International Journal of Neuroscience, 2014; Early Online: 1–8 Copyright © 2014 Informa Healthcare USA, Inc. ISSN: 0020-7454 print / 1543-5245 online DOI: 10.3109/00207454.2014.971787
ORIGINAL ARTICLE
Cerebrospinal fluid biomarkers for neuropsychological symptoms in early stage of late-onset Alzheimer’s disease
Int J Neurosci Downloaded from informahealthcare.com by University of Alberta on 12/01/14 For personal use only.
Hung-Chou Kuo,1 Hsiu-Chuan Yen,2 Chin-Chang Huang,1 Wen-Chuin Hsu,1 Hsing-Ju Wei,2 and Chih-Lung Lin3 1
Department of Neurology, Chang Gung Memorial Hospital and Chang Gung University College of Medicine, Taipei, Taiwan; 2 Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang Gung University, Taoyuan, Taiwan and 3 Department of Neurosurgery, Chang Gung Memorial Hospital, Taoyuan, Taiwan Purpose: In addition to testing blood, cerebrospinal fluid (CSF) has been analyzed in the search for biomarkers. The aim of this study was to identify biomarkers in CSF for neuropsychological symptoms in early-stage lateonset Alzheimer’s disease (LOAD). Methods: CSF levels of beta-amyloid 1-42 (Aβ42), F2 -isoprostanes (F2 -IsoPs) and F4 -neuroprostanes (F4 -NPs) were assayed in nine patients with mild Alzheimer’s disease (AD), nine patients with amnestic mild cognitive impairment (a-MCI) and nine individuals with normal mental function. The three groups underwent neuropsychological testing. Results: CSF levels of F2 -IsoPs and F4 -NPs did not significantly differ among the three groups. Aβ42 in CSF was significantly higher in the control group compared with the mild AD group (p < 0.001) and a-MCI group (p = 0.03). There was a significant positive correlation between the level of F2 -IsoPs and Aβ42 in the a-MCI group and between the level of F2 -IsoPs and F4 -NPs in the mild AD group. In comparisons between the mild AD group and a-MCI group combined, the cognitive impairment (CI) group, with the control group, the median levels of F2-IsoPs and F4 -NPs were significantly higher in the CI group and median level of Aβ42 was significantly lower in the CI group. Both the levels of F2 -IsoPs and Aβ42 were significantly negatively correlated with paranoid and delusional ideation and total score for the Behavioral Pathology in Alzheimer’s Disease Scale (BEHAVE-AD). Conclusions: The findings suggest CSF levels of Aβ42 and F2 -IsoPs are associated with the severity of neuropsychological symptoms. KEYWORDS: Alzheimer’s disease, neuropsychological symptom, cerebrospinal fluid biomarkers, neuroprostanes, isoprostanes, Aβ42
Introduction Patients with Alzheimer’s disease (AD) have brain lesions characterized by plaques containing beta-amyloid [Aβ], neurofibrillary tangles formed by microtubuleassociated tau protein, and loss of synapses and neurons [1–3]. It has been observed that formation of plaques and neurofibrillary tangles begins 10 to 15 years before there is any indication of dementia [4]. Received 26 December 2013; revised 9 September 2014; accepted 29 September 2014. Correspondence: Hung-Chou Kuo, MD, Department of Neurology, Chang Gung Memorial Hospital and Chang Gung University, 199, Tung Hwa North Road, Taipei, Taiwan. Tel: +886-3-3281200 ext. 8410, Fax: +886-3-3287226. E-mail: [email protected]
Biomarkers have been sought for diagnosing AD. In addition to testing blood, cerebrospinal fluid (CSF) has been analyzed in the search for biomarkers. Useful biomarkers in the CSF for diagnosing AD include Aβ42, total tau (tau), and phosphorylated tau (p-tau), which is tau phosphorylated at threonine 181 [3,4]. These biomarkers are representative of AD neuropathology [3]. There is increasing evidence that the brains of AD patients have oxidative damage [5]. Various types of studies indicate that lipid peroxidation is involved in the pathogenesis of neuron degeneration in such patients [6]. Lipid peroxidation occurs when lipids are attacked by free radicals and is self-propagating; it is ended by antioxidants [6]. Lipid peroxidation produces structural damage to membranes [5]. Also, highly reactive
1
Int J Neurosci Downloaded from informahealthcare.com by University of Alberta on 12/01/14 For personal use only.
