Neurobiology of Aging 35 (2014) 1519e1525
Contents lists available at ScienceDirect
Neurobiology of Aging journal homepage: www.elsevier.com/locate/neuaging
Association of homocysteine with hippocampal volume independent of cerebral amyloid and vascular burden Young Min Choe a, Bo Kyung Sohn b, Hyo Jung Choi a, Min Soo Byun a, Eun Hyun Seo a, Ji Young Han a, Yu Kyeong Kim c, Eun Jin Yoon c, Jong-Min Lee d, Jinsick Park d, Jong Inn Woo a, Dong Young Lee a, * a
Department of Neuropsychiatry, Seoul National University Hospital, Seoul, Korea Department of Neuropsychiatry, SMG-SNU Boramae Medical Center, Seoul, Korea Department of Nuclear Medicine, SMG-SNU Boramae Medical Center, Seoul, Korea d Department of Biomedical Engineering, Hanyang University, Seoul, Korea b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 April 2013 Received in revised form 20 December 2013 Accepted 12 January 2014 Available online 17 January 2014
This study aimed to clarify whether homocysteine has independent association, not mediated by cerebral beta amyloid protein deposition and vascular burden, with whole brain or hippocampal volume in elderly individuals with normal cognition, mild cognitive impairment, and Alzheimer’s disease. Nineteen mild cognitive impairment and 24 Alzheimer’s disease patients were recruited from the Dementia Clinic of the Seoul National University Hospital. Fourteen cognitively normal elderly subjects were also selected from a pool of elderly volunteers. Multiple linear regression analyses showed that plasma total homocysteine level was significantly associated with hippocampal volume even after controlling the degree of global cerebral beta amyloid deposition and vascular burden as well as other potential confounders including age, gender, education, and apolipoprotein E ε4 genotype. On the contrary, plasma total homocysteine level did not show any significant association with whole brain volume. Our finding of the independent negative association between homocysteine and hippocampal volume suggests that homocysteine has a direct adverse effect, not mediated by cerebral beta amyloid deposition and vascular burden, on the hippocampus. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Homocysteine Hippocampus Amyloid PIB Vascular burden Alzheimer’s disease Mild cognitive impairment
1. Introduction Homocysteine has been proposed to be one of the risk factors for cardiovascular disease (Stampfer et al., 1992), carotid stenosis (Selhub et al., 1995), and stroke (Perry et al., 1995). A series of studies also suggested that the elevation of plasma total homocysteine (tHcy) increases the risk of cognitive decline, dementia, and Alzheimer’s disease (AD) (Nurk et al., 2005; Prins et al., 2002; Ravaglia et al., 2005; Seshadri et al., 2002). An association between homocysteine and brain atrophy has also been reported. Elevation of plasma tHcy level was associated with more rapid atrophy of the medical temporal lobe (Clarke et al., 1998), and smaller hippocampus size in AD patients (den Heijer et al., 2003). The rate of brain atrophy in the elderly
* Corresponding author at: Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro Jongno-gu, Seoul, 110-744, Republic of Korea. Tel.: þ82 2 2072 2205; fax: þ82 2 744 7241. E-mail address: [email protected] (D.Y. Lee). 0197-4580/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.neurobiolaging.2014.01.013
individuals with mild cognitive impairment (MCI) was slowed by the treatment with homocysteine lowering B vitamins (Smith et al., 2010). Even in healthy old people, plasma tHcy levels were related to small hippocampal width (Williams et al., 2002), ventricular dilatation (Sachdev et al., 2002), and reduced white matter (WM) volume (Feng et al., 2013). The mechanism of pathogenetic linkage between homocysteine and brain atrophy is, however, still unclear (Sachdev, 2005). Homocysteine may cause brain atrophy through potentiating beta-amyloid protein (Ab) induced neurodegeneration in the pathophysiological process of AD (Hardy and Higgins, 1992; Ho et al., 2001). In addition, given that elevated plasma tHcy is known to be an independent risk factor for vascular disease (Boushey et al., 1995; Seshadri et al., 2008), and brain atrophy is also observed in vascular dementia (Gemmell et al., 2012), homocysteine may influence brain volume through vascular pathology. The possibility that homocysteine has direct neurotoxic effect, not mediated by Ab and vascular pathology, was also strongly suggested by preclinical studies (Kruman et al., 2000; Lipton et al., 1997; Perna et al., 2003).
