Neuropathological investigation of the hypometabolic regions on positron emission tomography with [18F] fluorodeoxyglucose in patients with dementia with Lewy bodies

Journal of the Neurological Sciences 314 (2012) 111–119

Contents lists available at SciVerse ScienceDirect

Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns

Neuropathological investigation of the hypometabolic regions on positron emission tomography with [ 18F] fluorodeoxyglucose in patients with dementia with Lewy bodies Koji Kasanuki a, b, Eizo Iseki a, b,⁎, Hiroshige Fujishiro a, b, Ryoko Yamamoto a, b, Shinji Higashi a, c, Michiko Minegishi a, Takashi Togo d, Omi Katsuse d, Hirotake Uchikado d, Yoshiko Furukawa d, Hiroaki Hino c, Kenji Kosaka c, Kiyoshi Sato a, Heii Arai b a

PET/CT Dementia Research Center, Juntendo Tokyo Koto Geriatric Medical Center, Juntendo University School of Medicine, 3-3-20 Shinsuna, Koto-ku, Tokyo 136-0075, Japan Department of Psychiatry, Juntendo University School of Medicine, Bunkyo-ku, Tokyo, Japan c Yokohama Houyuu Hospital, Asahi-ku, Yokohama, Japan d Department of Psychiatry, Yokohama City University School of Medicine, Kanazawa-ku, Yokohama, Japan b

a r t i c l e

i n f o

Article history: Received 25 April 2011 Received in revised form 6 September 2011 Accepted 7 October 2011 Available online 30 October 2011 Keywords: Dementia with Lewy bodies Alzheimer's disease 18 F-FDG PET Neuropathology α-Synuclein Tau Aβ

a b s t r a c t We performed a quantitative neuropathological examination of the hypometabolic regions on FDG PET in dementia with Lewy bodies (DLB), Alzheimer's disease (AD) and control cases. When the DLB cases were divided into two groups according to concomitant AD pathology (ADP), neuronal loss in the temporo-parietal association area was milder in the DLB groups than in the AD group, although there were no differences between the two DLB groups. Tau and Aβ immunoreactivities were observed in the AD group and the DLB group with ADP, but were rare in the DLB group without ADP. Tau and Aβ immunoreactivities as well as numbers of neurofibrillary tangles (NFTs) and neuritic plaques (NPs) were more common in the AD group than in the DLB group with ADP. There was no difference in neuronal loss in the occipital area among the three groups. α-Synuclein immunoreactivity was observed in the DLB groups but not in the AD group. There were no differences in α-synuclein immunoreactivity and number of Lewy bodies (LBs) between the two DLB groups. These findings indicate that the neuropathological bases of the hypometabolic regions in the temporoparietal association and occipital area in DLB may be AD pathology and Lewy pathology, respectively. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Over the past two decades, neuroimaging strategies for the diagnosis of neurodegenerative diseases have progressed dramatically. Many studies using single photon emission computed tomography (SPECT) and positron emission tomography with fluorodeoxyglucose (FDG PET) have produced images characteristic of Alzheimer's disease (AD) and dementia with Lewy bodies (DLB), the most frequent and second most frequent neurodegenerative dementia diseases, respectively [1–4]. In AD patients, FDG PET using the 3D-SSP analysis shows a regional hypometabolic pattern in the posterior cingulate cortex (PCC), precuneus, and temporo-parietal association area [2]. A similar pattern is also found in the PCC and precuneus in patients with mild cognitive impairment (MCI) developing AD [3]. The hypometabolism in these regions does not mean they are the regions initially affected by AD ⁎ Corresponding author at: PET/CT Dementia Research Center, Juntendo Tokyo Koto Geriatric Medical Center, Juntendo University School of Medicine, 3-3-20 Shinsuna, Koto-ku, Tokyo 136-0075, Japan. Tel.: + 81 3 5632 3111; fax: + 81 3 5632 3728. E-mail address: [email protected] (E. Iseki). 0022-510X/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2011.10.010

pathology, such as neurofibrillary tangles (NFTs) and senile plaques, but it is likely to reflect a secondary metabolic defect from AD pathology in the medial temporal area including the parahippocampal cortex [5–7]. In addition, the evolution of regional hypoperfusion on SPECT is correlated with the degree of Braak NFT stage [8]. These previous studies suggest a close relationship between functional neuroimaging and AD pathology in AD patients, because there is robust accordance of AD-specific patterns by SPECT and FDG PET [9]. In DLB patients, FDG PET hypometabolism in the occipital area, including the primary visual cortex (PVC) and visual association cortex (VAC), is a specific feature by which one can differentiate DLB from AD [4,10]. DLB patients show characteristic visual hallucinations, which may be related to the hypoperfusion/hypometabolism in the occipital area on SPECT/FDG PET [11–13]. Minoshima et al. [4] showed that the occipital hypometabolism on FDG PET occurs independently from the frontal or posterior hypometabolism in DLB patients by multi-variables analysis, suggesting some specific pathophysiology in the occipital area. In addition to the occipital area, a regional hypometabolic pattern in the PCC and temporo-parietal association area including the inferior lateral parietal cortex (ILP) is also found in DLB patients [14]. DLB neuropathologically shows