2
H.-C. Kuo et al.
secondary aldehyde products can be generated and may cause damage to cells because they can covalently modify important biomolecules [6]. F2 -isoprostanes, which are prostaglandin-like molecules that are synthesized via nonenzymatic peroxidation of arachidonic acid in a reaction catalyzed by free radicals, might be useful as biomarkers for AD [3,6,7]. The findings of human and transgenic animal studies have suggested that oxidative stress is an early event in the clinical course and pathogenesis of AD [8]. Besides, increased levels of isoprostane and prostaglandin in hippocampus of postmortem AD patients have suggested that free radical damage and inflammation process are prominent in the brains of AD patients [9]. Roberts et al. found that oxidation of docosahexaenoic acid resulted in the formation of compounds that were like F2 -isoprostane, which they named F4 neuroprostanes [10]. They found that CSF levels of F4 -neuroprostanes were significantly higher in AD patients compared with age-matched controls and therefore proposed that these compounds could be a unique marker for oxidative brain injury and clinical criteria for this diagnosis were suggested. The aim of this study was to compare CSF levels of Aβ42, F2 -isoprostanes (F2 -IsoPs), and F4 -neuroprostanes (F4 -NPs) in patients with mild stage of late-onset Alzheimer’s disease (LOAD), patients with mild amnestic cognitive impairment (a-MCI), the transitional state from normal cognition to mild AD, and controls to identify CSF biomarkers for neuropsychological symptoms in the early stage of LOAD.
Methods Subjects Participants were recruited from patients in the surrounding community who visited our hospital. Because a strong reluctance to undergo lumbar puncture is part of the culture in Taiwan, the study was limited in sample size. All participants underwent a standard neurological examination to rule out other neurological diseases, major psychiatric illness and severe visual or hearing impairment. After biochemistry testing to screen the underlying causes of cognitive impairment, all patients underwent brain imaging to rule out the possibility of other organic brain lesions. This study included nine patients with LOAD who had mild AD, nine patients with a-MCI, and nine controls (age > 60 years) with polyneuropathy who had intact mental status with a normal judgment, orientation, memory, abstract thinking and calculation (JOMAC) scale score on neurological examination and no significant
brain abnormalities detected on brain computed tomography (CT) or magnetic resonance imaging (MRI). The controls needed a CSF study to exclude chronic inflammatory demyelination polyneuropathy, paraprotein neuropathy and Gullain–Barr´e syndrome. Written informed consent was obtained by all participants and their spouse, next of kin or guardians in accordance with protocol approved by Chang Gung Memorial Hospital (IRB 98-0137B, 99-0014C, 99-3954C and 1001400C). The ethics committees of Chang Gung Memorial Hospital approved this consent procedure and the study. The dementia patients underwent standard neurological evaluation, including neuropsychological testing. After laboratory testing and brain imaging studies, patients were diagnosed with probable AD according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) and National Institute of Neurological and Communicative Disorders and Stroke/Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) criteria. Disease severity was graded by using the Clinical Dementia Rating (CDR) scale and Mini-Mental Status Examination (MMSE) [11,12]. A diagnosis of mild AD was determined by a score of 1 on the CDR scale. A diagnosis of a-MCI was defined by subjective memory impairment and having significant deficits in memory (i.e., < 1.5 standard deviations [SD] below the age and education control mean on the tests of delayed recall) while other neuropsychological functions remained within normal parameters (within 0.5 SD of appropriate controls) [13–15]. A diagnosis of a-MCI also required that the patient showed delayed recall, retention and recognition on the Word List Memory subtest of more than 1.5 SD below the age and education adjusted norm while also scoring within 0.5 SD of age and education norms on the MMSE and Boston Naming Test of Consortium to Establish Registry for Alzheimer’s Disease (CERAD) [16]. Other neuropsychological studies included Instrumental Activities of Daily Living (IADL) test, Behavioral Pathology in Alzheimer’s Disease Scale (BEHAVE–AD) [16], Cognitive Abilities Screening Instrument (CASI), Alzheimer’s Disease Assessment Scale Cognitive Subscale (ADAS-Cog) and Visual Association Memory Test (VAMT) [17]. The BEHAVE–AD, which includes assessment of the symptoms of paranoid and delusional ideation, hallucinations, diurnal rhythm disturbance, and anxieties and phobias, is used to evaluate the behavior and psychological status of AD patients. The VAMT includes immediate recall and second immediate recall, which represents remembering the event after several seconds and several minutes. These tests were performed by a well-trained neuropsychologist.
International Journal of Neuroscience
Biomarkers in late-onset Alzheimer’s disease
Collection of specimens All patients and controls were admitted to our hospital for CSF study and lumbar puncture was performed in the morning. Cerebrospinal fluid was collected using standard procedures. Collected CSF specimens are centrifuged at 3000 rpm for 10 min. The resulting supernatants are divided into aliquots and then transferred to −70 ◦ C freezers.
Int J Neurosci Downloaded from informahealthcare.com by University of Alberta on 12/01/14 For personal use only.