1520
Y.M. Choe et al. / Neurobiology of Aging 35 (2014) 1519e1525
We aimed to clarify whether plasma tHcy has any independent association, not mediated by cerebral Ab deposition and vascular burden, with whole brain or hippocampal volume in elderly individuals with normal cognition, MCI, and AD. We applied 11 C-labeled Pittsburgh Compound B (11C-PiB) positron emission tomography (PET) imaging (Klunk et al., 2004) to quantify the cerebral Ab deposition. 2. Methods 2.1. Subjects Nineteen MCI and 24 AD patients were recruited from the Dementia and Age-Associated Cognitive Decline Clinic of the Seoul National University Hospital. The AD patients met the criteria for dementia of the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) and the criteria of probable AD of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s disease and Related Disorders Association (NINCDS-ADRDA). All individuals with MCI met the current consensus criteria for amnestic MCI (Petersen, 2004) (1) memory complaint corroborated by an informant; (2) objective memory impairment for age, education, and gender; (3) essentially preserved general cognitive function; (4) largely intact functional activities; and (5) not demented. All amnestic MCI individuals had an overall Clinical Dementia Rating (CDR) (Morris, 1993) of 0.5. In terms of criterion (2), a performance score for at least 1 of the 4 episodic memory tests was 1.5 standard deviation below the respective age-, education-, and gender-specific normative mean (Lee et al., 2004). Those episodic memory tests were included in the Korean version of the Consortium to Establish a Registry for Alzheimer’s disease (CERAD-K) neuropsychological battery (namely: Word List Memory, Word List Recall, and Word List Recognition and Constructional Recall test) (Lee et al., 2002). Fourteen cognitively normal (CN) elderly subjects with an overall CDR of 0 were also selected from a pool of elderly volunteers. All subjects were included after a standardized clinical assessment, as described in the following. The following exclusion criteria were applied to all subjects: (1) any present serious medical, psychiatric, or neurologic disorder that could affect mental function other than MCI or AD; (2) evidence of focal brain lesions on magnetic resonance imaging (MRI) including lacunes and WM hyperintensity lesions of grade 2 or more by Fazeka scale (Fazekas et al., 1987); (3) the presence of severe behavioral or communication problems that would make a clinical or imaging examination difficult; presence of severe renal insufficiency defined as serum creatinine level above 1.5 mg/dL; and (4) the absence of a reliable informant. The Institutional Review Board of Seoul National University Hospital approved the study protocol and written informed consent was obtained from all study subjects or their relatives.
score that was the sum of the factors present ranging from 0 to 6 (DeCarli et al., 2004) and reported as a percentage. 2.3. MRI image acquisition and analysis 2.3.1. MRI acquisition MRI was performed using a whole-body 3T system (Signa VH/i; General Electric, Milwaukee, WI, USA). A dual spin-echo echoplanar imaging sequence was used to acquire diffusion tensor imaging (DTI). MR images with 25 noncollinear diffusion gradients and without diffusion gradient were acquired (repetition time ¼ 10,000 ms, echo time ¼ 77.1 ms, B-factor ¼ 1000 s/mm2, matrix ¼ 128 128, slice thickness/gap ¼ 3.5/0 mm, field of view ¼ 240 mm, slice number ¼ 38). A 3-dimensional T1-weighted spoiled gradient recalled echo sequence was obtained for volumetric tracing and anatomic localization (repetition time ¼ 22.0 ms, echo time ¼ 4.0 ms, slice thickness/gap ¼ 1.4/0 mm, matrix ¼ 256 192, field of view ¼ 240 mm, flip angle ¼ 40 ). Additionally, fluid-attenuated inversion recovery and T2-weighted images were also obtained for qualitative clinical reading. 2.3.1.1. Hippocampal volume. The anatomic boundaries of hippocampus were traced manually on T1-weighted images using Analyze AVW 5.0 (Biomedical Imaging Resource, Mayo Foundation, Rochester, MN, USA) and all traces were drawn blind to diagnosis, sex, or subject demographics. The details of tracing process including the borders of hippocampus were described previously (Choo et al., 2010). To determine the reliability of volumetric measurements, the same rater, unaware of previous readings, repeated volume tracing on 10 randomly selected subjects. Reliability, expressed as intraclass correlation coefficients, was 0.97 for the hippocampus. Mean volume of the left and right hippocampus was used for further analyses. 2.3.1.2. Whole brain and total intracranial volume. Images were processed using the standard Montreal Neurological Institute anatomic pipeline. The native T1 images were normalized into a standardized stereotaxic space using a linear transformation and corrected for intensity nonuniformity (Collins et al., 1994; Sled et al., 1998). Each subject’s brain, which was transformed and corrected, was classified into WM, gray matter, cerebrospinal fluid, and background using a 3-D stereotaxic brain mask and the intensitynormalized stereotaxic environment for classification of tissues (INSECT) algorithm (Zijdenbos et al., 1996). Whole brain volume (WBV) was calculated as sum of WM and gray matter volume, which was inverse-transformed to native space. We calculated the total intracranial volume by measuring the volume of voxels within the brain mask. The brain mask was generated using the brain extraction tool (Smith, 2002). WBV and hippocampal volume was normalized by intracranial volume to correct for possible differences in head size. 11
2.2. Clinical assessment
2.4.