112

K. Kasanuki et al. / Journal of the Neurological Sciences 314 (2012) 111–119

Lewy pathology such as Lewy bodies (LBs) and Lewy neurites (LNs), which are frequently accompanied by AD pathology [13]. Therefore, the regional hypometabolism in the PCC and temporo-parietal association area in DLB patients is likely to reflect concomitant AD pathology. There have been few studies of the relationship between functional neuroimaging and neuropathology in DLB patients [1,15]. The aim of this study was to perform a quantitative neuropathological examination of the regions where FDG PET hypometabolism is commonly detected in brains of DLB patients, and clarify the neuropathological features indicated by the regional hypometabolism on FDG-PET.

the design of the study: a DLB group without AD pathology (ADP), defined as Braak NFT stage I–II and CERAD amyloid stage 0–A, and a DLB group with ADP, defined as Braak NFT stage III–VI and CERAD amyloid stage B–C. Consequently, six cases were included in the DLB group without ADP, and 16 cases were included in the DLB group with ADP (Table 1). All autopsies were undertaken with written consent, and the study was approved by the ethics committee of the Juntendo Tokyo Koto Geriatric Medical Center.

2. Materials and methods

Brains were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Two coronal slices of the cerebral hemispheres through the posterior end of the putamen and the tip of the posterior horn were embedded in paraffin. Six-micrometer-thick sections were stained using the hematoxylin-eosin (HE) and Klϋver-Barrera (KB) methods for pathological examination. Serial sections were also immunostained with the following antibodies: anti-phosphorylated α-synuclein (PSer129: monoclonal, mouse, 1:20,000, donated by Dr. T. Iwatsubo) [19], anti-PHF tau (AT8: monoclonal, mouse, 1:2000, Innogenetics, Belgium), and anti-Aβ (polyclonal, rabbit, 1:5000, donated by Dr. T. Ishii). Immunolabeling was detected using the avidin–biotin–horseradish peroxidase complex (ABC) method (Elite Kit: Vector, USA) and visualized with diaminobenzidine (DAB: Wako, Japan). Before immunostaining with PSer129 and anti-Aβ antibodies, sections were pretreated with 70% formic acid for 10 min. Lewy pathology, namely LBs and LNs, was evaluated by PSer129 immunostaining, while AD pathology, namely NFTs and senile plaques, was evaluated by AT8- and Aβ immunostaining, respectively.

2.1. FDG PET images of DLB and AD patients In Fig. 1, we demonstrate representative FDG PET images using the 3D-SSP analysis of DLB and AD patients, which were compared with those in an age-matched normative database, as previously reported [16]. Fig. 1A and B are those of the DLB patients which show hypometabolism predominantly in the occipital area, and hypometabolism in the PCC and temporo-parietal association area in addition to the occipital area, respectively. Fig. 1C is those of the AD patient which shows hypometabolism in the PCC and temporo-parietal association area, but not in the occipital area. 2.2. Brain cases Autopsied brains from 22 DLB cases (14 males and eight females, mean age 74.5 years, mean brain weight 1146 g) and 10 AD cases (six males and four females, mean age 79.6 years, mean brain weight 1021 g), which were preserved in the PET/CT Dementia Research Center, were neuropathologically examined. These DLB and AD cases fulfilled the neuropathological criteria for DLB [13] and AD [17,18], respectively. In addition, brains from seven elderly control cases (four males and three females, mean age 74.1 years, mean brain weight 1276 g) who had no previous history of neurological diseases or no evidence of significant neuropathological abnormalities were examined. Subdivisions of these brain cases based on neuropathology are shown in Table 1. In this study, the 22 DLB cases were divided into two groups according to the severity of concomitant AD pathology to simplify

2.3. Neuropathological methods

2.4. Identification of regions of interest Four regions of interest were examined, namely, the PCC, ILP, PVC, and VAC. The PCC and ILP were examined in a coronal slice at the level of posterior end of the putamen. At this level, the region adjacent to the corpus callosum (Brodmann Area: BA 23) was identified as the PCC, and the angular gyrus (BA 39) and supramarginal gyrus (BA40) were represented as the ILP. In a coronal slice at the level of the tip of the posterior horn, the area striate (BA17) and the basal aspect of the occipital area (BA 19) were identified as the PVC and VAC, respectively [20].

Fig. 1. FDG PET images of DLB and AD patients (A)–(C) demonstrate representative FDG PET images using the 3D-SSP analysis of DLB and AD patients. (A) and (B) are those of the DLB patients which show hypometabolism predominantly in the occipital area, and hypometabolism in the posterior cingulate cortex and temporo-parietal association area in addition to the occipital area, respectively. (C) is those of the AD patient which shows hypometabolism in the posterior cingulate cortex and temporo-parietal association area, but not in the occipital area.

