[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.159.189.139. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Download PDF
Figure 1.
Kaplan-Meier Curve for Progression From Mild Cognitive Impairment (MCI) to Alzheimer Disease (AD), Stratified by Tertiles of Neurogranin Levels
Kaplan-Meier Curve for Progression From Mild Cognitive Impairment (MCI) to Alzheimer Disease (AD), Stratified by Tertiles of Neurogranin Levels

Neurogranin levels were evaluated at the lowest tertile (<1666 pg/mL), the middle tertile (1666-2734 pg/mL), and the highest tertile (>2734 pg/mL). The lowest tertile was used as reference.

Figure 2.
Annual Change in Neurogranin (NGRN) Levels
Annual Change in Neurogranin (NGRN) Levels

The annual change in NGRN levels obtained from samples of cerebrospinal fluid were assessed using age- and sex-adjusted linear mixed models. The NGRN level was the dependent variable, and clinical diagnosis (treated as a categorical variable), time (ie, the interval between lumbar punctures in years; treated as a continuous variable), and the interaction between diagnosis and time were the independent variables. The reported β represents the estimated change in NGRN levels per year, and the error bars represent the 95% CIs of the reported effects.

Table 1.  
Baseline Characteristics of Patients in the Separate Clinical Diagnostic Categories
Baseline Characteristics of Patients in the Separate Clinical Diagnostic Categories
Table 2.  
Correlations of CSF Neurogranin With Core CSF Biomarkers
Correlations of CSF Neurogranin With Core CSF Biomarkers
Table 3.  
Progression From MCI to AD
Progression From MCI to AD
1.
Blennow  K, Hampel  H.  CSF markers for incipient Alzheimer’s disease. Lancet Neurol. 2003;2(10):605-613.
PubMedArticle
2.
Kester  MI, Scheffer  PG, Koel-Simmelink  MJ,  et al.  Serial CSF sampling in Alzheimer’s disease: specific versus non-specific markers. Neurobiol Aging. 2012;33(8):1591-1598.
PubMedArticle
3.
Perrin  RJ, Fagan  AM, Holtzman  DM.  Multimodal techniques for diagnosis and prognosis of Alzheimer’s disease. Nature. 2009;461(7266):916-922.
PubMedArticle
4.
Zhou  B, Teramukai  S, Yoshimura  K, Fukushima  M.  Validity of cerebrospinal fluid biomarkers as endpoints in early-phase clinical trials for Alzheimer’s disease. J Alzheimers Dis. 2009;18(1):89-102.
PubMed
5.
Scheff  SW, Price  DA, Schmitt  FA, Mufson  EJ.  Hippocampal synaptic loss in early Alzheimer’s disease and mild cognitive impairment. Neurobiol Aging. 2006;27(10):1372-1384.
PubMedArticle
6.
Scheff  SW, Price  DA, Schmitt  FA, DeKosky  ST, Mufson  EJ.  Synaptic alterations in CA1 in mild Alzheimer disease and mild cognitive impairment. Neurology. 2007;68(18):1501-1508.
PubMedArticle
7.
Chang  JW, Schumacher  E, Coulter  PM  II, Vinters  HV, Watson  JB.  Dendritic translocation of RC3/neurogranin mRNA in normal aging, Alzheimer disease and fronto-temporal dementia. J Neuropathol Exp Neurol. 1997;56(10):1105-1118.
PubMedArticle
8.
Slemmon  JR, Feng  B, Erhardt  JA.  Small proteins that modulate calmodulin-dependent signal transduction: effects of PEP-19, neuromodulin, and neurogranin on enzyme activation and cellular homeostasis. Mol Neurobiol. 2000;22(1-3):99-113.
PubMedArticle
9.
Davidsson  P, Blennow  K.  Neurochemical dissection of synaptic pathology in Alzheimer’s disease. Int Psychogeriatr. 1998;10(1):11-23.
PubMedArticle
10.
Reddy  PH, Mani  G, Park  BS,  et al.  Differential loss of synaptic proteins in Alzheimer’s disease: implications for synaptic dysfunction. J Alzheimers Dis. 2005;7(2):103-117.
PubMed
11.
Thorsell  A, Bjerke  M, Gobom  J,  et al.  Neurogranin in cerebrospinal fluid as a marker of synaptic degeneration in Alzheimer’s disease. Brain Res. 2010;1362:13-22.
PubMedArticle
12.
Kvartsberg  H, Duits  FH, Ingelsson  M,  et al.  Cerebrospinal fluid levels of the synaptic protein neurogranin correlates with cognitive decline in prodromal Alzheimer’s disease [published online December 19, 2014]. Alzheimers Dement. doi:10.1016/j.jalz.2014.10.009.
PubMed
13.
McKhann  G, Drachman  D, Folstein  M, Katzman  R, Price  D, Stadlan  EM.  Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984;34(7):939-944.
PubMedArticle
14.
Petersen  RC, Smith  GE, Waring  SC, Ivnik  RJ, Tangalos  EG, Kokmen  E.  Mild cognitive impairment: clinical characterization and outcome. Arch Neurol. 1999;56(3):303-308.
PubMedArticle
15.
