Alzheimer's cerebrospinal biomarkers from Lumipulse fully automated immunoassay: concordance with amyloid-beta PET and manual immunoassay in Koreans : CSF AD biomarkers measured by Lumipulse in Koreans

Background

Alzheimer’s disease (AD) cerebrospinal fluid (CSF) biomarker cutoffs from immunoassays with low interlaboratory variability in diverse ethnic groups are necessary for their use in clinics and clinical trials. With lack of cutoffs from fully automated immunoassay platforms in diverse races, the aim of this study is to evaluate the clinical utility of CSF AD biomarkers from the Lumipulse fully automated immunoassay based on β-amyloid (Aβ) positron emission tomography (PET) status comparing with these from two manual immunoassays, in Koreans.

Methods

Among 331 Korean participants enrolled from a prospective, 3-year longitudinal observational study of the validation cohort of Korean Brain Aging Study for the Early Diagnosis and Prediction of AD, 139 (29 CN, 58 SCD, 29 MCI, and 23 AD) provided CSF and 271 underwent baseline amyloid PET (n = 128 with overlapping CSF and Aβ-PET, and 143 without CSFs). Three annual cognitive and neuropsychiatric function tests were conducted. Aβ42, Aβ40, total-tau, and phosphorylated-tau₁₈₁ were measured by Lumipulse fully automated immunoassay and two manual immunoassays (INNO-BIA AlzBio3, INNOTEST). Clinical utility of CSF biomarker cutoffs, based on 128 participants with Aβ-PET, was evaluated.

Results

Cognitive and neuropsychological scores differed significantly among the groups, with descending performance among CN>SCD>MCI>AD. Biomarker levels among immunoassays were strongly intercorrelated. We determined the Aβ-PET status in a subgroup without CSF (n = 143), and then when we applied CSF biomarker cutoffs determined based on the Aβ-PET status, the CSF biomarkers (cutoffs of 642.1 pg/mL for Aβ42, 0.060 for Aβ42/Aβ40, 0.315 for t-tau/Aβ42, and 0.051 for p-tau/Aβ42, respectively) showed good agreement with Aβ-PET (overall AUC ranges of 0.840–0.898). Use of the Aβ-PET-based CSF cutoffs showed excellent diagnostic discrimination between AD and CN (Aβ42, Aβ42/Aβ40, t-tau/Aβ42, and p-tau/Aβ42) with overall AUC ranges of 0.876–0.952. During follow-up, participants with AD-like CSF signature determined by Aβ-PET-based cutoffs from Lumipulse showed rapid progression of cognitive decline in 139 subjects, after adjustment for potential confounders, compared with those with a normal CSF signature.

Conclusion

CSF AD biomarkers measured by different immunoassay platforms show strong intercorrelated agreement with Aβ-PET in Koreans. The Korean-specific Aβ-PET-based CSF biomarker cutoffs measured by the Lumipulse assay strongly predicts progression of cognitive decline. The clinical utility of CSF biomarkers from fully-automated immunoassay platforms should be evaluated in larger, more diverse cohorts.

First amyloid β1-42 certified reference material for re-calibrating commercial immunoassays

INTRODUCTION

Reference materials based on human cerebrospinal fluid were certified for the mass concentration of amyloid beta (Aβ)1-42 (Aβ₄₂ ). They are intended to be used to calibrate diagnostic assays for Aβ₄₂ .

METHODS

The three certified reference materials (CRMs), ERM-DA480/IFCC, ERM-DA481/IFCC and ERM-DA482/IFCC, were prepared at three concentration levels and characterized using isotope dilution mass spectrometry methods. Roche, EUROIMMUN, and Fujirebio used the three CRMs to re-calibrate their immunoassays.