All subjects were examined by neuropsychiatrists with advanced training in dementia research according to the protocol of the Consortium to Establish a Registry for AD (CERAD) (Lee et al., 2002). A panel consisting 4 neuropsychiatrists made clinical decisions including the assignment of CDR. The presence or absence of 6 cerebrovascular risk factors including stroke, diabetes, dyslipidemia, transient ischemic attack, hypertension, and coronary artery disease was systematically assessed from subject and subject’s history provided by an informant as well as review of pertinent medical records. To calculate an overall measure of cerebrovascular burden, we created a composite
11 C-PiB PET was performed using the ECAT EXACT47 scanner (Siemens-CTI, Knoxville, TN, USA), which had an intrinsic resolution of 5.2-mm full width at half maximum. The complete details of 11C-PiB PET acquisition were described previously (Choo et al., 2011). PET quantification image analyses were performed using the fully automated processing as described previously (Choo et al., 2011). Briefly, image preprocessing for statistical analyses was performed using statistical parametric mapping2 (Wellcome Department of Cognitive Neurology, London, United Kingdom) implanted in the Matlab (Mathworks, Natick, MA, USA). 11C-PiB PET data of each subject were co-registered to
C-PiB PET image acquisition and analysis
Y.M. Choe et al. / Neurobiology of Aging 35 (2014) 1519e1525
individual volumetric magnetic resonance image and then automatically spatially normalized into the standard Montreal Neurological Institute template in statistical parametric mapping2 using transformation parameters derived from the normalization of individual magnetic resonance image to the template. All normalized images were reformatted with a voxel size of 2 2 2 mm. For quantitative normalization of cerebral 11 C-PiB uptake values, the cerebellum, which is relatively free of Ab deposition, was used as a reference region (Lopresti et al., 2005) and 11C-PiB retention maps as region-to-cerebellar ratio were generated by dividing regional uptake values by the individual mean cerebellar uptake values in the same images. The automatic anatomic labeling algorithm (Tzourio-Mazoyer et al., 2002) and a region combining method (Reiman et al., 2009) were applied to set region-of-interest (ROI) to characterize 11CPiB retention in the frontal, lateral parietal, posterior cingulateprecuneus (PC-PRC), lateral temporal and occipital, and basal ganglia regions, where prominent 11C-PiB retention was reported (Klunk et al., 2004). A global cortical ROI consisting of frontal, lateral parietal, PC-PRC, lateral temporal, and basal ganglia (BG) ROIs was also defined. For each ROI, mean 11C-PiB value was calculated by averaging 11C-PiB retention values for all voxels within ROI. Global amyloid burden was defined as a mean 11C-PiB retention value of the global cortical ROI. The image was classified as PiB-positive if mean 11C-PiB retention value was over 1.4 in one of the following ROIs: frontal, lateral temporal, lateral parietal, PC-PRC, and BG (Reiman et al., 2009). 2.5. Blood homocysteine, folate, and vitamin B12 level measurement, and apolipoprotein E genotyping Blood samples for measurement of plasma tHcy were collected in evacuated tubes containing ethylenediaminetetraacetic acid after an overnight fast. ethylenediaminetetraacetic acid samples were immediately placed on ice and the plasma was separated by refrigerated centrifugation within 4 hours. Separated plasma was stored for up to 3 days refrigerated at 4 C before being tested. Plasma tHcy concentrations were determined by using chemiluminescent microparticle immunoassay technology (ARCHITECT homocysteine reagent kit, Abbott, IL, USA). The coefficient of variation for plasma tHcy was 16.2 mmol/L, women >13.56 mmol/ L). No subjects had pathologic deficiency of folate (
Contents lists available at ScienceDirect
Neurobiology of Aging journal homepage: www.elsevier.com/locate/neuaging
Association of homocysteine with hippocampal volume independent of cerebral amyloid and vascular burden Young Min Choe a, Bo Kyung Sohn b, Hyo Jung Choi a, Min Soo Byun a, Eun Hyun Seo a, Ji Young Han a, Yu Kyeong Kim c, Eun Jin Yoon c, Jong-Min Lee d, Jinsick Park d, Jong Inn Woo a, Dong Young Lee a, * a