Neary  D, Snowden  JS, Gustafson  L,  et al.  Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria. Neurology. 1998;51(6):1546-1554.
PubMedArticle
16.
Román  GC, Tatemichi  TK, Erkinjuntti  T,  et al.  Vascular dementia: diagnostic criteria for research studies. Report of the NINDS-AIREN International Workshop. Neurology. 1993;43(2):250-260.
PubMedArticle
17.
McKeith  IG, Dickson  DW, Lowe  J,  et al; Consortium on DLB.  Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005;65(12):1863-1872.
PubMedArticle
18.
Bouwman  FH, Schoonenboom  NS, Verwey  NA,  et al.  CSF biomarker levels in early and late onset Alzheimer’s disease. Neurobiol Aging. 2009;30(12):1895-1901.
PubMedArticle
19.
Bouwman  FH, van der Flier  WM, Schoonenboom  NS,  et al.  Usefulness of longitudinal measurements of β-amyloid1-42 in cerebrospinal fluid of patients with various cognitive and neurologic disorders. Clin Chem. 2006;52(8):1604-1606.
PubMedArticle
20.
Verwey  NA, Bouwman  FH, van der Flier  WM, Veerhuis  R, Scheltens  P, Blankenstein  MA.  Variability in longitudinal cerebrospinal fluid tau and phosphorylated tau measurements. Clin Chem Lab Med. 2008;46(9):1300-1304.
PubMedArticle
21.
Twisk  JW. Applied Longitudinal Data Analysis for Epidemiology. Cambridge, England: Cambridge University Press; 2003.
22.
George  AJ, Gordon  L, Beissbarth  T,  et al.  A serial analysis of gene expression profile of the Alzheimer’s disease Tg2576 mouse model. Neurotox Res. 2010;17(4):360-379.
PubMedArticle
23.
Scheff  SW, Price  DA, Ansari  MA,  et al.  Synaptic change in the posterior cingulate gyrus in the progression of Alzheimer’s disease. J Alzheimers Dis. 2015;43(3):1073-1090.
PubMed
24.
Jack  CR  Jr, Knopman  DS, Jagust  WJ,  et al.  Tracking pathophysiological processes in Alzheimer’s disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol. 2013;12(2):207-216.
PubMedArticle
25.
Koob  AO, Shaked  GM, Bender  A, Bisquertt  A, Rockenstein  E, Masliah  E.  Neurogranin binds α-synuclein in the human superior temporal cortex and interaction is decreased in Parkinson’s disease. Brain Res. 2014;1591:102-110.
PubMedArticle
26.
Ingelsson  M, Fukumoto  H, Newell  KL,  et al.  Early Aβ accumulation and progressive synaptic loss, gliosis, and tangle formation in AD brain. Neurology. 2004;62(6):925-931.
PubMedArticle
27.
Blennow  K, Bogdanovic  N, Alafuzoff  I, Ekman  R, Davidsson  P.  Synaptic pathology in Alzheimer’s disease: relation to severity of dementia, but not to senile plaques, neurofibrillary tangles, or the ApoE4 allele. J Neural Transm. 1996;103(5):603-618.
PubMedArticle
28.
Braak  H, Braak  E.  Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. 1991;82(4):239-259.
PubMedArticle
29.
Buerger  K, Otto  M, Teipel  SJ,  et al.  Dissociation between CSF total tau and tau protein phosphorylated at threonine 231 in Creutzfeldt-Jakob disease. Neurobiol Aging. 2006;27(1):10-15.
PubMedArticle
30.
Riemenschneider  M, Wagenpfeil  S, Vanderstichele  H,  et al.  Phospho-tau/total tau ratio in cerebrospinal fluid discriminates Creutzfeldt-Jakob disease from other dementias. Mol Psychiatry. 2003;8(3):343-347.
PubMedArticle
31.
Kester  MI, van der Vlies  AE, Blankenstein  MA,  et al.  CSF biomarkers predict rate of cognitive decline in Alzheimer disease. Neurology. 2009;73(17):1353-1358.
PubMedArticle
Original Investigation
November 2015

Neurogranin as a Cerebrospinal Fluid Biomarker for Synaptic Loss in Symptomatic Alzheimer Disease

Author Affiliations
  • 1Alzheimer Center and Department of Neurology, VU University Medical Center, Amsterdam, the Netherlands
  • 2Department of Clinical Chemistry, VU University Medical Center, Amsterdam, the Netherlands
  • 3Department of Pathology and Immunology, Washington University School of Medicine, St Louis, Missouri
  • 4Department of Epidemiology and Biostatistics, VU University Medical Center, Amsterdam, the Netherlands
  • 5The Knight Alzheimer’s Disease Research Center, Washington University School of Medicine, St Louis, Missouri
  • 6Department of Neurology, Washington University School of Medicine, St Louis, Missouri
  • 7Hope Center for Neurological Disorders, Washington University School of Medicine, St Louis, Missouri
JAMA Neurol. 2015;72(11):1275-1280. doi:10.1001/jamaneurol.2015.1867
Abstract

Importance  Neurogranin (NGRN) seems to be a promising novel cerebrospinal fluid (CSF) biomarker for synaptic loss; however, clinical, and especially longitudinal, data are sparse.