RESULTS

The certified Aβ₄₂ mass concentrations in ERM-DA480/IFCC, ERM-DA481/IFCC, and ERM-DA482/IFCC are 0.45, 0.72, and 1.22 μg/L, respectively, with expanded uncertainties (k = 2) of 0.07, 0.11, and 0.18 μg/L, respectively. Before re-calibration, a good correlation (Pearson's r > 0.97), yet large biases, were observed between results from different commercial assays. After re-calibration the between-assay bias was reduced to < 5%.

DISCUSSION

The Aβ₄₂ CRMs can ensure the equivalence of results between methods and across platforms for the measurement of Aβ₄₂ .

Cerebrospinal fluid growth-associated protein 43 in multiple sclerosis

Neurodegeneration in multiple sclerosis (MS) correlates with disease progression and reparative processes may be triggered. Growth-associated protein 43 (GAP-43) exhibits induced expression during axonal growth and reduced expression during MS progression. We aimed to evaluate if GAP-43 can serve as a biomarker of regeneration in relapsing-remitting MS (RRMS) and whether disease-modifying therapies (DMTs) influence GAP-43 concentration in cerebrospinal fluid (CSF). GAP-43 was measured using an enzyme-linked immunosorbent assay in 105 MS patients (73 RRMS, 12 primary progressive MS, 20 secondary progressive MS) and 23 healthy controls (HCs). In 35 of the patients, lumbar puncture, clinical assessment, and magnetic resonance imaging was performed before initiation of therapeutic intervention, and at follow-up. CSF GAP-43 concentration was significantly lower in progressive MS compared with HCs (p = 0.004) and RRMS (p =  < 0.001) and correlated negatively with disability (p = 0.026). However, DMTs did not alter CSF GAP-43. Interestingly, in RRMS CSF GAP-43 levels were higher in patients with signs of active inflammatory disease than in patients in remission (p = 0.042). According to CSF GAP-43 concentrations, regeneration seems reduced in progressive MS, increased during disease activity in RRMS but is unaffected by treatment of highly active DMTs.

Transient increase in CSF GAP-43 concentration after ischemic stroke

BACKGROUND

Cerebrospinal fluid (CSF) biomarkers reflect ongoing processes in the brain. Growth-associated protein 43 (GAP-43) is highly upregulated in brain tissue shortly after experimental ischemia suggesting the CSF GAP-43 concentration may be altered in ischemic brain disorders. CSF GAP-43 concentration is elevated in Alzheimer's disease patients; however, patients suffering from stroke have not been studied previously.

METHODS

The concentration of GAP-43 was measured in longitudinal CSF samples from 28 stroke patients prospectively collected on days 0-1, 2-4, 7-9, 3 weeks, and 3-5 months after ischemia and cross-sectionally in 19 controls. The stroke patients were clinically evaluated using a stroke severity score system. The extent of the brain lesion, including injury size and degrees of white matter lesions and atrophy were evaluated by CT and magnetic resonance imaging.

RESULTS

Increased GAP-43 concentration was detected from day 7-9 to 3 weeks after stroke, compared to day 1-4 and to levels in the control group (P = 0.02 and P = 0.007). At 3-5 months after stroke GAP-43 returned to admission levels. The initial increase in GAP-43 during the nine first days was associated to stroke severity, the degree of white matter lesions and atrophy and correlated positively with infarct size (r_s = 0.65, P = 0.001).

CONCLUSIONS

The transient increase of CSF GAP-43 is important to take into account when used as a biomarker for other neurodegenerative diseases such as Alzheimer's disease. Furthermore, GAP-43 may be a marker of neuronal responses after stroke and additional studies confirming the potential of CSF GAP-43 to reflect severity and outcome of stroke in larger cohorts are warranted.

Assessing the commutability of reference material formats for the harmonization of amyloid-β measurements

The cerebrospinal fluid (CSF) amyloid-β (Aβ42) peptide is an important biomarker for Alzheimer’s disease (AD). Variability in measured Aβ42 concentrations at different laboratories may be overcome by standardization and establishing traceability to a reference system. Candidate certified reference materials (CRMs) are validated herein for this purpose.

