Abstract
Objectives: Decreased levels of Aβ1-42 are found in CSF of patients with AD. Because early stages of Creutzfeldt-Jakob disease (CJD) and AD share several clinical features, we investigated Aβ1-42 levels in CSF of these groups, inferring that this might give additional help in differentiating patients with CJD from AD patients.
Methods: We investigated 27 patients with CJD, 14 patients with AD, 19 patients with other dementias, and 20 nondemented controls (NDC) for Aβ1-42 in CSF. Twenty-four of the 27 CJD patients were neuropathologically verified. All the neuropathologically verified patients presented with a type 1 prion protein pattern. CJD patients were all homozygous for methionine at codon 129. Except in five CJD patients, no β-amyloid plaques were seen. Additionally, APOE status was determined in patients with CJD.
Results: Levels of Aβ1-42 in CSF were decreased in patients with AD as well as in CJD. Levels of Aβ1-42 in CSF of patients with CJD and AD were significantly different from the other dementia and NDC groups. There was no substantial difference between the CJD and AD groups (p = 0.66). Decreased levels of Aβ1-42 did not correlate with the APOE ε4 load in patients with CJD.
Conclusion: Low levels of Aβ1-42 in CSF do not exclude a diagnosis of CJD. Decreased levels of Aβ1-42 in CSF can occur without β-amyloid plaque formation in the brain. However, the underlying mechanism of this phenomenon must be elucidated.
The supposed pathophysiology of Creutzfeldt-Jakob disease (CJD) shares several features with nontransmissible amyloidoses, especially Alzheimer’s disease (AD).1 In both groups a normal or mutant precursor protein or peptide is incorporated into an aggregated, noncovalent linked polymeric fibril-like structure, often with a high level of beta-sheet formation. This order of aggregation can further induce polymerization of an additional precursor protein, thus providing a possible basis for propagation or transmission.1-4 Apart from these similarities, clinically sporadic CJD is difficult to differentiate from AD, especially in early stages of the disease.5 The diagnosis of CJD is made according to clinical symptoms and EEG.6,7 The diagnosis can be supported by a positive immunoblot against 14-3-3 proteins,8,9 elevated levels of S100 beta protein,10 neuron-specific enolase (NSE),11 or tau protein.12 The latter protein can also be found in elevated levels in CSF of patients with AD.13-15 However, the cutoff level of tau protein in CSF that supports the diagnosis is much higher in CSF of patients with CJD.12
Besides the histopathology of neurofibrillary tangles consisting of hyperphosphorylated tau protein,16 AD is characterized by a typical plaque formation of amyloid β protein (Aβ).17-19 These plaques consist of residues of peptides 39 to 43 derived from the Aβ-precursor protein (APP).20 Decreased levels of Aβ1-42 have been proposed as a candidate marker for diagnosis of AD.21 These results have been confirmed by several clinical studies.22-25 It is assumed that the decrease of Aβ1-42 in CSF might reflect the additional recruitment of Aβ1-42 to the intracerebral amyloid plaques.21 As β-amyloid plaque formation is not a typical event in CJD, we investigated CSF of patients with CJD for Aβ1-42 levels, inferring that this might give additional help in the neurochemical differentiation of patients with CJD from AD patients.
Patients and materials.
Patients.
We investigated 27 patients with CJD, 14 patients with AD, and 19 patients with other dementias (non-AD, non-CJD) for Aβ1-42 in CSF. These patients were initially seen under the differential diagnosis of having CJD. Additionally, we investigated 20 nondemented controls (NDC). The study was approved by the local ethics committee in Göttingen.
Patients with CJD.
Diagnosis of CJD was done according to clinical criteria.6,7 We investigated 21 women and 6 men. Median age was 68 years (range 31 to 82 years). Twenty-four patients were later neuropathologically verified as having definite CJD.26 The other three patients were clinically diagnosed as probable CJD cases. All CJD patients were homozygous for methionine at codon 129. All neuropathologically verified cases had a type 1 prion protein pattern according to Parchi (MM type 1).27,28 APOE status was available for 26 CJD patients. PCR was performed according to Hixson and Vernier.29
Patients with AD.
We investigated nine women and five men with a median age of 69.5 years (range 54 to 83 years). For the diagnosis of AD, patients had to fulfill the criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV)30 and the criteria published by McKhann et al.31
In two patients AD was later neuropathologically verified according to standard protocol.32
Other dementias.