Objective  To examine the utility of NGRN, with repeated CSF sampling, for diagnosis, prognosis, and monitoring of Alzheimer disease (AD).

Design, Setting, and Participants  Longitudinal study of consecutive patients who underwent 2 lumbar punctures between the beginning of 1995 and the end of 2010 within the memory clinic–based Amsterdam Dementia Cohort. The study included 163 patients: 37 cognitively normal participants (mean [SE] age, 64 [2] years; 38% female; and mean [SE] Mini-Mental State Examination [MMSE] score, 28 [0.3]), 61 patients with mild cognitive impairment (MCI) (mean [SE] age, 68 [1] years; 38% female; and mean [SE] MMSE score, 27 [0.3]), and 65 patients with AD (mean [SE] age, 65 [1] years; 45% female; and mean [SE] MMSE score, 22 [0.7]). The mean (SE) interval between lumbar punctures was 2.0 (0.1) years, and the mean (SE) duration of cognitive follow-up was 3.8 (0.2) years. Measurements of CSF NGRN levels were obtained in January and February 2014.

Main Outcome and Measure  Levels of NGRN in CSF samples.

Results  Baseline CSF levels of NGRN in patients with AD (median level, 2381 pg/mL [interquartile range, 1651-3416 pg/mL]) were higher than in cognitively normal participants (median level, 1712 pg/mL [interquartile range, 1206-2724 pg/mL]) (P = .04). Baseline NGRN levels were highly correlated with total tau and tau phosphorylated at threonine 181 in all patient groups (all P < .001), but not with Aβ42. Baseline CSF levels of NGRN were also higher in patients with MCI who progressed to AD (median level, 2842 pg/mL [interquartile range, 1882-3950 pg/mL]) compared with those with stable MCI (median level, 1752 pg/mL [interquartile range, 1024-2438 pg/mL]) (P = .004), and they were predictive of progression from MCI to AD (hazard ratio, 1.8 [95% CI, 1.1-2.9]; stratified by tertiles). Linear mixed-model analyses demonstrated that within-person levels of NGRN increased over time in cognitively normal participants (mean [SE] level, 90 [45] pg/mL per year; P < .05) but not in patients with MCI or AD.

Conclusions and Relevance  Neurogranin is a promising biomarker for AD because levels were elevated in patients with AD compared with cognitively normal participants and predicted progression from MCI to AD. Within-person levels of NGRN increased in cognitively normal participants but not in patients with later stage MCI or AD, which suggests that NGRN may reflect presymptomatic synaptic dysfunction or loss.

Introduction

The core cerebrospinal fluid (CSF) biomarkers Aβ42, total tau, and tau phosphorylated at threonine 181 (P-tau181) reflect the neuropathological hallmarks of Alzheimer disease (AD), amyloid plaques, and neurofibrillary tangles.1 Clinically, AD is characterized by cognitive decline, but once a patient has AD pathology, these core CSF biomarkers appear not to reflect further functional decline owing to their relative stability during clinical AD.24

The synapse plays a central and essential role in cognitive function because it subserves neuronal transmission. Synaptic loss is an early event in the pathogenesis of AD5 and has been shown to correlate with cognitive decline.6 Biomarkers that reflect synaptic integrity could therefore be useful for both an accurate, early diagnosis and disease prognosis. A promising biomarker candidate is the postsynaptic protein neurogranin (NGRN),7 which is expressed exclusively in the brain, particularly in dendritic spines.7 Neurogranin binds to calmodulin in the absence of calcium and is involved in synaptic plasticity and long-term potentiation, processes essential for learning.8 Decreased levels of NGRN have been reported in AD brain tissue samples compared with control samples,9,10 and recent studies have reported increases in CSF NGRN levels in patients with AD compared with controls.11,12 We aimed to evaluate the diagnostic and prognostic utility of NGRN as a CSF biomarker in a cohort of patients with AD or mild cognitive impairment (MCI) and cognitively normal participants, and to assess its dynamics during disease progression in longitudinal CSF samples obtained from participants over 2 years.