Commutability of 16 candidate CRM formats was assessed across five CSF Aβ42 immunoassays and one mass spectrometry (MS) method in a set of 48 individual clinical CSF samples. Promising candidate CRM formats (neat CSF and CSF spiked with Aβ42) were identified and subjected to validation across eight (Elecsys, EUROIMMUN, IBL, INNO-BIA AlzBio3, INNOTEST, MSD, Simoa, and Saladax) immunoassays and the MS method in 32 individual CSF samples. Commutability was evaluated by Passing-Bablok regression and the candidate CRM termed commutable when found within the prediction interval (PI). The relative distance to the regression line was assessed.

The neat CSF candidate CRM format was commutable for almost all method comparisons, except for the Simoa/MSD, Simoa/MS and MS/IBL where it was found just outside the 95% PI. However, the neat CSF was found within 5% relative distance to the regression line for MS/IBL, between 5% and 10% for Simoa/MS and between 10% and 15% for Simoa/MSD comparisons.

The neat CSF candidate CRM format was commutable for 33 of 36 method comparisons, only one comparison more than expected given the 95% PI acceptance limit. We conclude that the neat CSF candidate CRM can be used for value assignment of the kit calibrators for the different Aβ42 methods.

The influence of matrix type, diurnal rhythm and sample collection and processing on the measurement of plasma beta-amyloid isoforms using the INNO-BIA plasma Abeta forms multiplex assay

OBJECTIVE

The aim of the study was to determine the extent to which plasma matrix types, diurnal rhythm and sample collection and processing procedures contribute to overall variability of measurements with the INNO-BIA plasma Abeta forms assay.

METHODS

Plasma samples from healthy volunteers were collected at BARC-CRI. Analyte concentrations from various plasma matrix types (EDTA, heparin, fluoride) were compared to serum after collection of blood in commercial plastic and glass tubes. Sample processing variables including time and temperature before and after centrifugation, centrifugal force and plasma dilution factor were also investigated. Diurnal variability in plasma Abeta isoforms was determined in 29 healthy volunteers by analysis of EDTA plasma specimens serially collected over 24 hours and stored frozen following oral administration of a placebo treatment. All plasma samples from a given individual and experiment were analyzed in a single analytical run.

RESULTS

Highest Abeta levels were obtained using EDTA-plasma samples (in contrast to serum, heparin, citrate, or fluoride). Addition of aprotinin to EDTA plasma had no effect on Abeta peptide recovery. The elapsed time and temperature exposure, before and after sample processing affects the recovery of Abeta isoforms. Analyte recovery was not significantly affected by the presence of platelets in plasma samples. At the subject level, analysis of serially collected EDTA plasma specimens from healthy volunteers revealed no evidence of diurnal variation in any of the Abeta isoforms investigated and results from samples collected on a monthly basis showed only very limited intra-individual variation.

CONCLUSIONS

Optimal recovery of Abeta peptides was obtained from blood drawn into EDTA tubes and processed within 4 h. Plasma that was refrigerated after separation and analysed within 4 h gave comparable results to samples immediately processed and frozen at -70 degrees C.

Amino-Truncated β-Amyloid42 Peptides in Cerebrospinal Fluid and Prediction of Progression of Mild Cognitive Impairment

Background: Early identification of patients with mild cognitive impairment (MCI) progressing to Alzheimer disease (MCI-AD) by use of biomarkers in cerebrospinal fluid (CSF) is an essential step toward improving clinical diagnosis and drug development. We evaluated whether different β-amyloid42 (Aβ42) peptides can add further information to the combined use of tau and Aβ1–42 for predicting risk of progression of MCI to AD.

Methods: We used xMAP® technology to simultaneously quantify different Aβ42 peptides modified at the amino terminus, tau, and phosphorylated tau (P-tau181P) in CSF. Aβ42 peptide concentrations were measured by use of immunoreactivity toward Aβ monoclonal antibodies [3D6 (Aβ42-3D6), WO2 (Aβ42-WO2), 6E10 (Aβ42-6E10), and 4G8 (Aβ42-4G8)]. The discriminant ability of the markers was evaluated by ROC curve analysis.