Patients with other dementias were all initially investigated under the differential diagnosis of CJD. During the evaluating process, it turned out that these patients did not fulfill the criteria for AD or CJD. Final diagnosis of these patients was made according to the latest available data. This group comprised 19 patients (13 women, 6 men) with a median age of 67 years (range 27 to 77 years): 5 with parkinsonian syndrome and dementia, 3 with multisystem atrophy, 3 with multi-infarct dementia, 3 with a postencephalitic syndrome, 2 with a leukoencephalopathy, and 1 each with MS, epilepsy, and vasculitis.
Nondemented controls.
This group comprised 20 patients (11 women, 9 men) with a median age of 57 years (range 45 to 76 years). These patients did not show any clinical signs of dementia and had a lumbar puncture for other differential diagnostic reasons.
Aβ1-42 assay.
The concentration of Aβ1-42 in CSF was measured using a sandwich ELISA obtained from Innogenetics, Belgium.23,25 The assay detection limit is 50 pg/mL, with a linearity in the range of 200 pg/mL to 1500 pg/mL. An artificial decrease of Aβ1-42 by repeated freeze-thawing cycles, especially after the first cycle, was described earlier.33 As in measurement of CJD, AD and other dementia samples were done after the second freezing cycle; values for NDCs were also measured after two freeze-thawing cycles.
14-3-3 immunoblot.
The 14-3-3 immunoblot was performed on all samples according to the previously published standard method.8,9
Statistical analysis.
Comparison of the Aβ1-42 distribution between subgroups in the study population was based on nonparametric rank tests (for two groups, Wilcoxon-Mann-Whitney U-test; for more than two groups, Kruskal-Wallis test).
Results.
Levels of Aβ1-42 in CSF.
Levels of Aβ1-42 in patients with CJD ranged from 58 to 823 pg/mL (median 352 pg/mL; mean 390 pg/mL, SD 170 pg/mL; figure). Patients with AD had levels of 217 to 836 pg/mL in CSF (median 318 pg/mL; mean 361 pg/mL, SD 153 pg/mL). The range of Aβ1-42 in CSF of patients with other dementias ranged from 189 to 1209 pg/mL (median 557 pg/mL; mean 575 pg/mL, SD 227 pg/mL). Levels of Aβ1-42 in the NDC group was between 631 and 1223 pg/mL (median 878 pg/mL; mean 903 pg/mL, SD 163 pg/mL).
Figure. Boxplot of Aβ1-42 levels in CSF. Plot shows 25th, 50th, and 75th percentiles in the boxes (lower, middle, upper crosslines, respectively); 10th and 90th percentiles are indicated by whiskers and outliers by circles. NDC = nondemented controls; OD = other dementias; CJD = Creutzfeldt-Jakob disease.
There was a difference of Aβ1-42 levels in CSF between these groups (p < 0.0001). Moreover, there was a difference between the AD and other dementias group (p = 0.0013) and the AD and NDC group (p < 0.0001). There was also a difference between the CJD and the other dementias group (p = 0.0022) and the CJD and NDC group (p < 0.0001). There was no difference between the CJD and the AD group (p = 0.66).
Comparison of Aβ1-42 with APOE-4 load.
APOE status was available for 26 CJD patients. Two patients showed an APOE-2/3 type, 17 patients an APOE-3/3 type, two patients an APOE-2/4 status, four patients an APOE-3/4 status, and two patients an APOE-4/4 status. Comparing the Aβ1-42 level of patients carrying at least one APOE-4 allele with the CJD patients carrying no APOE-4 allele showed no difference (p = 0.24).
Comparison of Aβ1-42 with neuropathologic data.
Twenty-four of the 27 CJD patients were neuropathologically verified. All 24 patients had a prion protein type 1 pattern and presented with typical histologic changes for the MM type 1 CJD.28 Five of the 24 patients had additional β-amyloid plaques. Reduction of CSF Aβ1-42 was not pronounced in these five patients as compared with other CJD patients (median 354 pg/mL; range 303 to 686 pg/mL).
Comparison of Aβ1-42 with 14-3-3 immunoblot.