CSF assays CSF Aβ42 The prevailing method for detection of Aβ42 is sandwich ELISA. We used this assay (INNOTEST Aβ142; Innogenetics, Gent, Belgium) with monoclonal antibody 21F12 specific for the free C-end of Aβ42 as the capturing antibody. In addition, monoclonal antibody 3D6, which is specific for the N-end of Aβ42, was used as the detector. After washing five times at room temperature, horseradish peroxidase-labeled streptavidin and 3,5,3 ,5 -tetramethylbenzidine were added. Absorbance was read at 450 nm on a microplate reader. Analysis of Free F2 -IsoPs and F4 -NPs in CSF The procedures of sample purification by solid phase extraction and thin-layer chromatography, sample derivatization and equipment settings for detection of free F4 NPs in human CSF by gas chromatograph/negative-ion chemical-ionization mass spectrometry (GC/NICI-MS) were carried out according to the previously described Table 1.
3
methods [18]. The 6890 GC/5975 MS and DB-1701 capillary columns from Agilent were used. Because the amount of CSF samples was very limited, we analyzed both F2 -IsoPs and F4 -NPs from the same samples processed for F4 -NPs analysis. The structure and molecular weight of the detected ions for trimethylsilyl-derivatized F2 -IsoPs, F4 -NPs and [2 H4 ]-15-F2t -IsoP (the internal standard) have been previously illustrated [18]. F4 -NPs concentrations in CSF samples were obtained by multiplying the concentration of the internal standard with the ratio of integrated area of the peaks of F4 -NPs (m/z 593.5), which was determined by the peak range of docosahexaenoic acid oxidized in vitro, to the area of the [2 H4 ]-15-F2t -IsoP peak (m/z 573.4) [17]. F2 -IsoPs concentrations in CSF samples were obtained by multiplying the concentration of the internal standard with the ratio of the peak height of the F2 -IsoP peak (m/z 569.4) to the peak height of the [2 H4 ]-15-F2t -IsoP peak (m/z 573.4) [19].
Statistical analysis Data were presented as median with interquartile range (IQR) in Table 1 and the box plot in Figure 1. Comparisons between the mild AD and a-MCI groups were performed with the Mann–Whitney U-test. In addition, Kruskal–Wallis tests were used to determine the difference among the control, mild AD and a-MCI groups. When a significant difference between groups was apparent, multiple comparisons were performed using the
Characteristics of the controls, mild AD patients and a-MCI patients.
Age Male Education MMSE CDR VAMT Immediate recall 2nd Immediate recall BEHAVE-AD Paranoid and delusional ideation (21) Hallucinations (15) Activity disturbances (9) Aggression (9) Diurnal rhythm disturbances (3) Affective disturbances (6) Anxieties and phobias (12) Total (75)
Control (n = 9)
Mild AD (n = 9)
a-MCI (n = 9)
p-Value
69 (65.5–74.5) 6 6 (6–7) 29 (28.3–30) 0 (0–0)
77 (64–79) 5 6 (6–9) 18 (14.5–19)∗ 1 (1–1)∗
73 (70.5–79) 3 9 (6–10.5) 20 (18.5–23)∗ † 0.5 (0.5–0.5)∗ †
0.205 0.885 0.357 < 0.001 < 0.001
– –
1 (0–1.8) 1 (0.3–3)
– – – – – – – –
1 (0–2) 0 (0–0.8) 0 (0–0) 0 (0–1.8) 0 (0–0) 0 (0–0.8) 1 (0–1.8) 4 (0.5–7.8)
1.5 (0–3.3) 2 (0–5.3)
0.635 0.642
0 (0–1) 0 (0 − 0) 0 (0–0) 0 (0–1.5) 0 (0–0.8) 0 (0–0.8) 0 (0–0.8) 1 (0–5)
0.449 0.144 1.000 0.655 0.643 1.000 0.199 0.310
a-MCI: amnestic MCI; mild AD: mild Alzheimer’s disease, CDR: Clinical Dementia Rating scale, MMSE: Mini-Mental Status Examination, VAMT: Visual Association Memory Test, BEHAVE-AD: Behavioral Pathology in Alzheimer’s Disease. The score of each corresponding part was included in the parenthesis. ∗ p < 0.05 significantly different compared with control group. † p < 0.05 significantly different compared with mild AD group. C
2014 Informa Healthcare USA, Inc.
4
H.-C. Kuo et al.
man’s correlation. Data were analyzed using SPSS 15.0 statistics software (SPSS Inc., Chicago, IL). All statistical assessments were two-sided and evaluated at the 0.05 level of significant difference.
Int J Neurosci Downloaded from informahealthcare.com by University of Alberta on 12/01/14 For personal use only.
Results
Figure 1. Cerebrospinal fluid level of biomarkers. (A) F2 -
isoprostanes, (B) F4 -neuroprostanes, and (C) β-amyloid (142).