Department of Neuropsychiatry, Seoul National University Hospital, Seoul, Korea Department of Neuropsychiatry, SMG-SNU Boramae Medical Center, Seoul, Korea Department of Nuclear Medicine, SMG-SNU Boramae Medical Center, Seoul, Korea d Department of Biomedical Engineering, Hanyang University, Seoul, Korea b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 April 2013 Received in revised form 20 December 2013 Accepted 12 January 2014 Available online 17 January 2014
This study aimed to clarify whether homocysteine has independent association, not mediated by cerebral beta amyloid protein deposition and vascular burden, with whole brain or hippocampal volume in elderly individuals with normal cognition, mild cognitive impairment, and Alzheimer’s disease. Nineteen mild cognitive impairment and 24 Alzheimer’s disease patients were recruited from the Dementia Clinic of the Seoul National University Hospital. Fourteen cognitively normal elderly subjects were also selected from a pool of elderly volunteers. Multiple linear regression analyses showed that plasma total homocysteine level was significantly associated with hippocampal volume even after controlling the degree of global cerebral beta amyloid deposition and vascular burden as well as other potential confounders including age, gender, education, and apolipoprotein E ε4 genotype. On the contrary, plasma total homocysteine level did not show any significant association with whole brain volume. Our finding of the independent negative association between homocysteine and hippocampal volume suggests that homocysteine has a direct adverse effect, not mediated by cerebral beta amyloid deposition and vascular burden, on the hippocampus. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Homocysteine Hippocampus Amyloid PIB Vascular burden Alzheimer’s disease Mild cognitive impairment
1. Introduction Homocysteine has been proposed to be one of the risk factors for cardiovascular disease (Stampfer et al., 1992), carotid stenosis (Selhub et al., 1995), and stroke (Perry et al., 1995). A series of studies also suggested that the elevation of plasma total homocysteine (tHcy) increases the risk of cognitive decline, dementia, and Alzheimer’s disease (AD) (Nurk et al., 2005; Prins et al., 2002; Ravaglia et al., 2005; Seshadri et al., 2002). An association between homocysteine and brain atrophy has also been reported. Elevation of plasma tHcy level was associated with more rapid atrophy of the medical temporal lobe (Clarke et al., 1998), and smaller hippocampus size in AD patients (den Heijer et al., 2003). The rate of brain atrophy in the elderly
* Corresponding author at: Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro Jongno-gu, Seoul, 110-744, Republic of Korea. Tel.: þ82 2 2072 2205; fax: þ82 2 744 7241. E-mail address: [email protected] (D.Y. Lee). 0197-4580/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.neurobiolaging.2014.01.013
individuals with mild cognitive impairment (MCI) was slowed by the treatment with homocysteine lowering B vitamins (Smith et al., 2010). Even in healthy old people, plasma tHcy levels were related to small hippocampal width (Williams et al., 2002), ventricular dilatation (Sachdev et al., 2002), and reduced white matter (WM) volume (Feng et al., 2013). The mechanism of pathogenetic linkage between homocysteine and brain atrophy is, however, still unclear (Sachdev, 2005). Homocysteine may cause brain atrophy through potentiating beta-amyloid protein (Ab) induced neurodegeneration in the pathophysiological process of AD (Hardy and Higgins, 1992; Ho et al., 2001). In addition, given that elevated plasma tHcy is known to be an independent risk factor for vascular disease (Boushey et al., 1995; Seshadri et al., 2008), and brain atrophy is also observed in vascular dementia (Gemmell et al., 2012), homocysteine may influence brain volume through vascular pathology. The possibility that homocysteine has direct neurotoxic effect, not mediated by Ab and vascular pathology, was also strongly suggested by preclinical studies (Kruman et al., 2000; Lipton et al., 1997; Perna et al., 2003).