K. Kasanuki et al. / Journal of the Neurological Sciences 314 (2012) 111–119

113

Table 1 Neuropathological characteristics of brain cases. DLB without: DLB group without ADP. DLB with: DLB group with ADP. AD: AD group. Cont: Control group. Lewy: Lewy pathology type. Diffuse: diffuse neocortical type. Lim: limbic type. NFT: Braak NFT stage. Amyloid: CERAD amyloid stage. Likelihood: likelihood of Lewy pathology for clinical features typical of DLB. Group

Case

Age (y.o.)

Sex

Brain weight (g)

Lewy

NFT

Amyloid

Likelihood

DLB without

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7

47 50 72 76 81 84 62 67 71 74 74 75 75 77 78 79 79 79 82 84 84 88 56 76 79 79 81 81 83 86 86 89 64 65 68 79 79 80 84

F M M M M M F M M M F F F F M M M F F M M M M M F F F M M M M F F F M F M M M

1060 1050 1370 1320 1400 1145 980 1030 1320 1300 1010 1240 1065 900 1230 1165 1060 1173 1050 1060 1140 1145 1240 1190 970 950 1070 1020 940 950 950 930 1320 1350 1350 1160 1259 1260 1230

Diffuse Diffuse Lim Lim Lim Diffuse Diffuse Diffuse Diffuse Diffuse Lim Diffuse Diffuse Diffuse Diffuse Diffuse Diffuse Diffuse Diffuse Diffuse Lim Lim – – – – – – – – – – – – – – – – –

II II II II II II VI V III III IV IV III III IV VI IV III VI IV III IV V VI VI IV VI VI VI VI VI VI I I I I I II II

0 0 0 0 0 A C C B C B C B C C C B C C B C C C C C C C C C C C C 0 0 0 0 0 0 0

High High High High High High Intermediate Intermediate High High Intermediate High High High High Intermediate High High Intermediate High Intermediate Intermediate – – – – – – – – – – – – – – – – –

DLB with

AD

Cont

2.5. Quantification of neuron

2.7. Statistical analysis

In each region, the number of neurons from two cortical layers (layers III and V) was measured. Neurons were differentiated from glial cells on the basis of morphological features such as somal size, pyramidal somas and nuclei containing clear chromatin. For each region, three randomly selected images at a 400× magnification were captured by a CCD color camera coupled to an Olympus microscope, and the number of neurons in the obtained images was automatically calculated using imaging software (Image Pro Plus7.0®, Media Cybernetics, MD, USA) (Fig. 2A).

Statistical analysis was performed using Graph Pad Prism5® (GraphPad Software, Inc., San Diego, CA) and PASW program version 18 for Windows XP (SPSS, Inc., Chicago, IL). For the neuropathological quantification, a digital photomicrograph of 2776 × 2074 pixels, which corresponded to 990 μm × 1290 μm in tissue sections, was taken from two cortical layers in each case. The number of neurons was represented as the median. Because of the lack of Gaussian distributions by Shapiro–Wilk test, the variables were examined by the statistical analysis using nonparametric procedures. The differences in the neuronal number, and the amounts of α-synuclein-, tau- and Aβ immunoreactivities were analyzed among the DLB group without ADP, the DLB group with ADP, the AD group, and the control group using the Kruskal–Wallis test, followed by the Dunn post-hoc test for multiple group comparisons. The differences in the numbers of LBs, NFTs, and NPs were also analyzed by the same method. The differences between two regions of interest were determined by the Mann–Whitney test. p-Values less than 0.05 or less than 0.10 were accepted as significant or showing a tendency in these nonparametric tests, respectively. Spearman correlations and multiple regression analysis were used to test the correlation between the neuronal number and the amounts of α-synuclein-, tau- and Aβ immunoreactivities as well as the numbers of LBs, NFTs, and SPs. Confidence intervals for regression coefficients were calculated with 95% confidence limits (p b 0.05).

2.6. Quantification of α-synuclein, tau- and Aβ-immunoreactivity In each region, α-synuclein-, tau and Aβ immunoreactivity were measured in two cortical layers (layers III and V). First, three randomly selected images at a 400× magnification were captured by a CCD color camera in each region, and the amount of α-synuclein, tau and Aβ immunoreactivity in these obtained images was automatically calculated according to the color density using the imaging soft. Second, LBs, NFTs and neuritic plaques (NPs) were identified from αsynuclein, tau and Aβ immunoreactivities according to morphological features, respectively, and the average numbers of these specific structures in each region were calculated (Fig. 2B–D).

114

K. Kasanuki et al. / Journal of the Neurological Sciences 314 (2012) 111–119

Fig. 2. Computerized quantitative analysis using Image Pro Plus7.0® (A) Neurons (KB stain) surrounded by red circles. (B) α-Synuclein immunoreactivity detected by color density (red dots) and Lewy bodies detected by morphological feature (red circles). (C) Tau immunoreactivity detected by color density (red dots) and neurofibrillary tangles detected by morphological feature (red circles). (D) Aβ immunoreactivity detected by color density (red dots) and neuritic plaques detected by morphological feature (red circles).