Bonferroni procedure with Type-I error adjustment. Comparisons between the cognitive impairment group (mild AD and a-MCI groups combined) and control group were performed using the Mann–Whitney U-test. The correlation between biomarker and severity of neuropsychological symptoms was evaluated using Spear-
The demographic characteristics of all three groups and the neuropsychological characteristics of the mild AD and a-MCI groups are listed in Table 1. The median MMSE score in the mild AD group was 18 and 20 in the a-MCI group, which were significantly lower than the score of 29 in the control group (p ≤ 0.03). The median CDR score in the mild AD group was 1 and in the a-MCI group 0.5, which were significantly higher compared with the score of 0 in the control group (p < 0.001). The CSF levels of F2 -IsoPs, F4 -NPs and Aβ42 are shown in Figure 1A–C. The median level of F2 -IsoPs was 14.3 pg/mL in the control group, 22.6 pg/mL in the mild AD group, and 21.8 pg/mL in the a-MCI group. There was no statistically significant difference in the level of F2 -IsoPs among the three groups (p = 0.057). The median CSF level of F4 -NPs was 35.7, 49.9 and 38.6 pg/mL in the control, mild AD and a-MCI groups, respectively. There was no significant difference in the level of F4 -NPs among the three groups (p = 0.057). The median CSF level of Aβ42 was 109.4 pg/mL in the control group, 46.2 pg/mL in the mild AD group and 55.5 pg/mL in the a-MCI group. The Aβ42 level in the control group was significantly higher compared with the level in the mild AD group (p < 0.001) and a-MCI group (p = 0.031). There was a significant positive correlation between the level of F2 -IsoPs and Aβ42 in the a-MCI group as well as a significant positive correlation between the level of F2 -IsoPs and F4 -NPs in the mild AD group. Also, there were significant positive correlations between the level of F2 -IsoPs and F4 -NPs in all groups combined. Table 2 presents the results of comparisons between the CI group and control group with regard to median levels of biomarkers and scores on the MMSE and CDR scale. The median levels of F2-IsoPs, F4-NPs and CDR scale scores were significantly higher in the CI group compared with the control group (F2-IsoPs: 22.2 vs. 14.3 pg/mL, p = 0.018; F4-NPs: 47.9 vs. 35.7 pg/mL, p = 0.018; CDR: 0.8 vs. 0, p < 0.001). The median level of β-amyloid and median MMSE score were significantly lower in the CI group compared with the control group (β-amyloid: 50.2 vs. 109.4, p < 0.001; MMSE: 19 vs. 29, p < 0.001). Table 3 shows the relationships between biomarkers F2 -IsoPs, F4 -NPs and Aβ42 and clinical severity according to the results on the MMSE, CDR scale, VAMT International Journal of Neuroscience
Biomarkers in late-onset Alzheimer’s disease Table 2.
We found that Aβ42 levels were significantly lower in the mild AD and a-MCI patients compared with the controls and that the CI and control groups significantly differed in levels of all three biomarkers and MMSE and CDR scale scores. We also found that levels of Aβ42 and F2 -IsoPs were significantly negatively correlated with paranoid and delusional ideation and total BEHAVEAD score, Aβ42 was significantly negatively correlated with aggression, and F2 -IsoPs were significantly negatively correlated with affective disturbances. We measured CSF biomarkers in patients with sporadic LOAD. Liu et al. compared CSF biomarker levels between elderly subjects who were cognitively normal but had a family history of LOAD and elderly subjects who were cognitively normal and did not have a family history of LOAD [4]. They found that the level of Aβ42 and the Aβ42/40 ratio in subjects with a maternal history of LOAD were lower than in subjects with a paternal history of LOAD or no family history of LOAD. There were no differences between the groups in tau or p-tau levels. The results suggested that individuals with a family history of LOAD on the maternal side may be at increased risk for developing AD. The authors speculated that the lower level of Aβ42 in CSF might be a reflection of early Aβ deposition in the brain. In a study by van Harten et al. CSF levels of Aβ42, tau and p-tau were measured in non-demented patients who were seen at a memory clinic because of subjective complaints [20]. Among 127 patients, 11 progressed to MCI and 2 progressed to AD. It was found that Aβ42 was the strongest predictor of progression, which the authors noted was consistent with the hypothesis that the pathologic changes in AD begin with Aβ42 deposition while in contrast neuronal degeneration and tau hyperphosphorylation occur closer to the time that AD becomes clinically manifested. Our results showing that Aβ42 levels in the CSF in LOAD patients and a-MCI
Comparisons between the controls and CI patients. Controls (n = 9)
CI patients (n = 18)
pValue
F2-IsoPs (pg/mL) 14.3 (12.7–18.1) 22.2 (15.2–26.2) 0.018 F4-NPs (pg/mL) 35.7 (30.6–46.3) 47.9 (37.1–57.2) 0.018 β-amyloid (1-42) 109.4 (102.8–140.8) 50.2 (37.0–94.8)
ORIGINAL ARTICLE
Cerebrospinal fluid biomarkers for neuropsychological symptoms in early stage of late-onset Alzheimer’s disease
Int J Neurosci Downloaded from informahealthcare.com by University of Alberta on 12/01/14 For personal use only.