Results: The areas under the curves for the separation of MCI-AD from nonprogressing MCI (MCI-N) were significantly higher when we used Aβ42-3D6/Aβ42-WO2, Aβ42-3D6/Aβ42-6E10, or Aβ42-3D6/Aβ42-4G8 compared with Aβ42-3D6. In addition, differentiation of MCI-N from MCI-AD was improved by quantification of full-length Aβ1–42 (Aβ42-3D6) compared with Aβ42-WO2, Aβ42-6E10, or Aβ42-4G8. Several Aβ42 peptides truncated at the amino terminus (Aβ11–42 and Aβ8–42) were identified in CSF by surface-enhanced laser desorption/ionization time-of-flight technology.

Conclusion: The CSF markers tau, Aβ42 forms, and P-tau181P, when used as adjuncts to clinical diagnosis, have the potential to help identify AD pathology and could be a valuable asset for early AD diagnosis.

Phosphorylation of amyloid precursor carboxy-terminal fragments enhances their processing by a gamma-secretase-dependent mechanism

In Alzheimer's disease, the complex catabolism of amyloid precursor protein (APP) leads to the production of amyloid-beta (Abeta) peptide, the major component of amyloid deposits. APP is cleaved by beta- and alpha-secretases to generate APP carboxy-terminal fragments (CTFs). Abeta peptide and amyloid intracellular domain are resulting from the cleavage of APP-CTFs by the gamma-secretase. In the present study, we hypothesize that post-translational modification of APP-CTFs could modulate their processing by the gamma-secretase. Inhibition of the gamma-secretase was shown to increase the total amount of APP-CTFs. Moreover, we showed that this increase was more marked among the phosphorylated variants and directly related to the activity of the gamma-secretase, as shown by kinetics analyses. Phosphorylated CTFs were shown to associate to presenilin 1, a major protein of the gamma-secretase complex. The phosphorylation of CTFs at the threonine 668 resulting of the c-Jun N-terminal kinase activation was shown to enhance their degradation by the gamma-secretase. Altogether, our results demonstrated that phosphorylated CTFs can be the substrates of the gamma-secretase and that an increase in the phosphorylation of APP-CTFs facilitates their processing by gamma-secretase.

Truncated beta-amyloid peptide species in pre-clinical Alzheimer's disease as new targets for the vaccination approach

Vaccination against human beta-amyloid peptide (A beta) has been shown to remove the amyloid burden produced in transgenic mice overexpressing the mutated human amyloid precursor protein (APP) gene. For human beings, the efficiency of this therapeutic strategy has to take into account the specificities of human amyloid, especially at the early stages of 'sporadic' Alzheimer's disease (AD). A beta 40/42 were previously quantified in tissues from our well-established brain bank, including non-demented individuals with both mild amyloid and tau pathologies, hence corresponding to the earliest stages of Alzheimer pathology. Herein, we have adapted a proteomic method combined with western blotting and mass spectrometry for the characterization of insoluble A beta extracted in pure-formic acid. We demonstrated that amino-truncated A beta species represented more than 60% of all A beta species, not only in full blown AD, but also, and more interestingly, at the earliest stage of Alzheimer pathology. At this stage, A beta oligomers were exclusively made of A beta-42 species, most of them being amino-truncated. Thus, our results strongly suggest that amino-truncated A beta-42 species are instrumental in the amyloidosis process. In conclusion, a vaccine specifically targeting these pathological amino-truncated species of A beta-42 are likely to be doubly beneficial, by inducing the production of specific antibodies against pathological A beta products that are, in addition, involved in the early and basic mechanisms of amyloidosis in humans.