3. Results 3.1. Characteristics of brain cases The six cases in the DLB group without ADP comprised three cases of the limbic type, and three cases of the diffuse neocortical type; there were six cases of high likelihood of Lewy pathology for clinical features typical of DLB. The sixteen cases in the DLB group with ADP comprised three cases of the limbic type, and 13 cases of the diffuse neocortical type; there were nine cases of high likelihood, and seven cases of intermediate likelihood. Consequently, the proportion of limbic type was higher in the DLB without ADP group (50%) than in the DLB with ADP group (18.8%),

while that of the diffuse neocortical type was higher in the DLB with ADP group (81.2%) than in the DLB without ADP group (50%). In addition, the proportion of high likelihood of Lewy pathology for clinical features typical of DLB was higher in the DLB without ADP group (100%) than in the DLB with ADP group (56.3%). 3.2. Neuronal number The regional neuronal numbers in each group are shown in Fig. 3. In the PCC and ILP, the neuronal number was significantly less (p b 0.0001, p b 0.0001) in the AD group than in the two DLB groups, although there were no significant differences among the DLB group without ADP, the DLB group with ADP, and the control group. In the

K. Kasanuki et al. / Journal of the Neurological Sciences 314 (2012) 111–119

115

PVC and VAC (p b 0.0001). There was no significant difference in the amount of α-synuclein immunoreactivity between the PCC and the ILP, or between the PVC and the VAC. The number of LBs tended to be higher in the PCC and ILP than in the PVC (p b 0.01), but was not higher than that in the VAC (Fig. 5A). The amount of α-synuclein immunoreactivity in the DLB group with ADP was significantly higher in the PCC and ILP than that in the PVC and VAC (p b 0.0001). There was no significant difference in the amount of α-synuclein immunoreactivity between the PCC and ILP, or between the PVC and VAC. The number of LBs was significantly higher in the PCC, ILP and VAC than in the PVC (p b 0.0001) (Fig. 5A). 3.4. Tau immunoreactivity Fig. 3. Neuronal number. Neuronal numbers in the PCC, ILP, PVC, and VAC in the DLB group without ADP (DLB without ADP), the DLB group with ADP (DLB with ADP), the AD group (AD), and the control group (Cont). The columns represent the average neuronal number. Error bars give +1 SD for each group. PCC: posterior cingulate cortex, ILP: inferior lateral parietal cortex, PVC: primary visual cortex, VAC: visual association cortex.

PVC and VAC, no significant differences were found among the four groups. 3.3. α-Synuclein immunoreactivity α-Synuclein immunoreactivity was found in all four regions in the two DLB groups, but not in the AD and control groups. LBs were detected as intracytoplasmic round or oval structures, whereas LNs were detected as curly or dot-like structures surrounding LBs. Both LBs and LNs were seen more commonly in the ILP (Fig. 4A) than in the PVC (Fig. 4B). In all four regions, there were no significant differences in both the amount of α-synuclein immunoreactivity and the number of LBs between the two DLB groups, except that the number of LBs in the ILP was significantly higher in the DLB group with ADP than in the DLB group without ADP (p b 0.05) (Fig. 5A). In the DLB group without ADP, the amount of α-synuclein immunoreactivity was significantly higher in the PCC and ILP than in the

Tau immunoreactivity was found in all four regions in the AD group and the DLB group with ADP, but rare in the DLB group without ADP and the control group. In all four regions, both the amount of tau immunoreactivity and the number of NFTs were significantly higher in the AD group than in the DLB group with ADP (p b 0.0001) (Fig. 5B). In the DLB group with ADP, both the amount of tau immunoreactivity and the number of NFTs were significantly lower in the PVC than in the PCC, ILP and VAC (p b 0.01), although there were no significant differences between the PCC, ILP and VAC except that the number of NFTs was significantly more in the PCC than in the PVC (p b 0.01) (Fig. 5B). The amount of tau immunoreactivity in the AD group was significantly higher in the PCC and ILP than in the PVC and VAC (p b 0.001). There was no significant difference in the amount of tau immunoreactivity between the PCC and ILP or between the PVC and VAC. The number of NFTs was significantly lower in the PVC than in the PCC and ILP (p b 0.05), although there were no significant differences among the PCC, ILP and VAC. There was no significant difference in the number of NFTs between the PVC and VAC (Fig. 5B). 3.5. Aβ immunoreactivity Aβ immunoreactivity was found in all four regions in the AD group and the DLB group with ADP, but rare in the DLB group without ADP and the control group. The amount of Aβ immunoreactivity in the PCC was significantly higher in the AD group than in the DLB group with ADP (p b 0.05), although there was no significant difference between the AD group and the DLB group with ADP in terms of Aβ immunoreactivity in the ILP. The numbers of NPs in the PCC and ILP were significantly higher in the AD group than in the DLB group with ADP (p b 0.0001, p b 0.05). Both the amount of Aβ immunoreactivity and the number of NPs in the PVC (p b 0.0001, p b 0.0001) and VAC (p b 0.0001, p b 0.0001) were significantly higher in the AD group than in the DLB group with ADP (Fig. 5C). In the DLB group with ADP, there were no significant differences in either the amount of Aβ immunoreactivity or the number of NPs among the four regions, although the number of NPs tended to be higher in the ILP than in the PCC, PVC and VAC (p b 0.10) (Fig. 5C). In the AD group, both the amount of Aβ immunoreactivity and the number of NPs were significantly higher in the PVC and VAC than in the ILP (p b 0.05, p b 0.05), although there were no significant differences among the PCC, PVC, and VAC (Fig. 5C). 3.6. Correlation between neuronal number and α-synuclein, tau or Aβ immunoreactivity