1520
Y.M. Choe et al. / Neurobiology of Aging 35 (2014) 1519e1525
We aimed to clarify whether plasma tHcy has any independent association, not mediated by cerebral Ab deposition and vascular burden, with whole brain or hippocampal volume in elderly individuals with normal cognition, MCI, and AD. We applied 11 C-labeled Pittsburgh Compound B (11C-PiB) positron emission tomography (PET) imaging (Klunk et al., 2004) to quantify the cerebral Ab deposition. 2. Methods 2.1. Subjects Nineteen MCI and 24 AD patients were recruited from the Dementia and Age-Associated Cognitive Decline Clinic of the Seoul National University Hospital. The AD patients met the criteria for dementia of the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) and the criteria of probable AD of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s disease and Related Disorders Association (NINCDS-ADRDA). All individuals with MCI met the current consensus criteria for amnestic MCI (Petersen, 2004) (1) memory complaint corroborated by an informant; (2) objective memory impairment for age, education, and gender; (3) essentially preserved general cognitive function; (4) largely intact functional activities; and (5) not demented. All amnestic MCI individuals had an overall Clinical Dementia Rating (CDR) (Morris, 1993) of 0.5. In terms of criterion (2), a performance score for at least 1 of the 4 episodic memory tests was 1.5 standard deviation below the respective age-, education-, and gender-specific normative mean (Lee et al., 2004). Those episodic memory tests were included in the Korean version of the Consortium to Establish a Registry for Alzheimer’s disease (CERAD-K) neuropsychological battery (namely: Word List Memory, Word List Recall, and Word List Recognition and Constructional Recall test) (Lee et al., 2002). Fourteen cognitively normal (CN) elderly subjects with an overall CDR of 0 were also selected from a pool of elderly volunteers. All subjects were included after a standardized clinical assessment, as described in the following. The following exclusion criteria were applied to all subjects: (1) any present serious medical, psychiatric, or neurologic disorder that could affect mental function other than MCI or AD; (2) evidence of focal brain lesions on magnetic resonance imaging (MRI) including lacunes and WM hyperintensity lesions of grade 2 or more by Fazeka scale (Fazekas et al., 1987); (3) the presence of severe behavioral or communication problems that would make a clinical or imaging examination difficult; presence of severe renal insufficiency defined as serum creatinine level above 1.5 mg/dL; and (4) the absence of a reliable informant. The Institutional Review Board of Seoul National University Hospital approved the study protocol and written informed consent was obtained from all study subjects or their relatives.
score that was the sum of the factors present ranging from 0 to 6 (DeCarli et al., 2004) and reported as a percentage. 2.3. MRI image acquisition and analysis 2.3.1. MRI acquisition MRI was performed using a whole-body 3T system (Signa VH/i; General Electric, Milwaukee, WI, USA). A dual spin-echo echoplanar imaging sequence was used to acquire diffusion tensor imaging (DTI). MR images with 25 noncollinear diffusion gradients and without diffusion gradient were acquired (repetition time ¼ 10,000 ms, echo time ¼ 77.1 ms, B-factor ¼ 1000 s/mm2, matrix ¼ 128 128, slice thickness/gap ¼ 3.5/0 mm, field of view ¼ 240 mm, slice number ¼ 38). A 3-dimensional T1-weighted spoiled gradient recalled echo sequence was obtained for volumetric tracing and anatomic localization (repetition time ¼ 22.0 ms, echo time ¼ 4.0 ms, slice thickness/gap ¼ 1.4/0 mm, matrix ¼ 256 192, field of view ¼ 240 mm, flip angle ¼ 40 ). Additionally, fluid-attenuated inversion recovery and T2-weighted images were also obtained for qualitative clinical reading. 2.3.1.1. Hippocampal volume. The anatomic boundaries of hippocampus were traced manually on T1-weighted images using Analyze AVW 5.0 (Biomedical Imaging Resource, Mayo Foundation, Rochester, MN, USA) and all traces were drawn blind to diagnosis, sex, or subject demographics. The details of tracing process including the borders of hippocampus were described previously (Choo et al., 2010). To determine the reliability of volumetric measurements, the same rater, unaware of previous readings, repeated volume tracing on 10 randomly selected subjects. Reliability, expressed as intraclass correlation coefficients, was 0.97 for the hippocampus. Mean volume of the left and right hippocampus was used for further analyses. 2.3.1.2. Whole brain and total intracranial volume. Images were processed using the standard Montreal Neurological Institute anatomic pipeline. The native T1 images were normalized into a standardized stereotaxic space using a linear transformation and corrected for intensity nonuniformity (Collins et al., 1994; Sled et al., 1998). Each subject’s brain, which was transformed and corrected, was classified into WM, gray matter, cerebrospinal fluid, and background using a 3-D stereotaxic brain mask and the intensitynormalized stereotaxic environment for classification of tissues (INSECT) algorithm (Zijdenbos et al., 1996). Whole brain volume (WBV) was calculated as sum of WM and gray matter volume, which was inverse-transformed to native space. We calculated the total intracranial volume by measuring the volume of voxels within the brain mask. The brain mask was generated using the brain extraction tool (Smith, 2002). WBV and hippocampal volume was normalized by intracranial volume to correct for possible differences in head size. 11
2.2. Clinical assessment
2.4.
