neuropathological assessment of Alzheimer’s disease. Neurobiol Aging 18, S1-S2.

20.

Klucken

, McLean

, Gomez-Tortosa

, Ingelsson

, Hyman

(2003) Neuritic alterations and neural system dysfunction in Alzheimer’s disease and dementia with Lewy bodies. Neurochem Res 28, 1683-1691.

21.

Teipel

, Flatz

, Heinsen

, Bokde

ALW

, Schoenberg

, Stockel

, Dietrich

, Reiser

, Moller

, Hampel

(2005) Measurement of basal forebrain atrophy in Alzheimer’s disease using MRI. Brain 128, 2626-2644.

22.

Terry

, Peck

, DeTeresa

, Schechter

, Horoupian

(1981) Some morphometric aspects of the brain in senile dementia of the Alzheimer type. Ann Neurol 10, 184-192.

23.

Masliah

, Terry

(1993) The role of synaptic proteins in the pathogenesis of disorders of the central nervous system. Brain Pathol 3, 77-85.

24.

Masliah

(2000) The role of synaptic proteins in Alzheimer’s disease. Ann N Y Acad Sci 924, 68-75.

25.

Crews

, Masliah

(2010) Molecular mechanisms of neurodegeneration in Alzheimer’s disease. Hum Mol Genet 19, R12-R20.

26.

Vogels

, Broere

, ter

Laak HJ

, ten

Donkelaar HJ

, Nieuwenhuys

, Schulte

(1990) Cell loss and shrinkage in the nucleus basalis Meynert complex in Alzheimer’s disease. Neurobiol Aging 11, 3-13.

27.

Simic

, Kostovic

, Winblad

, Bogdanovic

(1997) Volume and number of neurons of the human hippocampal formation in normal aging and Alzheimer’s disease. J Comp Neurol 379, 482-494.

28.

Masliah

(1995) Mechanisms of synaptic dysfunction in Alzheimer’s disease. Histol Histopathol 10, 509-519.

29.

Coleman

, Perry

(2002) Axon pathology in neurological disease: A neglected therapeutic target. Trends Neurosci 25, 532-537.

30.

Higuchi

, Lee

VMY

, Trojanowski

(2002) Tau and axonopathy in neurodegenerative disorders. Neuromolecular Med 2, 131-150.

31.

Grutzendler

, Helmin

, Tsai

, Gan

(2007) Various dendritic abnormalities are associated with fibrillar amyloid deposits in Alzheimer’s disease. Ann N Y Acad Sci 1097, 30-39.

32.

Perlson

, Maday

, Fu

, Moughamian

, Holzbaur

ELF

(2010) Retrograde axonal transport: Pathways to cell death? Trends Neurosci 33, 335-344.

33.

Hyman

, Van Hoesen

, Kromer

, Damasio

(1986) Perforant pathway changes and the memory impairment of Alzheimer’s disease. Ann Neurol 20, 472-481.

34.

Lippa

, Hamos

, Pulaski-Salo

, DeGennaro

, Drachman

(1992) Alzheimer’s disease and aging: Effects on perforant pathway perikarya and synapses. Neurobiol Aging 13, 405-411.

35.

DeKosky

, Scheff

(1990) Synapse loss in frontal cortex biopsies in Alzheimer’s disease: Correlation with cognitive severity. Ann Neurol 27, 457-464.

36.

DeKosky

, Scheff

, Styren

(1996) Structural correlates of cognition in dementia: Quantification and assessment of synapse change. Neurodegeneration 5, 417-421.

37.

Iacono

, Resnick

, O’Brien

, Zonderman

, An

, Pletnikova

, Rudow

, Crain

, Troncoso

(2014) Mild cognitive impairment and asymptomatic Alzheimer disease subjects: Equivalent β-amyloid and tau loads with divergent cognitive outcomes. J Neuropathol Exp Neurol 73, 295-304.

38.