Fig. 4. α-Synuclein immunoreactivities. LBs were detected as intracytoplasmic round or oval structures (arrows), whereas LNs were detected as curly or dot-like structures surrounding LBs. Both LBs and LNs were seen more commonly in the ILP (A) than in the PVC (B). Bar = 100 μm.

In the DLB groups, the correlations between neuronal numbers and α-synuclein, tau and Aβ immunoreactivities were analyzed. Negative correlations between neuronal number and the amounts of αsynuclein and Aβ immunoreactivity were found in the VAC and IPL, respectively. The neuronal number was significantly negatively correlated with the amount of tau immunoreactivity, the number of NFTs,

116

K. Kasanuki et al. / Journal of the Neurological Sciences 314 (2012) 111–119

Fig. 5. α-Synuclein, tau and Aβ immunoreactivities. (A) The amount of α-synuclein immunoreactivity and the number of LBs in the PCC, ILP, PVC, and VAC in the DLB group without ADP, the DLB group with ADP, the AD group, and the control group. (B) The amount of tau immunoreactivity and the number of NFTs. (C) The amount of Aβ immunoreactivity and the number of NPs. The columns represent the average amounts and numbers. Error bars give + 1 SD for each group. α-Synuclein immunoreactivity: the amount of α-synuclein immunoreactivity. Tau immunoreactivity: the amount of tau-immunoreactivity. Aβ immunoreactivity: the amount of Aβ immunoreactivity. LBs: the number of Lewy bodies. NFTs: the number of neurofibrillary tangles. NPs: the number of neuritic plaques.

and the NFT Braak stage in the PCC and IPL, although there were no such correlations in the PVC and VAC (Fig. 6). In the AD group, negative correlations between neuronal number and the amount of tau immunoreactivity, the number of NFTs, and the NFT Braak stage were found in all four regions. To clarify the neuropathological features related to neuronal number in each region in the DLB groups, multiple regression analysis with forward–backward stepwise selection was applied for the amounts of α-synuclein, tau and Aβ immunoreactivity (Table 2). Consequently, the amount of tau immunoreactivity was regarded as a significant variable (partial regression coefficient − 0.844) in the PCC. In the ILP, the amount of tau or Aβ immunoreactivity was detected as a significant variable (partial regression coefficient −0.380 or − 0.491, respectively). In the PVC, the amount of tau-immunoreactivity was detected as a significant variable (partial regression coefficient −0.520). In the VAC, the amount of α-synuclein-immunoreactivity

was detected as a significant variable (partial regression coefficient −0.454). 4. Discussion DLB is a neurodegenerative dementia disease that is clinically characterized by cognitive decline, visual hallucinations, and parkinsonism. DLB is neuropathologically characterized by neuronal loss with Lewy pathology such as LBs and LNs throughout the brain, which is frequently accompanied with AD pathology such as NFTs and senile plaques [13,21]. In DLB, the relationship between clinical features and neuropathology remains unclear, although the conception of likelihood was introduced to determine the neuropathological basis of clinical features typical of DLB [13]. Functional neuroimaging methods such as SPECT and FDG PET are helpful for the clinical diagnosis of DLB. FDG PET characteristically

K. Kasanuki et al. / Journal of the Neurological Sciences 314 (2012) 111–119

117

Fig. 6. Correlation between neuronal number and NFT Braak stag. (A) PCC, (B) ILP, (C) PVC, (D) VAC. A correlation is seen in the PCC and ILP.

reveals hypometabolism in the occipital area, which is frequently accompanied by hypometabolism in the PCC and temporo-parietal association area in DLB patients [10,15]. Minoshima et al. [4] indicated with multivariate analysis that there is a pathophysiological difference between FDG PET hypometabolism in the temporo-parietal association area and that in the occipital area, especially in the PVC, in DLB patients. The detection of Lewy pathology in brains is essential for the pathological diagnosis of DLB. Previous studies, however, have shown that Lewy pathology and AD pathology mostly coexist in DLB brains [22,23]. Fujishiro et al. [23] reported that, among 42 autopsied DLB

Table 2 Data of multiple regression analysis between number of neurons and neuropathological variables. Neuropathological variables