Methods
Participants

From the Amsterdam Dementia Cohort, we included 65 patients with AD, 61 patients with MCI, and 37 cognitively normal participants, all of whom had CSF samples obtained at 2 time points.2 At baseline, all patients underwent standard dementia screening, including physical and neurological examinations, laboratory tests, electroencephalography, and magnetic resonance imaging. Cognitive screening included a Mini-Mental State Examination but usually also involved comprehensive neuropsychological testing. The diagnosis of probable AD was made according to the criteria of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association.13 The diagnosis of MCI was made according to the criteria of Petersen et al.14 All of the patients with probable AD or MCI experienced subjective cognitive decline, and, in addition, they scored in a cognitive domain below expected (<1 SD), and yet they did not have dementia. When the results of all examinations were normal, the patients were considered to have subjective memory complaints. The cognitively normal group consisted of 31 participants with subjective memory complaints, 2 participants with a psychiatric disorder (eg, depression), 2 participants with temporal epilepsy (treated with medication), and 2 healthy volunteers. Diagnoses were made by consensus of a multidisciplinary team. Our study was approved by the ethical review board of the VU University Medical Center in Amsterdam, the Netherlands, and all participants provided written informed consent.

Follow-up

Patients were followed up clinically on an annual basis. Of the 61 patients with MCI (with a mean [SE] follow-up of 2.7 [0.3] years), 17 remained stable, 36 progressed to AD,13 and 8 progressed to other types of dementia (2 patients progressed to frontotemporal lobar degeneration,15 3 patients to vascular dementia,16 1 patient to dementia with Lewy bodies, 1 patients to progressive supranuclear palsy,17 and 1 patient to normal-pressure hydrocephalus). Of the 37 cognitively normal participants (with a mean [SE] follow-up of 4.0 [0.5] years for 31 cognitively normal participants), 6 with subjective memory complaints progressed to MCI, 3 progressed to AD, 1 progressed to vascular dementia, and 27 remained stable. During follow-up, patients were asked to undergo a second lumbar puncture (minimum interval, 6 months). Owing to technical reasons (inadequate amount of CSF aliquoted in selected vials), NGRN was unavailable for 1 patient at baseline and 3 patients at follow-up.

Analyses of CSF Samples

Samples of CSF were obtained by standard lumbar puncture, using a 25-gauge Quincke needle, and collected in 10-mL polypropylene tubes. Within 2 hours, CSF samples were centrifuged at 1800g for 10 minutes at 4°C, aliquoted in polypropylene tubes of 0.5 or 1 mL, and stored at −80°C until further analysis of NRGN. Baseline CSF Aβ42, total tau, and P-tau181 were measured with an enzyme-linked immunosorbent assay (INNOTEST ELISA; Fujirebio [formerly Innogenetics]) at the VU University Medical Center.18 Because the manufacturer does not supply controls, the consistency of the assay’s performance was monitored using pools of surplus CSF samples. The mean (SD) intra-assay coefficient of variation was 2.0% (0.5%) for Aβ42, 3.2% (1.3%) for total tau, and 2.9% (0.8%) for P-tau181 as calculated from averaging the coefficient of variation of duplicates from 5 runs (with 36 samples each) randomly selected over 2 years. The mean (SD) interassay coefficient of variation was 10.9% (1.8%) for Aβ42, 9.9% (2.1%) for total tau, and 9.1% (1.8%) for P-tau181, as analyzed in a high and low pool from 13 consecutive pool preparations used in total in 189 to 231 runs. Samples of CSF were analyzed for NGRN using a sandwich immunoassay developed on a Singulex Erenna system at Washington University in St Louis, Missouri.

NGRN Assay

Recombinant glutathione S-transferase (GST)–NGRN fusion protein was produced in pGEX-4T-1 (GE Healthcare Biosciences), expressed in Escherichia coli, and then purified according to the manufacturer. Rabbits were immunized with the GST-NGRN fusion protein at Harlan Bioproducts for Science (Madison, Wisconsin). Antiserum samples were first passed over a glutathione-GST column to remove anti-GST antibodies and then a glutathione-GST-NGRN column to obtain affinity-purified anti-NGRN antibodies. This material was epitope-mapped using spot-peptide membrane arrays, where spot 1 comprised residues 1 to 10, spot 2 residues 2 to 11, spot 3 residues 3 to 12, and so on, until the entire sequence of 78 residues was covered. Two peptides were synthesized as a result of this mapping experiment. Epitope affinity columns were prepared with N-terminal peptide S10-D23 and C-terminal peptide G49-G60, each with a nonnative N-terminal cysteine for conjugation to the column. Synthetic 78-mer human NGRN was prepared and characterized by AAPPTec using C18-reversed phase-high performance liquid chromatography and electrospray ionization-mass spectrometry, and this material was used as the immunoassay standard. The stock mean (SD) concentration, prepared in phosphate-buffered saline with azide, was determined by amino acid analysis in triplicate (at AAA Service Laboratory, Inc, in Damascus, Oregon) and found to be 1.12 (0.03) mg/mL.