Hung-Chou Kuo,1 Hsiu-Chuan Yen,2 Chin-Chang Huang,1 Wen-Chuin Hsu,1 Hsing-Ju Wei,2 and Chih-Lung Lin3 1
Department of Neurology, Chang Gung Memorial Hospital and Chang Gung University College of Medicine, Taipei, Taiwan; 2 Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang Gung University, Taoyuan, Taiwan and 3 Department of Neurosurgery, Chang Gung Memorial Hospital, Taoyuan, Taiwan Purpose: In addition to testing blood, cerebrospinal fluid (CSF) has been analyzed in the search for biomarkers. The aim of this study was to identify biomarkers in CSF for neuropsychological symptoms in early-stage lateonset Alzheimer’s disease (LOAD). Methods: CSF levels of beta-amyloid 1-42 (Aβ42), F2 -isoprostanes (F2 -IsoPs) and F4 -neuroprostanes (F4 -NPs) were assayed in nine patients with mild Alzheimer’s disease (AD), nine patients with amnestic mild cognitive impairment (a-MCI) and nine individuals with normal mental function. The three groups underwent neuropsychological testing. Results: CSF levels of F2 -IsoPs and F4 -NPs did not significantly differ among the three groups. Aβ42 in CSF was significantly higher in the control group compared with the mild AD group (p < 0.001) and a-MCI group (p = 0.03). There was a significant positive correlation between the level of F2 -IsoPs and Aβ42 in the a-MCI group and between the level of F2 -IsoPs and F4 -NPs in the mild AD group. In comparisons between the mild AD group and a-MCI group combined, the cognitive impairment (CI) group, with the control group, the median levels of F2-IsoPs and F4 -NPs were significantly higher in the CI group and median level of Aβ42 was significantly lower in the CI group. Both the levels of F2 -IsoPs and Aβ42 were significantly negatively correlated with paranoid and delusional ideation and total score for the Behavioral Pathology in Alzheimer’s Disease Scale (BEHAVE-AD). Conclusions: The findings suggest CSF levels of Aβ42 and F2 -IsoPs are associated with the severity of neuropsychological symptoms. KEYWORDS: Alzheimer’s disease, neuropsychological symptom, cerebrospinal fluid biomarkers, neuroprostanes, isoprostanes, Aβ42
Introduction Patients with Alzheimer’s disease (AD) have brain lesions characterized by plaques containing beta-amyloid [Aβ], neurofibrillary tangles formed by microtubuleassociated tau protein, and loss of synapses and neurons [1–3]. It has been observed that formation of plaques and neurofibrillary tangles begins 10 to 15 years before there is any indication of dementia [4]. Received 26 December 2013; revised 9 September 2014; accepted 29 September 2014. Correspondence: Hung-Chou Kuo, MD, Department of Neurology, Chang Gung Memorial Hospital and Chang Gung University, 199, Tung Hwa North Road, Taipei, Taiwan. Tel: +886-3-3281200 ext. 8410, Fax: +886-3-3287226. E-mail: [email protected]
Biomarkers have been sought for diagnosing AD. In addition to testing blood, cerebrospinal fluid (CSF) has been analyzed in the search for biomarkers. Useful biomarkers in the CSF for diagnosing AD include Aβ42, total tau (tau), and phosphorylated tau (p-tau), which is tau phosphorylated at threonine 181 [3,4]. These biomarkers are representative of AD neuropathology [3]. There is increasing evidence that the brains of AD patients have oxidative damage [5]. Various types of studies indicate that lipid peroxidation is involved in the pathogenesis of neuron degeneration in such patients [6]. Lipid peroxidation occurs when lipids are attacked by free radicals and is self-propagating; it is ended by antioxidants [6]. Lipid peroxidation produces structural damage to membranes [5]. Also, highly reactive
1
Int J Neurosci Downloaded from informahealthcare.com by University of Alberta on 12/01/14 For personal use only.