Standardized regression coefficient

PCC Aβ Tau α-Synuclein

− 0.096 − 0.844 − 0.162

ILP Aβ Tau α-Synuclein

− 0.491 − 0.380 − 0.138

PVC Aβ Tau α-Synuclein

− 0.050 − 0.520 − 0.359

VAC Aβ Tau α-Synuclein

0.104 − 0.263 − 0.454

a

Regression coefficient

95% confidence intervala

− 0.393

− 0.510

− 0.276

− 0.150 − 0.250

− 0.262 − 0.403

− 0.038 − 0.007

− 0.479

− 0.845

− 0.112

0.807 0.013 0.060

− 0.253

− 4.850

− 0.211

0.034

Per 1000 pixel × pixel unit increment for the continuous variables.

p Value

0.512 b0.0001 0.187

0.011 0.0043 0.366

brains, about 80% had a moderate to severe extent of concomitant AD pathology (NIA Intermediate to High), and the remaining 20% of brains had only mild AD pathology (NIA Low). In this study, moderate to severe AD pathology (NIA Intermediate to High) was found in 16 of the 22 DLB cases (72.2%), consistent with the previous findings. Few studies have investigated the relation between FDG PET neuroimaging and neuropathology in DLB patients [1,15]. Albin et al. [15] reported that the differences in both the regional hypometabolic pattern and neuropathological distribution were unclear between the DLB cases with AD pathology and DLB cases without AD pathology, although this study was based on a small number of cases. Higuchi et al. [1] showed that the spongiform change with gliosis was found throughout the subcortical white matter in DLB brains, prominently in the occipital area, and the severity of spongiform change paralleled the degree of hypometabolism on FDG PET, suggesting that disconnection of cortico-cortical or cortico-subcortical long projection fibers may lead to secondary impairment of cortical glucose metabolism in DLB patients. In this study, we quantitatively assessed α-synuclein, tau, and Aβ immunoreactivities in addition to neuronal number in the occipital area, PCC and temporo-parietal association area, and compared them among the DLB group without ADP, the DLB group with ADP, and the AD group. The DLB group with ADP had a higher proportion of the diffuse neocortical type, and low likelihood of Lewy pathology for clinical features typical of DLB than the DLB group without ADP, indicating that the DLB group with ADP had not only more severe AD pathology, but also more severe Lewy pathology than the DLB group without ADP. This finding supports recent increasing evidence that cortical amyloid deposits accelerate the accumulation of αsynuclein [24,25]. In the PCC and temporo-parietal association area, neuronal loss was milder in the DLB groups than in the AD group, although there were no differences between the two DLB groups. Both tau and Aβ immunoreactivities differed between the two DLB groups, being much higher in the DLB group with ADP than in the DLB group without ADP, although they were rather lower in the DLB group with ADP

118

K. Kasanuki et al. / Journal of the Neurological Sciences 314 (2012) 111–119

than in the AD group. In the DLB groups, tau and Aβ immunoreactivities participated in neuronal loss by multivariate analysis, and the severity of NFT pathology was correlated with the degree of neuronal loss. Chételat et al. [26] reported that local atrophy of the PCC based on neuronal loss partly contributes to the hypometabolism in the PCC in AD patients. With regard to the hypometabolism in the PCC, the role of white matter disruption and disconnection with the hippocampus is also emphasized [27]. In addition, the hypometabolism in the PCC and temporo-parietal association area in AD patients may reflect a secondary defect from AD pathology in the medial temporal area [5–7]. These findings suggest that both primary and secondary effects of AD pathology may play an important role in the hypometabolism in the PCC and temporo-parietal association area in DLB patients. In contrast, α-synuclein immunoreactivity in the PCC and temporo-parietal association area showed no difference between the two DLB groups, except that the number of LBs in the ILP was significantly higher in the DLB group with ADP than in the DLB group without ADP. In addition, α-synuclein immunoreactivity did not correlate with neuronal loss by multivariate analysis. In short, the neuropathological basis of FDG PET hypometabolism in the PCC and temporo-parietal association area in DLB patients may be AD pathology, but not Lewy pathology. In the occipital area, there were no differences in neuronal loss among the three groups. Both tau and Aβ immunoreactivities showed significant differences between the two DLB groups, being higher in the DLB group with ADP than in the DLB group without ADP, although they were rather lower in the DLB group with ADP than in the AD group. In the DLB groups, tau immunoreactivity participated in neuronal loss only in the PVC by multivariate analysis, while the severity of NFT showed no correlation with the degree of neuronal loss. In contrast, α-synuclein immunoreactivity showed no difference between the two DLB groups. In the DLB groups, α-synuclein immunoreactivity participated in neuronal loss only in the VAC by multivariate analysis, while the severity of LB pathology showed no correlation with the degree of neuronal loss. Previous studies have detected NFTs in the occipital area in 24–52% of brains of nondemented subjects [19,28]. In contrast, incidental Lewy pathology was found in 24% of brains of longitudinally followed elderly subjects, but the occipital area contained no Lewy pathology [29]. These findings suggest that Lewy pathology may occur independently from AD pathology in the occipital area, and that primary and secondary effects of Lewy pathology may play an important role in the hypometabolism in the occipital area in DLB patients. The pathomechanism of the secondary effect of Lewy pathology in the occipital area remains unclear. More severely involved regions by Lewy pathology, such as the amygdala, Meynert nucleus and insular cortex, do not show prominent hypometabolism on FDG PET in DLB patients. Mori et al. [30] reported that the administration of donepezil improved hypoperfusion in the occipital area in DLB patients, suggesting a secondary acetylcholinergic defect from Lewy pathology in the forebrain or brainstem. We showed that Lewy pathology in DLB brains first occurs in the axonal terminals including LNs, but not in the neuronal cell bodies including LBs [31], that secondary LBs may be formed by transneuronal degeneration in the regions where the degenerative axonal terminals are found [31,32], and that Lewy pathology in the occipital area may arise along with degeneration of the visuo-amygdaloid pathway [33]. These findings suggest that the occipital hypometabolism on FDG PET in DLB patients may reflect a secondary metabolic defect similar to the hypometabolism in the PCC and temporo-parietal association area in AD patients, which are caused by degeneration of long projection fibers to the occipital area. In short, the neuropathological basis of the FDG PET hypometabolism in the occipital area in DLB patients may be Lewy pathology rather than AD pathology. In conclusion, this study provides further evidence that the neuropathological bases of the regions where FDG PET hypometabolism is