Arriagada

, Marzloff

, Hyman

(1992) Distribution of Alzheimer-type pathologic changes in nondemented elderly individuals matches the pattern in Alzheimer’s disease. Neurology 42, 1681-1688.

39.

Bierer

, Hof

, Purohit

, Carlin

, Schmeidler

, Davis

, Perl

(1995) Neocortical neurofibrillary tangles correlate with dementia severity in Alzheimer’s disease. Arch Neurol 52, 81-88.

40.

Giannakopoulos

, Herrmann

, Bussiere

, Bouras

, Kovari

, Perl

, Morrison

, Gold

, Hof

(2003) Tangle and neuron numbers, but not amyloid load, predict cognitive status in Alzheimer’s disease. Neurology 60, 1495-1500.

41.

Murray

, Lowe

, Graff-Radford

, Liesinger

, Cannon

, Przybelski

, Rawal

, Parisi

, Petersen

, Kantarci

, Ross

, Duara

, Knopman

, Jack

, Dickson

(2015) Clinicopathologic and ¹¹C-Pittsburgh compound B implications of Thal amyloid phase across the Alzheimer’s disease spectrum. Brain 138, 1370-1381.

42.

Nagy

, Esiri

, Jobst

, Morris

, King

, McDonald

, Litchfield

, Smith

, Barnetson

, Smith

(1995) Relative roles of plaques and tangles in the dementia of Alzheimer’s disease: Correlations using three sets of neuropathological criteria. Dementia 6, 21-31.

43.

Roth

, Tomlinson

, Blessed

(1966) Correlation between scores for dementia and counts of “senile plaques” in cerebral grey matter of elderly subjects. Nature 209, 109-110.

44.

Mandler

, Walker

, Santic

, Hanson

, Upadhaya

, Colloby

, Morris

, Thal

, Thomas

, Schneeberger

, Attems

(2014) Pyroglutamylated amyloid-β is associated with hyperphosphorylated tau and severity of Alzheimer’s disease. Acta Neuropathol 128, 67-79.

45.

Busch

, Bohl

, Ohm

(1997) Spatial, temporal and numeric analysis of Alzheimer changes in the nucleus coeruleus. Neurobiol Aging 18, 401-406.

46.

Mufson

, Ma

, Cochran

, Bennett

, Beckett

, Jaffar

, Saragovi

, Kordower

(2000) Loss of nucleus basalis neurons containing trkA immunoreactivity in individuals with mild cognitive impairment and early Alzheimer’s disease. J Comp Neurol 427, 19-30.

47.

West

, Coleman

, Flood

, Troncoso

(1994) Differences in the pattern of hippocampal neuronal loss in normal ageing and Alzheimer’s disease. Lancet 344, 769-772.

48.

West

, Kawas

, Stewart

, Rudow

, Troncoso

(2004) Hippocampal neurons in pre-clinical Alzheimer’s disease. Neurobiol Aging 25, 1205-1212.

49.

Dickerson

, Bakkour

, Salat

, Feczko

, Pacheco

, Greve

, Grodstein

, Wright

, Blacker

, Rosas

, Sperling

, Atri

, Growdon

, Hyman

, Morris

, Fischl

, Buckner

(2009) The cortical signature of Alzheimer’s disease: Regionally specific cortical thinning relates to symptom severity in very mild to mild AD dementia and is detectable in asymptomatic amyloid-positive individuals. Cereb Cortex 19, 497-510.

50.

Regeur

(2000) Increasing loss of brain tissue with increasing dementia: A stereological study of post-mortem brains from elderly females. Eur J Neurol 7, 47-54.

51.

Sabuncu

, Desikan

(2011) The dynamics of cortical and hippocampal atrophy in Alzheimer disease. Arch Neurol 68, 1040-1048.

52.

Thompson

, Mega

, Woods

, Zoumalan

, Lindshield

, Blanton

, Moussai

, Holmes

, Cummings

, Toga

(2001) Cortical change in Alzheimer’s disease detected with a disease-specific population-based brain atlas. Cereb Cortex 11, 1-16.