2
H.-C. Kuo et al.
secondary aldehyde products can be generated and may cause damage to cells because they can covalently modify important biomolecules [6]. F2 -isoprostanes, which are prostaglandin-like molecules that are synthesized via nonenzymatic peroxidation of arachidonic acid in a reaction catalyzed by free radicals, might be useful as biomarkers for AD [3,6,7]. The findings of human and transgenic animal studies have suggested that oxidative stress is an early event in the clinical course and pathogenesis of AD [8]. Besides, increased levels of isoprostane and prostaglandin in hippocampus of postmortem AD patients have suggested that free radical damage and inflammation process are prominent in the brains of AD patients [9]. Roberts et al. found that oxidation of docosahexaenoic acid resulted in the formation of compounds that were like F2 -isoprostane, which they named F4 neuroprostanes [10]. They found that CSF levels of F4 -neuroprostanes were significantly higher in AD patients compared with age-matched controls and therefore proposed that these compounds could be a unique marker for oxidative brain injury and clinical criteria for this diagnosis were suggested. The aim of this study was to compare CSF levels of Aβ42, F2 -isoprostanes (F2 -IsoPs), and F4 -neuroprostanes (F4 -NPs) in patients with mild stage of late-onset Alzheimer’s disease (LOAD), patients with mild amnestic cognitive impairment (a-MCI), the transitional state from normal cognition to mild AD, and controls to identify CSF biomarkers for neuropsychological symptoms in the early stage of LOAD.
Methods Subjects Participants were recruited from patients in the surrounding community who visited our hospital. Because a strong reluctance to undergo lumbar puncture is part of the culture in Taiwan, the study was limited in sample size. All participants underwent a standard neurological examination to rule out other neurological diseases, major psychiatric illness and severe visual or hearing impairment. After biochemistry testing to screen the underlying causes of cognitive impairment, all patients underwent brain imaging to rule out the possibility of other organic brain lesions. This study included nine patients with LOAD who had mild AD, nine patients with a-MCI, and nine controls (age > 60 years) with polyneuropathy who had intact mental status with a normal judgment, orientation, memory, abstract thinking and calculation (JOMAC) scale score on neurological examination and no significant
brain abnormalities detected on brain computed tomography (CT) or magnetic resonance imaging (MRI). The controls needed a CSF study to exclude chronic inflammatory demyelination polyneuropathy, paraprotein neuropathy and Gullain–Barr´e syndrome. Written informed consent was obtained by all participants and their spouse, next of kin or guardians in accordance with protocol approved by Chang Gung Memorial Hospital (IRB 98-0137B, 99-0014C, 99-3954C and 1001400C). The ethics committees of Chang Gung Memorial Hospital approved this consent procedure and the study. The dementia patients underwent standard neurological evaluation, including neuropsychological testing. After laboratory testing and brain imaging studies, patients were diagnosed with probable AD according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) and National Institute of Neurological and Communicative Disorders and Stroke/Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) criteria. Disease severity was graded by using the Clinical Dementia Rating (CDR) scale and Mini-Mental Status Examination (MMSE) [11,12]. A diagnosis of mild AD was determined by a score of 1 on the CDR scale. A diagnosis of a-MCI was defined by subjective memory impairment and having significant deficits in memory (i.e., < 1.5 standard deviations [SD] below the age and education control mean on the tests of delayed recall) while other neuropsychological functions remained within normal parameters (within 0.5 SD of appropriate controls) [13–15]. A diagnosis of a-MCI also required that the patient showed delayed recall, retention and recognition on the Word List Memory subtest of more than 1.5 SD below the age and education adjusted norm while also scoring within 0.5 SD of age and education norms on the MMSE and Boston Naming Test of Consortium to Establish Registry for Alzheimer’s Disease (CERAD) [16]. Other neuropsychological studies included Instrumental Activities of Daily Living (IADL) test, Behavioral Pathology in Alzheimer’s Disease Scale (BEHAVE–AD) [16], Cognitive Abilities Screening Instrument (CASI), Alzheimer’s Disease Assessment Scale Cognitive Subscale (ADAS-Cog) and Visual Association Memory Test (VAMT) [17]. The BEHAVE–AD, which includes assessment of the symptoms of paranoid and delusional ideation, hallucinations, diurnal rhythm disturbance, and anxieties and phobias, is used to evaluate the behavior and psychological status of AD patients. The VAMT includes immediate recall and second immediate recall, which represents remembering the event after several seconds and several minutes. These tests were performed by a well-trained neuropsychologist.
International Journal of Neuroscience
Biomarkers in late-onset Alzheimer’s disease
Collection of specimens All patients and controls were admitted to our hospital for CSF study and lumbar puncture was performed in the morning. Cerebrospinal fluid was collected using standard procedures. Collected CSF specimens are centrifuged at 3000 rpm for 10 min. The resulting supernatants are divided into aliquots and then transferred to −70 ◦ C freezers.
Int J Neurosci Downloaded from informahealthcare.com by University of Alberta on 12/01/14 For personal use only.