A sandwich immunoassay was developed for CSF on a Singulex Erenna system using the 2 epitope-specific rabbit antibodies (recognizing N-terminal epitope S11-D23 and C-terminal epitope G49-G60 antibodies). The C-terminal–specific antibody (P-4793) was coupled to magnetic beads and used as the capture antibody, and the N-terminal–specific antibody (P-4794) was labeled with a fluorescent dye and used as the capping/detection antibody. Three control CSF pools were prepared at concentrations corresponding to low (approximately 400 pg/mL), medium (approximately 1500 pg/mL), and high (approximately 2900 pg/mL) levels of NGRN. These were aliquoted and stored at −80°C for one-time use and included in every assay. For these 3 control CSF pools, the intra-assay precision (coefficient of variation) for 15 replicates was 6.6%, 7.1%, and 5.8%, respectively, and the interassay coefficient of variation for 23 assays was 5.7%, 6.2%, and 5.4%, respectively. The mean (SD) dilutional linearity recovery for 5 samples was 98.2% (9.45%), the mean (SD) spike recovery for 5 different samples was 104.8% (2.5%), and the lower limit of quantitation was approximately 2 pg/mL. To circumvent interassay variability, baseline and follow-up samples were analyzed on the same assay plate.19,20 The teams involved in the CSF analysis were not aware of the clinical diagnoses.

Statistical Analysis

Cross-sectional differences among groups were assessed using analysis of variance, with post hoc Bonferroni corrections, or the Fisher exact test when applicable. The CSF biomarkers were log-transformed to fit the assumptions needed for analysis of variance and were adjusted for sex and age. Pearson correlations were assessed per diagnostic group using baseline log-transformed CSF biomarker levels. Cox proportional hazards models, adjusted for sex and age, were performed to analyze the predictive value of the CSF biomarkers for progression of MCI to AD and for progression of cognitively normal to either MCI or AD. For the Cox analyses, NGRN was evaluated as tertiles (<1666 pg/mL at tertile 1, 1666-2734 pg/mL at tertile 2, and >2734 pg/mL at tertile 3). Hazard ratios are presented with 95% CIs. Kaplan-Meier curves were created for illustrative purposes. Finally, age- and sex-adjusted linear mixed models were applied to assess within-person annual changes over time in CSF biomarker levels by diagnosis. The CSF NGRN level was the dependent variable, while diagnosis (treated as a categorical variable), time (ie, the interval between lumbar punctures in years; treated as a continuous variable), and the interaction between diagnosis and time were the independent variables. Diagnostic categories were recoded as dummy variables in order to estimate the mean (SE) values for each category. All linear mixed models were specified with a random intercept and/or slope based on −2LL (minus twice the log likelihood) criteria.21 For statistical analyses, we used SPSS Statistics 21.0 (for Windows; IBM). Statistical significance was set at P < .05.

Results
Baseline Characteristics

The baseline characteristics of the patients are shown in Table 1. The baseline levels of NGRN in the patients with AD were higher than those in the cognitively normal participants (P = .04). There were no differences in NGRN levels between the patients with MCI and the patients with AD or between the patients with MCI and the cognitively normal participants (P = .25 and P = .31). Baseline NGRN levels were strongly positively correlated with total tau and P-tau181 levels in all clinical groups but were not correlated with Aβ42 level, as shown in Table 2.

Predictive Value of Baseline Levels of Novel Biomarkers for Progression

Further analyses with analysis of variance showed that baseline levels of NGRN were higher in patients with MCI who progressed to AD than in patients with MCI who remained stable. Cox regression analyses revealed that baseline NGRN levels were predictive of progression from MCI to AD in the same order of magnitude as the core biomarkers Aβ42, total tau, and P-tau181, as shown in Table 3 and illustrated by Figure 1. In the cognitively normal group, there was a trend, albeit nonsignificant, of NGRN levels predicting progression within the AD continuum (hazard ratio, 1.6 [95% CI, 0.64-4.0]).

Longitudinal Changes in Novel Biomarkers

Linear mixed-model analyses showed that levels of NGRN increased within cognitively normal participants (mean [SE] level, 90 [45] pg/mL per year; P < .05) but not in patients with MCI (mean [SE] level, 53 [42] pg/mL per year; P = .22) or AD (mean [SE] level, 14 [45] pg/mL per year; P = .75), as shown in Figure 2. This indicates that NGRN levels increase in a very early (asymptomatic) stage but not in later stages of the AD continuum.

Discussion

In our longitudinal study with repeated CSF sampling, we showed that NGRN has potential as a novel biomarker for synaptic dysfunction in AD. Baseline NGRN levels are higher in patients with MCI who progressed to AD than in patients with MCI who were clinically stable, and these higher levels predicted progression to AD. In addition, levels of NGRN were higher in patients with AD than in cognitively normal participants. Longitudinally, we showed that levels of NGRN increased over time in participants who were cognitively normal but not in patients with MCI or AD who were already cognitively impaired.