Identification of a thiosulfate utilization gene cluster from the green phototrophic bacterium Chlorobium limicola

Chlorobium is an autotrophic, green phototrophic bacterium which uses reduced sulfur compounds to fix carbon dioxide in the light. The pathways for the oxidation of sulfide, sulfur, and thiosulfate have not been characterized with certainty for any species of bacteria. However, soluble cytochrome c-551 and flavocytochrome c (FCSD) have previously been implicated in the oxidation of thiosulfate and sulfide on the basis of enzyme assays in Chlorobium. We have now made a number of observations relating to the oxidation of reduced sulfur compounds. (1) Western analysis shows that soluble cytochrome c-551 in Chlorobium limicola is regulated by thiosulfate, consistent with a role in the utilization of thiosulfate. (2) A membrane-bound flavocytochrome c-sulfide dehydrogenase (which is normally a soluble protein in other species) is constitutive and not regulated by sulfide as expected for an obligately autotrophic species dependent upon sulfide. (3) We have cloned the cytochrome c-551 gene from C. limicola and have found seven other genes, which are also presumably involved in sulfur metabolism and located near that for cytochrome c-551 (SoxA). These include genes for a flavocytochrome c flavoprotein homologue (SoxF2), a nucleotidase homologue (SoxB), four small proteins (including SoxX, SoxY, and SoxZ), and a thiol-disulfide interchange protein homologue (SoxW). (4) We have established that the constitutively expressed FCSD genes (soxEF1) are located elsewhere in the genome. (5) Through a database search, we have found that the eight thiosulfate utilization genes are clustered in the same order in the Chlorobium tepidum genome (www.tigr.org). Similar thiosulfate utilization gene clusters occur in at least six other bacterial species but may additionally include genes for rhodanese and sulfite dehydrogenase.

A novel system for heterologous expression of flavocytochrome c in phototrophic bacteria using the Allochromatium vinosum rbcA promoter

Flavocytochrome c-sulfide dehydrogenase (FCSD), an enzyme that catalyzes the reversible conversion of sulfide to elemental sulfur in vitro, is common to bacteria that utilize reduced sulfur compounds as electron donors in the process of carbon dioxide fixation. FCSD is a heterodimer containing two different cofactors, a flavin (FAD) and one or two heme c groups, located on the separate protein subunits. Efforts to produce the holoproteins of the soluble Allochromatium vinosum FCSD and the membrane-bound Ectothiorhodospira vacuolata protein in Escherichia coli using several expression systems were unsuccessful. Although all systems used were able to export the recombinant FCSDs to the periplasm, the proteins did not incorporate heme. In order to develop a new expression system involving photosynthetic hosts (Rhodobacter capsulatus, Rhodobacter sphaeroides and Ect. vacuolata), plasmid mobilisation from E. coli donors was studied. In the search for efficient promoters for such hosts, a system was developed combining the broad-host-range plasmid pGV910 and the promoter of the A. vinosum RuBisCo gene, rbcA. Conjugation was used to enable transfer from the expression plasmid of E. coli into Rba. capsulatus, Rba. sphaeroides strains and into Ect. vacuolata. Both Rhodobacter hosts were able to transcribe the genes coding for FCSD from the rbcA promoter and to produce detectable amounts of recombinant FCSD holoprotein. Western blots showed that the best production was obtained from cells grown photosynthetically on malate or acetate with sulfide. This system may prove to be of general use for the production of recombinant c-type cytochromes in homologous or related host systems.