CSF assays CSF Aβ42 The prevailing method for detection of Aβ42 is sandwich ELISA. We used this assay (INNOTEST Aβ142; Innogenetics, Gent, Belgium) with monoclonal antibody 21F12 specific for the free C-end of Aβ42 as the capturing antibody. In addition, monoclonal antibody 3D6, which is specific for the N-end of Aβ42, was used as the detector. After washing five times at room temperature, horseradish peroxidase-labeled streptavidin and 3,5,3 ,5 -tetramethylbenzidine were added. Absorbance was read at 450 nm on a microplate reader. Analysis of Free F2 -IsoPs and F4 -NPs in CSF The procedures of sample purification by solid phase extraction and thin-layer chromatography, sample derivatization and equipment settings for detection of free F4 NPs in human CSF by gas chromatograph/negative-ion chemical-ionization mass spectrometry (GC/NICI-MS) were carried out according to the previously described Table 1.
3
methods [18]. The 6890 GC/5975 MS and DB-1701 capillary columns from Agilent were used. Because the amount of CSF samples was very limited, we analyzed both F2 -IsoPs and F4 -NPs from the same samples processed for F4 -NPs analysis. The structure and molecular weight of the detected ions for trimethylsilyl-derivatized F2 -IsoPs, F4 -NPs and [2 H4 ]-15-F2t -IsoP (the internal standard) have been previously illustrated [18]. F4 -NPs concentrations in CSF samples were obtained by multiplying the concentration of the internal standard with the ratio of integrated area of the peaks of F4 -NPs (m/z 593.5), which was determined by the peak range of docosahexaenoic acid oxidized in vitro, to the area of the [2 H4 ]-15-F2t -IsoP peak (m/z 573.4) [17]. F2 -IsoPs concentrations in CSF samples were obtained by multiplying the concentration of the internal standard with the ratio of the peak height of the F2 -IsoP peak (m/z 569.4) to the peak height of the [2 H4 ]-15-F2t -IsoP peak (m/z 573.4) [19].
Statistical analysis Data were presented as median with interquartile range (IQR) in Table 1 and the box plot in Figure 1. Comparisons between the mild AD and a-MCI groups were performed with the Mann–Whitney U-test. In addition, Kruskal–Wallis tests were used to determine the difference among the control, mild AD and a-MCI groups. When a significant difference between groups was apparent, multiple comparisons were performed using the
Characteristics of the controls, mild AD patients and a-MCI patients.
Age Male Education MMSE CDR VAMT Immediate recall 2nd Immediate recall BEHAVE-AD Paranoid and delusional ideation (21) Hallucinations (15) Activity disturbances (9) Aggression (9) Diurnal rhythm disturbances (3) Affective disturbances (6) Anxieties and phobias (12) Total (75)
Control (n = 9)
Mild AD (n = 9)
a-MCI (n = 9)
p-Value
69 (65.5–74.5) 6 6 (6–7) 29 (28.3–30) 0 (0–0)
77 (64–79) 5 6 (6–9) 18 (14.5–19)∗ 1 (1–1)∗
73 (70.5–79) 3 9 (6–10.5) 20 (18.5–23)∗ † 0.5 (0.5–0.5)∗ †
0.205 0.885 0.357 < 0.001 < 0.001
– –
1 (0–1.8) 1 (0.3–3)
– – – – – – – –
1 (0–2) 0 (0–0.8) 0 (0–0) 0 (0–1.8) 0 (0–0) 0 (0–0.8) 1 (0–1.8) 4 (0.5–7.8)
1.5 (0–3.3) 2 (0–5.3)
0.635 0.642
0 (0–1) 0 (0 − 0) 0 (0–0) 0 (0–1.5) 0 (0–0.8) 0 (0–0.8) 0 (0–0.8) 1 (0–5)
0.449 0.144 1.000 0.655 0.643 1.000 0.199 0.310
a-MCI: amnestic MCI; mild AD: mild Alzheimer’s disease, CDR: Clinical Dementia Rating scale, MMSE: Mini-Mental Status Examination, VAMT: Visual Association Memory Test, BEHAVE-AD: Behavioral Pathology in Alzheimer’s Disease. The score of each corresponding part was included in the parenthesis. ∗ p < 0.05 significantly different compared with control group. † p < 0.05 significantly different compared with mild AD group. C
2014 Informa Healthcare USA, Inc.
4
H.-C. Kuo et al.
man’s correlation. Data were analyzed using SPSS 15.0 statistics software (SPSS Inc., Chicago, IL). All statistical assessments were two-sided and evaluated at the 0.05 level of significant difference.
Int J Neurosci Downloaded from informahealthcare.com by University of Alberta on 12/01/14 For personal use only.
Results
Figure 1. Cerebrospinal fluid level of biomarkers. (A) F2 -
isoprostanes, (B) F4 -neuroprostanes, and (C) β-amyloid (142).