commonly detected in brains of DLB patients are different between the PCC and parieto-temporal association area and the occipital area, and that these neuropathological bases differed between DLB groups with and without AD pathology for tau and Aβ but not αsynuclein. Accordingly, the different patterns of FDG PET hypometabolism seem to reflect the heterogeneity of DLB, both AD pathology and Lewy pathology. The chief limitations of this study are that none of the examined brain cases had undertaken the FDG PET, and that this neuropathological study examined the primary effect of AD pathology or Lewy pathology on the regional FDG-PET hypometabolism but not the secondary effect which has not been clarified yet. In the future, a followup neuropathological study of examined cases on FDG PET with a postmortem autopsy and a dynamic functional imaging study which reflects the secondary effect of AD pathology or Lewy pathology should be performed. Conflict of interest The authors have no conflicts of interests to declare. Acknowledgments This study was supported, in part, by the Anti-aging Research Center of Juntendo University School of Medicine, the Ogasawara Foundation for the promotion of Science and Engineering, Nihon MediPhisics, and a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology in Japan. References [1] Higuchi M, Tashiro M, Arai H, Okamura N, Hara S, Higuchi S, et al. Glucose hypometabolism and neuropathological correlates in brains of dementia with Lewy bodies. Exp Neurol 2000;162:247–56. [2] Minoshima S, Foster NL, Kuhl DE. Posterior cingulate cortex in Alzheimer's disease. Lancet 1994;344:895. [3] Minoshima S, Giordani B, Berent S, Frey KA, Foster NL, Kuhl DE. Metabolic reduction in the posterior cingulate cortex in very early Alzheimer's disease. Ann Neurol 1997;42:85–94. [4] Minoshima S, Foster NL, Sima AA, Frey KA, Albin RL, Kuhl DE. Alzheimer's disease versus dementia with Lewy bodies: cerebral metabolic distinction with autopsy confirmation. Ann Neurol 2001;50:358–65. [5] Haxby JV, Grady CL, Koss E, Horwitz B, Heston L, Schapiro M, et al. Longitudinal study of cerebral metabolic asymmetries and associated neuropsychological patterns in dementia of the Alzheimer type. Arch Neurol 1990;47:977–80. [6] Mega MS, Chen SS, Thompson PM, Woods RP, Karaca TJ, Tiwari A, et al. Mapping histology to metabolism: coregistration of stained whole-brain sections to premortem PET in Alzheimer's disease. Neuroimage 1997;5:147–53. [7] Mosconi L, Pupi A, De Cristofaro MT, Fayyaz M, Sorbi S, Herholz K, et al. Functional interactions of the entorhinal cortex: an 18F-FDG PET study on normal aging and Alzheimer's disease. J Nucl Med 2004;45:382–92. [8] Bradley KM, O'Sullivan VT, Soper ND, Nagy EMF, Smith AD, Shepstone BJ, et al. Cerebral perfusion SPET correlated with Braak pathological stage in Alzheimer's disease. Brain 2002;125:1772–81. [9] Herholz K, Schopphoff H, Schmidt M, Mielke R, Eschner W, Sheidhauer K, et al. Direct comparison of spatially normalized PET and SPECT scans in Alzheimer's disease. J Nucl Med 2002;43:21–6. [10] Ishii K, Imamura T, Sasaki M, Yamaji S, Sakamoto S, Kitagaki H, et al. Regional cerebral glucose metabolism in dementia with Lewy bodies and Alzheimer's disease. Neurology 1998;51:125–30. [11] Gilman S, Koeppe RA, Little R, An H, Junck L, Giordani B, et al. Differentiation of Alzheimer's disease from dementia with Lewy bodies utilizing positron emission tomography with [18F] fluorodeoxyglucose and neuropsychological testing. Exp Neurol 2005;191:S95–S103. [12] Imamura T, Ishii K, Hirono N, Hashimoto M, Tanimukai S, Kazuai H, et al. Visual hallucinations and regional cerebral metabolism in dementia with Lewy bodies (DLB). Neuroreport 1999;10:1903–7. [13] McKeith IG, Dickson DW, Lowe J, Emre M, O'Brien JT, Feldman H, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB consortium. Neurology 2005;65:1863–72. [14] Mosconi L, Tsui WH, Herholz K, Pupi A, Drzezga A, Lucignani G, et al. Multicenter standardized 18F-FDG PET diagnosis of mild cognitive impairment, Alzheimer's disease, and other dementias. J Nucl Med 2008;49:390–8. [15] Albin RL, Minoshima S, D'Amato CJ, Frey KA, Kuhl DA, Sima AAF, et al. Fluorodeoyglucose positron emission tomography in diffuse Lewy body disease. Neurology 1996;47:462–8.