Our study confirms that CSF levels of NGRN are increased in patients with AD compared with cognitively normal participants.11,12 This is consistent with postmortem studies in both mouse models22 and human tissue.10 The measurement of CSF NGRN level could possibly aid in the early diagnosis of AD. In addition, we show that NGRN levels are higher in patients with MCI who progress to AD than in patients with MCI who remain stable and that levels are predictive of progression from MCI to AD, as has recently also been reported by another group in a similar cohort using a different assay.12 Thus, the CSF NGRN level appears to have a prognostic as well as a diagnostic value.

Levels of NGRN in CSF were much higher in the present study than in the study by Kvartsberg et al,12 which also used an enzyme-linked immunosorbent assay. Differences in absolute values may be a result of the different antibodies used in the 2 assays. Further studies comparing both enzyme-linked immunosorbent assays are needed to clarify this matter. However, the direction and order of magnitude of the variance between patient groups were the same. It is possible that differences in the calibration and/or antibodies used contributed to the absolute differences. Among the limitations, the cognitively normal group was a mixed group of participants that also included patients with psychiatric disorders and temporal lobe epilepsy, which may hamper generalizability. In addition, our cognitively normal group was biased toward participants who showed decline (6 progressed to MCI, and 4 to dementia), which could have even diluted the baseline effect. This bias could be due to the fact that progressors are more likely to return to our clinic for a second lumbar puncture.2 However, thanks to this follow-up, we were able to evaluate change over time in NGRN levels for all stages of the AD continuum. Another limitation was that the size of the group of cognitively normal participants was too small to reliably analyze for risk of progression with Cox proportional hazards models. However, the trend was in the same order of magnitude as shown for the group of patients with MCI. This supports the notion that synapse loss is a very early process during AD pathogenesis.

With our longitudinal analyses, we found that NGRN levels increased in time in cognitively normal participants but not in patients with MCI or AD. This pattern is also consistent with the view that synaptic changes mainly occur in the earliest phase of the AD continuum, even before the stage of MCI.23 Within the AD continuum (as proposed by Jack et al24), it could be hypothesized that CSF NGRN levels increase very early. Further studies are needed to confirm our findings. Additional cross-sectional studies are needed to clarify whether NGRN is specific for AD, or whether it also reflects synaptic changes in other neurodegenerative diseases.25

We observed high positive correlations between CSF NGRN levels and both total tau and P-tau181 levels, but not with Aβ42 levels. The lack of correlation between NGRN level and Aβ42 level is in line with studies showing that both synapse loss and clinical stage are unrelated to the amount of amyloid plaques.2628 The CSF total tau level, on the other hand, is related to cognitive deterioration, likely reflecting neuronal cell death.2931 The NGRN level could be a more specific marker for pathological changes that lead to cognitive deterioration because it represents the more specific, and potentially earlier, process of synapse loss. Importantly, it reflects a mechanism that could be useful in treatment trials to monitor the effects of drugs on synaptic integrity.

Conclusions

In summary, NGRN levels are lower in the cognitively normal participants than in patients with AD, and then they increase over time. Furthermore, increased levels of NGRN are associated with progression to AD in patients with MCI. In addition to the “core CSF biomarkers” Aβ42, total tau, and P-tau181, the NGRN level could have added value because it is a reflection of a pathophysiological process that is directly related to cognitive changes (ie, synapse function).

Back to top
Article Information

Accepted for Publication: June 17, 2015.

Corresponding Author: Maartje I. Kester, MD, PhD, Alzheimer Center and Department of Neurology, VU University Medical Center, PO Box 7057, 1007 MB Amsterdam, the Netherlands (m.kester@vumc.nl).

Published Online: September 14, 2015. doi:10.1001/jamaneurol.2015.1867.

Author Contributions: Drs Kester and van der Flier had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Kester, Teunissen, Crimmins, Holtzman.

Acquisition, analysis, or interpretation of data: Kester, Teunissen, Herries, Ladenson, Scheltens, van der Flier, Morris, Fagan.

Drafting of the manuscript: Kester, Teunissen, Crimmins.

Critical revision of the manuscript for important intellectual content: Teunissen, Herries, Ladenson, Scheltens, van der Flier, Morris, Holtzman, Fagan.

Statistical analysis: Kester.

Obtained funding: Kester, Scheltens, Holtzman.

Administrative, technical, or material support: Kester, Teunissen, Crimmins, Herries, Ladenson, Morris, Fagan.

Study supervision: Teunissen, Scheltens, van der Flier.