Bonferroni procedure with Type-I error adjustment. Comparisons between the cognitive impairment group (mild AD and a-MCI groups combined) and control group were performed using the Mann–Whitney U-test. The correlation between biomarker and severity of neuropsychological symptoms was evaluated using Spear-
The demographic characteristics of all three groups and the neuropsychological characteristics of the mild AD and a-MCI groups are listed in Table 1. The median MMSE score in the mild AD group was 18 and 20 in the a-MCI group, which were significantly lower than the score of 29 in the control group (p ≤ 0.03). The median CDR score in the mild AD group was 1 and in the a-MCI group 0.5, which were significantly higher compared with the score of 0 in the control group (p < 0.001). The CSF levels of F2 -IsoPs, F4 -NPs and Aβ42 are shown in Figure 1A–C. The median level of F2 -IsoPs was 14.3 pg/mL in the control group, 22.6 pg/mL in the mild AD group, and 21.8 pg/mL in the a-MCI group. There was no statistically significant difference in the level of F2 -IsoPs among the three groups (p = 0.057). The median CSF level of F4 -NPs was 35.7, 49.9 and 38.6 pg/mL in the control, mild AD and a-MCI groups, respectively. There was no significant difference in the level of F4 -NPs among the three groups (p = 0.057). The median CSF level of Aβ42 was 109.4 pg/mL in the control group, 46.2 pg/mL in the mild AD group and 55.5 pg/mL in the a-MCI group. The Aβ42 level in the control group was significantly higher compared with the level in the mild AD group (p < 0.001) and a-MCI group (p = 0.031). There was a significant positive correlation between the level of F2 -IsoPs and Aβ42 in the a-MCI group as well as a significant positive correlation between the level of F2 -IsoPs and F4 -NPs in the mild AD group. Also, there were significant positive correlations between the level of F2 -IsoPs and F4 -NPs in all groups combined. Table 2 presents the results of comparisons between the CI group and control group with regard to median levels of biomarkers and scores on the MMSE and CDR scale. The median levels of F2-IsoPs, F4-NPs and CDR scale scores were significantly higher in the CI group compared with the control group (F2-IsoPs: 22.2 vs. 14.3 pg/mL, p = 0.018; F4-NPs: 47.9 vs. 35.7 pg/mL, p = 0.018; CDR: 0.8 vs. 0, p < 0.001). The median level of β-amyloid and median MMSE score were significantly lower in the CI group compared with the control group (β-amyloid: 50.2 vs. 109.4, p < 0.001; MMSE: 19 vs. 29, p < 0.001). Table 3 shows the relationships between biomarkers F2 -IsoPs, F4 -NPs and Aβ42 and clinical severity according to the results on the MMSE, CDR scale, VAMT International Journal of Neuroscience
Biomarkers in late-onset Alzheimer’s disease Table 2.
We found that Aβ42 levels were significantly lower in the mild AD and a-MCI patients compared with the controls and that the CI and control groups significantly differed in levels of all three biomarkers and MMSE and CDR scale scores. We also found that levels of Aβ42 and F2 -IsoPs were significantly negatively correlated with paranoid and delusional ideation and total BEHAVEAD score, Aβ42 was significantly negatively correlated with aggression, and F2 -IsoPs were significantly negatively correlated with affective disturbances. We measured CSF biomarkers in patients with sporadic LOAD. Liu et al. compared CSF biomarker levels between elderly subjects who were cognitively normal but had a family history of LOAD and elderly subjects who were cognitively normal and did not have a family history of LOAD [4]. They found that the level of Aβ42 and the Aβ42/40 ratio in subjects with a maternal history of LOAD were lower than in subjects with a paternal history of LOAD or no family history of LOAD. There were no differences between the groups in tau or p-tau levels. The results suggested that individuals with a family history of LOAD on the maternal side may be at increased risk for developing AD. The authors speculated that the lower level of Aβ42 in CSF might be a reflection of early Aβ deposition in the brain. In a study by van Harten et al. CSF levels of Aβ42, tau and p-tau were measured in non-demented patients who were seen at a memory clinic because of subjective complaints [20]. Among 127 patients, 11 progressed to MCI and 2 progressed to AD. It was found that Aβ42 was the strongest predictor of progression, which the authors noted was consistent with the hypothesis that the pathologic changes in AD begin with Aβ42 deposition while in contrast neuronal degeneration and tau hyperphosphorylation occur closer to the time that AD becomes clinically manifested. Our results showing that Aβ42 levels in the CSF in LOAD patients and a-MCI
Comparisons between the controls and CI patients. Controls (n = 9)
CI patients (n = 18)
pValue
F2-IsoPs (pg/mL) 14.3 (12.7–18.1) 22.2 (15.2–26.2) 0.018 F4-NPs (pg/mL) 35.7 (30.6–46.3) 47.9 (37.1–57.2) 0.018 β-amyloid (1-42) 109.4 (102.8–140.8) 50.2 (37.0–94.8)