53.

Pelvig

, Pakkenberg

, Regeur

, Oster

, Pakkenberg

(2003) Neocortical glial cell numbers in Alzheimer’s disease. A stereological study. Dement Geriatr Cogn Disord 16, 212-219.

54.

Regeur

, Jensen

, Pakkenberg

, Evans

, Pakkenberg

(1994) No global neocortical nerve cell loss in brains from patients with senile dementia of Alzheimer’s type. Neurobiol Aging 15, 347-352.

55.

Jørgensen

AMB

, Marner

, Pakkenberg

(2008) No change in total length of white matter fibers in Alzheimer’s disease. Neuroscience 157, 878-883.

56.

Stark

, Pelvig

, Jørgensen

, Andersen

(2005) Measuring morphological and cellular changes in Alzheimer’s dementia: A review emphasizing stereology. Curr Alzheimer Res 2, 449-481.

57.

Andersen

, Andersen

, Pakkenberg

(2012) Stereological quantification of the cerebellum in patients with Alzheimer’s disease. Neurobiol Aging 33, 197.e11-20.

58.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189-198.

59.

Morris

, Heyman

, Mohs

, Hughes

, van Belle

, Fillenbaum

, Mellits

, Clark

(1989) The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology 39, 1159-1165.

60.

Mohs

, Cohen

(1988) Alzheimer’s Disease Assessment Scale (ADAS). Psychopharmacol Bull 24, 627-628.

61.

Reisberg

, Ferris

, de Leon

, Crook

(1982) The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry 139, 1136-1139.

62.

Berg

, McKeel

DWJ

, Miller

, Storandt

, Rubin

, Morris

, Baty

, Coats

, Norton

, Goate

, Price

, Gearing

, Mirra

, Saunders

(1998) Clinicopathologic studies in cognitively healthy aging and Alzheimer’s disease: Relation of histologic markers to dementia severity, age, sex, and apolipoprotein E genotype. Arch Neurol 55, 326-335.

63.

Price

, Davis

, Morris

, White

(1991) The distribution of tangles, plaques and related immunohistochemical markers in healthy aging and Alzheimer’s disease. Neurobiol Aging 12, 295-312.

64.

Karlsen

, Pakkenberg

(2011) Total numbers of neurons and glial cells in cortex and basal ganglia of aged brains with down syndrome-a stereological study. Cereb Cortex 21, 2519-2524.

65.

Karlsen

, Korbo

, Uylings

HBM

, Pakkenberg

(2014) A stereological study of the mediodorsal thalamic nucleus in Down syndrome. Neuroscience 279, 253-259.

66.

Salvesen

, Ullerup

, Sunay

, Brudek

, Løkkegaard

, Agander

, Winge

, Pakkenberg

(2015) Changes in total cell numbers of the basal ganglia in patients with multiple system atrophy – A stereological study. Neurobiol Dis 74, 104-113.

67.

Pakkenberg

, Gundersen

(1997) Neocortical neuron number in humans: Effect of sex and age. J Comp Neurol 384, 312-320.

68.

Gundersen

, Jensen

(1987) The efficiency of systematic sampling in stereology and its prediction. J Microsc 147, 229-263.

69.

Regeur

, Pakkenberg

(1989) Optimizing sampling designs for volume measurements of components of human brain using a stereological method. J Microsc 155, 113-121.

70.

Haga

, Ikeda

, Iwabuchi

, Akiyama

, Kondoh

, Kosaka

(1994) Methenamine-silver staining: A simple and sensitive staining method for senile plaques and neurofibrillary tangles. Biotech Histochem 69, 295-300.

71.

Manaye

, Wang

, O’Neil

, Huang

, Xu

, Lei

, Tizabi

, Ottinger

, Ingram

, Mouton

(2007) Neuropathological quantification of dtg APP/PS1: Neuroimaging, stereology, and biochemistry. Age 29, 87-96.