Conflict of Interest Disclosures: Dr van der Flier performs contract research for, and has been an invited speaker at, Boehringer Ingelheim. Dr Teunissen serves on the advisory board of Fujirebio and Roche; received research consumables from Euroimmun, IBL International, Fujirebio, Invitrogen, and Mesoscale Discovery; performed contract research for IBL International, Shire, Boehringer Ingelheim, Roche, and Probiodrug; and received grants from the European Commission, the Dutch Research Council (ZonMW), the ISAO, and the Alzheimer’s Drug Discovery Foundation. Dr Ladenson reports being named on patents related to the use of brain biomarkers. These are being managed by Washington University in accordance with university policy. Dr Scheltens has received grant support (for VU University Medical Center) from GE Healthcare, Danone Research, Piramal, and Merck. In the past 2 years, he has received consultancy/speaker fees (paid to the VU University Medical Center) from Eli Lilly, GE Healthcare, Novartis, Forum, Sanofi, and Nutricia. Dr Morris has participated and is currently participating in clinical trials of antidementia drugs sponsored by the following companies: Janssen Immunotherapy, Pfizer, Eli Lilly/Avid Radiopharmaceuticals, the Study of Nasal Insulin to Fight Forgetfulness, and the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease trial. He has served as a consultant for Lilly USA, ISIS Pharmaceuticals, and the Charles Dana Foundation. Dr Holtzman is a cofounder of C2N Diagnostics LLC; is on the scientific advisory boards of AstraZeneca, Genentech, Neurophage, and C2N Diagnostics; and is a consultant for Eli Lilly. Washington University receives grants to the laboratory of Dr Holtzman from the Tau Consortium, the Cure Alzheimer’s Fund, the JPB Foundation, Eli Lilly, Janssen, and C2N Diagnostics. Dr Fagan is on the scientific advisory boards of IBL International and Roche and is a consultant for AbbVie. No other disclosures were reported.

Funding/Support: Research at the Alzheimer’s Center of the VU University Medical Center is part of the Neurodegeneration Research Program of the Neuroscience Campus Amsterdam. The Alzheimer Center is supported by Alzheimer Nederland and Stichting VU University Medical Center funds. The clinical database structure was developed with funding from Stichting Dioraphte. Dr Kester was financially supported by a research fellowship from Alzheimer Nederland (grant WE 15-2012-03). The analyses of CSF at Washington University in St Louis were supported by National Institute on Aging grant P01 AG026276 (Dr Morris).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