14-3-3 immunoblot was positive in all but two CJD patients. The two 14-3-3 immunoblot negative patients had levels of 268 pg/mL (APOE-4/4) and 352 pg/mL (APOE-3/3). A positive 14-3-3 immunoblot was also present in two patients of the other dementias group. One patient had MS and had an Aβ1-42 level of 344 pg/mL. The second patient had an Aβ1-42 level of 852 pg/mL and had epilepsy. APOE typing was not available on these patients.
Discussion.
Several studies have described decreased levels of Aβ1-42 in CSF of patients with AD.21-25 Our study confirms this finding. A possible explanation for this phenomenon is a supposed disturbance of Aβ clearance in CSF of patients with AD, as Aβ1-42 might be sampled in the plaques of AD patients.21 However, single cases of decreased levels in nonplaque-forming diseases (bacterial meningitis and subacute sclerosing panencephalitis) have been reported25; this may give rise to doubts concerning this theory. Although there are several similarities in the pathophysiology and symptomatology of CJD and AD, β-amyloid plaque formation is not a typical event in CJD. For this reason, we investigated Aβ1-42 in CSF of patients with CJD and AD in order to find an additional marker that will help differentiate between these diseases. We found decreased levels of Aβ1-42 in patients with CJD. Levels in CJD patients were significantly different compared with several of our control groups (other dementias and NDC) but not with AD. Typical β-amyloid plaques that might explain this finding were not seen in most of our patients with CJD. A coimmunostaining of prion protein plaques by anti–β-amyloid antibodies is only described for the Gerstmann-Straussler syndrome, whereas such a coimmunostaining is not observed in sporadic CJD.34
A decreased production of Aβ1-42 and release or enhanced clearance into blood or both35 seems unlikely. An alternative explanation might be that in some pathologic conditions a fraction of Aβ1-42 cannot be detected by conventional ELISA methods as epitopes may be masked. This alternative explanation is supported by findings of Pitschke et al., who detected supramolecular Aβ aggregates in CSF of patients with AD using fluorescence correlation spectroscopy.36 Alternatively, a fraction of Aβ1-42 might bind with high affinity to other CSF proteins,37,38 which may be elevated under certain pathologic conditions. The presence of a large pool of Aβ1-42 not accessible to monoclonal antibodies was recently discovered in human blood.39 However, according to our findings in CSF of CJD patients, this multimer or complex formation is not restricted to AD solely and can be influenced by other pathologic conditions.
In contrast to AD, most studies on APOE-4 occurrence in CJD do not find a susceptibility role of this allele for developing CJD.40,41 It is also described that the APOE-4 load determines the level of Aβ1-42 in CSF. We could not find such a relation in our group of CJD patients. However, this result might be biased due to the low number of patients with an APOE-4 allele in our study (n = 8). The low number of non-CJD patients investigated with a positive 14-3-3 immunoblot make it difficult to determine whether a positive 14-3-3 immunoblot may predict a low Aβ1-42. As one patient with a positive 14-3-3 immunoblot presented with a relatively high Aβ1-42, however, this does not seem very likely.
We suggest that decreased levels of Aβ1-42 in CSF are due to an increased shift of Aβ1-42 into a pool that is not accessible to monoclonal antibodies (Wiltfang et al., article in preparation, 2000). This fraction of Aβ1-42 may be generated by formation of supramolecular autoaggregates or high affinity binding to a carrier.
Acknowledgments
Supported in part by grants from the Bundesministerium für Gesundheit (M.O., S.P.) and VerUm Stiftung (J.W., M.O., S.P., J.K.).
Acknowledgment
The authors thank all physicians notifying suspect cases to the German CJD surveillance unit for providing pertinent clinical, neuroradiologic, and biochemical data. They gratefully acknowledge the help of physicians Dr. Maria Seipelt, Kati Weidehaas, Dr. Karsten Henkel, Dr. Henriette Tschampa, and Dr. Bernhard Steinhoff, and the technical assistance of Sabine Paul, Barbara Ciesielzcyk, and Monika Bodemer. Aβ1-42 kits were obtained from Innogenetics, Belgium.
- Received September 10, 1999.
- Accepted December 10, 1999.
References
Letters: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Hemiplegic Migraine Associated With PRRT2 Variations A Clinical and Genetic Study
Dr. Robert Shapiro and Dr. Amynah Pradhan