A membrane-bound flavocytochrome c-sulfide dehydrogenase from the purple phototrophic sulfur bacterium Ectothiorhodospira vacuolata

The amino acid sequence of Ectothiorhodospira vacuolata cytochrome c-552, isolated from membranes with n-butanol, shows that it is a protein of 77 amino acid residues with a molecular mass of 9,041 Da. It is closely related to the cytochrome subunit of Chlorobium limicola f. sp. thiosulfatophilum flavocytochrome c-sulfide dehydrogenase (FCSD), having 49% identity. These data allowed isolation of a 5.5-kb subgenomic clone which contains the cytochrome gene and an adjacent flavoprotein gene as in other species which have an FCSD. The cytochrome subunit has a signal peptide with a normal cleavage site, but the flavoprotein subunit has a signal sequence which suggests that the mature protein has an N-terminal cysteine, characteristic of a diacyl glycerol-modified lipoprotein. The membrane localization of FCSD was confirmed by Western blotting with antibodies raised against Chromatium vinosum FCSD. When aligned according to the three-dimensional structure of Chromatium FCSD, all but one of the side chains near the flavin are conserved. These include the Cys 42 flavin adenine dinucleotide binding site; the Cys 161-Cys 337 disulfide; Glu 167, which modulates the reactivity with sulfite; and aromatic residues which may function as charge transfer acceptors from the flavin-sulfite adduct (C. vinosum numbering). The genetic context of FCSD is different from that in other species in that flanking genes are not conserved. The transcript is only large enough to encode the two FCSD subunits. Furthermore, Northern hybridization showed that the production of E. vacuolata FCSD mRNA is regulated by sulfide. All cultures that contained sulfide in the medium had elevated levels of FCSD RNA compared with cells grown on organics (acetate, malate, or succinate) or thiosulfate alone, consistent with the role of FCSD in sulfide oxidation.

Structure and characterization of Ectothiorhodospira vacuolata cytochrome b(558), a prokaryotic homologue of cytochrome b(5)

A soluble cytochrome b(558) from the purple phototropic bacterium Ectothiorhodospira vacuolata was completely sequenced by a combination of automated Edman degradation and mass spectrometry. The protein, with a measured mass of 10,094.7 Da, contains 90 residues and binds a single protoheme. Unexpectedly, the sequence shows homology to eukaryotic cytochromes b(5). As no prokaryotic homologue had been reported so far, we developed a protocol for the expression, purification, and crystallization of recombinant cytochrome b(558). The structure was solved by molecular replacement to a resolution of 1.65 A. It shows that cytochrome b(558) is indeed the first bacterial cytochrome b(5) to be characterized and differs from its eukaryotic counterparts by the presence of a disulfide bridge and a four-residue insertion in front of the sixth ligand (histidine). Eukaryotes contain a variety of b(5) homologues, including soluble and membrane-bound multifunctional proteins as well as multidomain enzymes such as sulfite oxidase, fatty-acid desaturase, nitrate reductase, and lactate dehydrogenase. A search of the Mycobacterium tuberculosis genome showed that a previously unidentified gene encodes a fatty-acid desaturase with an N-terminal b(5) domain. Thus, it may provide another example of a bacterial b(5) homologue.

Two new aminopeptidases from Ochrobactrum anthropi active on D-alanyl-p-nitroanilide

Two new enzymes which hydrolyse D-alanyl-p-nitroanilide have been detected in Ochrobactrum anthropi LMG7991 extracts. The first enzyme, DmpB, was purified to homogeneity and found to be homologous to the Dap protein produced by O. anthropi SCRC C1-38 (ATCC49237). The second enzyme, DmpA, exhibits a similar substrate profile when tested on p-nitroanilide derivatives of glycine and L/D-alanine, but the amounts produced by the Ochrobactrum strain were not sufficient to allow complete purification. Interestingly, the DmpA preparation also exhibited an L-aminopeptidase activity on the tripeptide L-Ala-Gly-Gly but it was not possible to be certain that the same protein was responsible for both p-nitroanilide and peptide hydrolysing activities. The gene encoding the DmpA protein was cloned and sequenced. The deduced protein sequence exhibits varying degrees of similarity with those corresponding to several open reading frames found in the genomes of other prokaryotic organisms, including Mycobacteria. None of these gene products has been isolated or characterised, but a tentative relationship can be proposed with the NylC amidase from Flavobacterium sp. K172.