All subjects were examined by neuropsychiatrists with advanced training in dementia research according to the protocol of the Consortium to Establish a Registry for AD (CERAD) (Lee et al., 2002). A panel consisting 4 neuropsychiatrists made clinical decisions including the assignment of CDR. The presence or absence of 6 cerebrovascular risk factors including stroke, diabetes, dyslipidemia, transient ischemic attack, hypertension, and coronary artery disease was systematically assessed from subject and subject’s history provided by an informant as well as review of pertinent medical records. To calculate an overall measure of cerebrovascular burden, we created a composite
11 C-PiB PET was performed using the ECAT EXACT47 scanner (Siemens-CTI, Knoxville, TN, USA), which had an intrinsic resolution of 5.2-mm full width at half maximum. The complete details of 11C-PiB PET acquisition were described previously (Choo et al., 2011). PET quantification image analyses were performed using the fully automated processing as described previously (Choo et al., 2011). Briefly, image preprocessing for statistical analyses was performed using statistical parametric mapping2 (Wellcome Department of Cognitive Neurology, London, United Kingdom) implanted in the Matlab (Mathworks, Natick, MA, USA). 11C-PiB PET data of each subject were co-registered to
C-PiB PET image acquisition and analysis
Y.M. Choe et al. / Neurobiology of Aging 35 (2014) 1519e1525
individual volumetric magnetic resonance image and then automatically spatially normalized into the standard Montreal Neurological Institute template in statistical parametric mapping2 using transformation parameters derived from the normalization of individual magnetic resonance image to the template. All normalized images were reformatted with a voxel size of 2 2 2 mm. For quantitative normalization of cerebral 11 C-PiB uptake values, the cerebellum, which is relatively free of Ab deposition, was used as a reference region (Lopresti et al., 2005) and 11C-PiB retention maps as region-to-cerebellar ratio were generated by dividing regional uptake values by the individual mean cerebellar uptake values in the same images. The automatic anatomic labeling algorithm (Tzourio-Mazoyer et al., 2002) and a region combining method (Reiman et al., 2009) were applied to set region-of-interest (ROI) to characterize 11CPiB retention in the frontal, lateral parietal, posterior cingulateprecuneus (PC-PRC), lateral temporal and occipital, and basal ganglia regions, where prominent 11C-PiB retention was reported (Klunk et al., 2004). A global cortical ROI consisting of frontal, lateral parietal, PC-PRC, lateral temporal, and basal ganglia (BG) ROIs was also defined. For each ROI, mean 11C-PiB value was calculated by averaging 11C-PiB retention values for all voxels within ROI. Global amyloid burden was defined as a mean 11C-PiB retention value of the global cortical ROI. The image was classified as PiB-positive if mean 11C-PiB retention value was over 1.4 in one of the following ROIs: frontal, lateral temporal, lateral parietal, PC-PRC, and BG (Reiman et al., 2009). 2.5. Blood homocysteine, folate, and vitamin B12 level measurement, and apolipoprotein E genotyping Blood samples for measurement of plasma tHcy were collected in evacuated tubes containing ethylenediaminetetraacetic acid after an overnight fast. ethylenediaminetetraacetic acid samples were immediately placed on ice and the plasma was separated by refrigerated centrifugation within 4 hours. Separated plasma was stored for up to 3 days refrigerated at 4 C before being tested. Plasma tHcy concentrations were determined by using chemiluminescent microparticle immunoassay technology (ARCHITECT homocysteine reagent kit, Abbott, IL, USA). The coefficient of variation for plasma tHcy was 16.2 mmol/L, women >13.56 mmol/ L). No subjects had pathologic deficiency of folate (