References
1.
Blennow  K, Hampel  H.  CSF markers for incipient Alzheimer’s disease. Lancet Neurol. 2003;2(10):605-613.
PubMedArticle
2.
Kester  MI, Scheffer  PG, Koel-Simmelink  MJ,  et al.  Serial CSF sampling in Alzheimer’s disease: specific versus non-specific markers. Neurobiol Aging. 2012;33(8):1591-1598.
PubMedArticle
3.
Perrin  RJ, Fagan  AM, Holtzman  DM.  Multimodal techniques for diagnosis and prognosis of Alzheimer’s disease. Nature. 2009;461(7266):916-922.
PubMedArticle
4.
Zhou  B, Teramukai  S, Yoshimura  K, Fukushima  M.  Validity of cerebrospinal fluid biomarkers as endpoints in early-phase clinical trials for Alzheimer’s disease. J Alzheimers Dis. 2009;18(1):89-102.
PubMed
5.
Scheff  SW, Price  DA, Schmitt  FA, Mufson  EJ.  Hippocampal synaptic loss in early Alzheimer’s disease and mild cognitive impairment. Neurobiol Aging. 2006;27(10):1372-1384.
PubMedArticle
6.
Scheff  SW, Price  DA, Schmitt  FA, DeKosky  ST, Mufson  EJ.  Synaptic alterations in CA1 in mild Alzheimer disease and mild cognitive impairment. Neurology. 2007;68(18):1501-1508.
PubMedArticle
7.
Chang  JW, Schumacher  E, Coulter  PM  II, Vinters  HV, Watson  JB.  Dendritic translocation of RC3/neurogranin mRNA in normal aging, Alzheimer disease and fronto-temporal dementia. J Neuropathol Exp Neurol. 1997;56(10):1105-1118.
PubMedArticle
8.
Slemmon  JR, Feng  B, Erhardt  JA.  Small proteins that modulate calmodulin-dependent signal transduction: effects of PEP-19, neuromodulin, and neurogranin on enzyme activation and cellular homeostasis. Mol Neurobiol. 2000;22(1-3):99-113.
PubMedArticle
9.
Davidsson  P, Blennow  K.  Neurochemical dissection of synaptic pathology in Alzheimer’s disease. Int Psychogeriatr. 1998;10(1):11-23.
PubMedArticle
10.
Reddy  PH, Mani  G, Park  BS,  et al.  Differential loss of synaptic proteins in Alzheimer’s disease: implications for synaptic dysfunction. J Alzheimers Dis. 2005;7(2):103-117.
PubMed
11.
Thorsell  A, Bjerke  M, Gobom  J,  et al.  Neurogranin in cerebrospinal fluid as a marker of synaptic degeneration in Alzheimer’s disease. Brain Res. 2010;1362:13-22.
PubMedArticle
12.
Kvartsberg  H, Duits  FH, Ingelsson  M,  et al.  Cerebrospinal fluid levels of the synaptic protein neurogranin correlates with cognitive decline in prodromal Alzheimer’s disease [published online December 19, 2014]. Alzheimers Dement. doi:10.1016/j.jalz.2014.10.009.
PubMed
13.
McKhann  G, Drachman  D, Folstein  M, Katzman  R, Price  D, Stadlan  EM.  Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984;34(7):939-944.
PubMedArticle
14.
Petersen  RC, Smith  GE, Waring  SC, Ivnik  RJ, Tangalos  EG, Kokmen  E.  Mild cognitive impairment: clinical characterization and outcome. Arch Neurol. 1999;56(3):303-308.
PubMedArticle
15.
Neary  D, Snowden  JS, Gustafson  L,  et al.  Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria. Neurology. 1998;51(6):1546-1554.
PubMedArticle
16.
Román  GC, Tatemichi  TK, Erkinjuntti  T,  et al.  Vascular dementia: diagnostic criteria for research studies. Report of the NINDS-AIREN International Workshop. Neurology. 1993;43(2):250-260.
PubMedArticle
17.
McKeith  IG, Dickson  DW, Lowe  J,  et al; Consortium on DLB.  Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005;65(12):1863-1872.
PubMedArticle
18.
Bouwman  FH, Schoonenboom  NS, Verwey  NA,  et al.  CSF biomarker levels in early and late onset Alzheimer’s disease. Neurobiol Aging. 2009;30(12):1895-1901.
PubMedArticle
19.
Bouwman  FH, van der Flier  WM, Schoonenboom  NS,  et al.  Usefulness of longitudinal measurements of β-amyloid1-42 in cerebrospinal fluid of patients with various cognitive and neurologic disorders. Clin Chem. 2006;52(8):1604-1606.
PubMedArticle
20.
Verwey  NA, Bouwman  FH, van der Flier  WM, Veerhuis  R, Scheltens  P, Blankenstein  MA.  Variability in longitudinal cerebrospinal fluid tau and phosphorylated tau measurements. Clin Chem Lab Med. 2008;46(9):1300-1304.
PubMedArticle
21.
Twisk  JW. Applied Longitudinal Data Analysis for Epidemiology. Cambridge, England: Cambridge University Press; 2003.
22.
George  AJ, Gordon  L, Beissbarth  T,  et al.  A serial analysis of gene expression profile of the Alzheimer’s disease Tg2576 mouse model. Neurotox Res. 2010;17(4):360-379.
PubMedArticle
23.
Scheff  SW, Price  DA, Ansari  MA,  et al.  Synaptic change in the posterior cingulate gyrus in the progression of Alzheimer’s disease. J Alzheimers Dis. 2015;43(3):1073-1090.
PubMed
24.
Jack  CR  Jr, Knopman  DS, Jagust  WJ,  et al.  Tracking pathophysiological processes in Alzheimer’s disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol. 2013;12(2):207-216.
PubMedArticle
25.
Koob  AO, Shaked  GM, Bender  A, Bisquertt  A, Rockenstein  E, Masliah  E.  Neurogranin binds α-synuclein in the human superior temporal cortex and interaction is decreased in Parkinson’s disease. Brain Res. 2014;1591:102-110.
PubMedArticle
26.
Ingelsson  M, Fukumoto  H, Newell  KL,  et al.  Early Aβ accumulation and progressive synaptic loss, gliosis, and tangle formation in AD brain. Neurology. 2004;62(6):925-931.
PubMedArticle
27.
Blennow  K, Bogdanovic  N, Alafuzoff  I, Ekman  R, Davidsson  P.  Synaptic pathology in Alzheimer’s disease: relation to severity of dementia, but not to senile plaques, neurofibrillary tangles, or the ApoE4 allele. J Neural Transm. 1996;103(5):603-618.
PubMedArticle
28.
Braak  H, Braak  E.  Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. 1991;82(4):239-259.
PubMedArticle
29.
Buerger  K, Otto  M, Teipel  SJ,  et al.  Dissociation between CSF total tau and tau protein phosphorylated at threonine 231 in Creutzfeldt-Jakob disease. Neurobiol Aging. 2006;27(1):10-15.
PubMedArticle
30.
Riemenschneider  M, Wagenpfeil  S, Vanderstichele  H,  et al.  Phospho-tau/total tau ratio in cerebrospinal fluid discriminates Creutzfeldt-Jakob disease from other dementias. Mol Psychiatry. 2003;8(3):343-347.
PubMedArticle
31.
Kester  MI, van der Vlies  AE, Blankenstein  MA,  et al.  CSF biomarkers predict rate of cognitive decline in Alzheimer disease. Neurology. 2009;73(17):1353-1358.
PubMedArticle
×