72.

Lemmens

MAM

, Sierksma

ASR

, Rutten

BPF

, Dennissen

, Steinbusch

HWM

, Lucassen

, Schmitz

(2011) Age-related changes of neuron numbers in the frontal cortex of a transgenic mouse model of Alzheimer’s disease. Brain Struct Funct 216, 227-237.

73.

Olesen

LØ

, Sivasaravanaparan

, Severino

, Babcock

, Bouzinova

, West

, Wiborg

, Finsen

(2017) Neuron and neuroblast numbers and cytogenesis in the dentate gyrus of aged APP swe /PS1 dE9 transgenic mice: Effect of long-term treatment with paroxetine. Neurobiol Dis 104, 50-60.

74.

Heggland

, Storkaas

, Soligard

, Kobro-Flatmoen

, Witter

(2015) Stereological estimation of neuron number and plaque load in the hippocampal region of a transgenic rat model of Alzheimer’s disease. Eur J Neurosci 41, 1245-1262.

75.

Koson

, Zilka

, Kovac

, Kovacech

, Korenova

, Filipcik

, Novak

(2008) Truncated tau expression levels determine life span of a rat model of tauopathy without causing neuronal loss or correlating with terminal neurofibrillary tangle load. Eur J Neurosci 28, 239-246.

76.

Jobst

, Smith

, Szatmari

, Esiri

, Jaskowski

, Hindley

, McDonald

, Molyneux

(1994) Rapidly progressing atrophy of medial temporal lobe in Alzheimer’s disease. Lancet 343, 829-830.

77.

Philippi

, Rousseau

, Noblet

, Botzung

, Després

, Cretin

, Kremer

, Blanc

, Manning

(2015) Different temporal patterns of specific and general autobiographical memories across the lifespan in Alzheimer’s disease. Behav Neurol 2015, 963460.

78.

Yotter

, Doshi

, Clark

, Sojkova

, Zhou

, Wong

, Ferrucci

, Resnick

, Davatzikos

(2013) Memory decline shows stronger associations with estimated spatial patterns of amyloid deposition progression than total amyloid burden. Neurobiol Aging 34, 2835-2842.

79.

Davies

, Mann

DMA

, Sumpter

, Yates

(1987) A quantitative morphometric analysis of the neuronal and synaptic content of the frontal and temporal cortex in patients with Alzheimer’s disease. J Neurol Sci 78, 151-164.

80.

Leuba

, Kraftsik

(1994) Visual cortex in Alzheimer’s disease: Occurencee of neuronal death and glial proliferation, and correlation with pathological hallmarks. Neurobiol Aging 15, 29-43.

81.

Rex

Sheu KF

, Kim

, Blass

, Weksler

(1985) An immunochemical study of the pyruvate dehydrogenase deficit in Alzheimer’s disease brain. Ann Neurol 17, 444-449.

82.

Nielson

, Cummings

, Cotman

(1996) Constructional apraxia in Alzheimer’s disease correlates with neuritic neuropathology in occipital cortex. Brain Res 741, 284-293.

83.

Hof

, Bouras

(1991) Object recognition deficit in Alzheimer’s disease: Possible disconnection of the occipito-temporal component of the visual system. Neurosci Lett 122, 53-56.

84.

Jackson

, Rudinskiy

, Herrmann

, Croft

, Kim

JSM

, Petrova

, Ramos-Rodriguez

, Pitstick

, Wegmann

, Garcia-Alloza

, Carlson

, Hyman

, Spires-Jones

(2016) Human tau increases amyloid β plaque size but not amyloid β-mediated synapse loss in a novel mouse model of Alzheimer’s disease. Eur J Neurosci 44, 3056-3066.

85.

Wang

, Tanila

, Puoliväli

, Kadish

, Van Groen

(2003) Gender differences in the amount and deposition of amyloid-β in APPswe and PS1 double transgenic mice. Neurobiol Dis 14, 318-327.