K. Kasanuki et al. / Journal of the Neurological Sciences 314 (2012) 111–119 [16] Iseki E, Murayama N, Yamamoto R, Fujishiro H, Suzuki M, Kawano M, et al. Construction of a 18F-FDG PET normative database of Japanese healthy elderly subjects and its application to demented and mild cognitive impairment patients. Int J Geriatr Psychiatry 2010;25:352–61. [17] Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol (Berl) 1991;82:239–59. [18] Mirra SS, Heyman A, McKeel D, Sumi MD, Crain BJ, Brownlee LM, et al. The Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer's disease. Neurology 1991;41:479–86. [19] Saito Y, Kawashima A, Ruberu NN, Fujiwara H, Koyama S, Sawabe M, et al. Accumulation of phosphorylated alpha-syunuclein in aging human brain. J Neuropathol Exp Neurol 2003;62:644–54. [20] Pikkarainen M, Kauppinen T, Alafuzoff I. Hyperphosphorylated tau in the occipital cortex in aged nondemented subjects. J Neuropathol Exp Neurol 2009;68:653–60. [21] McKeith IG, Galasko D, Kosaka K, Perry EK, Dickson DW, Hansen LA, et al. Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB). Neurology 1996;47:1113–24. [22] Aarsland D, Perry R, Brown A, Larsen JP, Ballard C. Neuropathology of dementia in Parkinson's disease: a prospective, community-based study. Ann Neurol 2005;58: 773–6. [23] Fujishiro H, Ferman TJ, Boeve BF, Smith GE, Graff-Radford NR, Uitti RJ, et al. Validation of the neuropathologic criteria of the third consortium for dementia with Lewy bodies for prospectively diagnosed cases. J Neuropathol Exp Neurol 2008;67:649–56. [24] Lashley T, Holton JL, Gray E, Kirkham K, O'Sullivan S, Hilbig A, et al. Cortical alphasynuclein load is associated with amyloid-beta plaque burden in a subset of Parkinson's disease patients. Acta Neuropathol (Berl) 2008;115:417–25.

119

[25] Obi K, Akiyama H, Kondo H, Shimomura Y, Hasegawa M, Iwatsubo T, et al. Relationship of phosphorylated alpha-synuclein and tau accumulation to Abeta deposition in the cerebral cortex of dementia with Lewy Bodies. Exp Neurol 2008;210: 409–20. [26] Chételat G, Villain N, Desgranges B, Eustache F, Baron JC. Posterior cingulate hypometabolism in early Alzheimer's disease: what is the contribution local atrophy versus disconnection? Brain 2009;132:1–2. [27] Villain N, Desgranges B, Viader F, de la Sayette V, Mézenge F, Landeau B, et al. Relationships between hippocampal atrophy, white matter disruption, and gray matter hypometabolism in Alzheimer's disease. J Neurosci 2008;24:6174–81. [28] McKee AC, Au R, Cabral HJ, Kowall NW, Seshadri S, Kubilus CK, et al. Visual association pathology in preclinical Alzheimer disease. J Neuropathol Exp Neurol 2006;65:621–30. [29] Markesbery WR, Jicha GA, Liu H, Schmitt FA. Lewy body pathology in normal elderly subjects. J Neuropathol Exp Neurol 2009;68:816–22. [30] Mori T, Ikeda M, Fukuhara R, Nestor PJ, Tanabe H. Correlation of visual hallucination with occipital rCBF changes by donepezil in DLB. Neurology 2006;66:935–7. [31] Iseki E, Marui W, Kosaka K, Akiyama H, Ueda K, Iwatsubo T. Degenerative terminals of the perforant pathway are human α-synuclein-immunoreactive in the hippocampus of patients with diffuse Lewy body disease. Neurosci Lett 1998;258:81–4. [32] Katsuse O, Iseki E, Marui W, Kosaka K. Developmental stages of cortical Lewy bodies and their relation to axonal transport blockage in brains of patients with dementia with Lewy bodies. J Neurosci 2003;211:29–35. [33] Yamamoto R, Iseki E, Murayama N, Minegishi M, Marui W, Togo T, et al. Investigation of Lewy pathology in the visual pathway of brains of dementia with Lewy bodies. J Neurol Sci 2006;246:95–101.