Normal Human Cerebrospinal Fluid Research Articles

Highly-sensitive simultaneous quantitation of glucosylsphingosine and galactosylsphingosine in human cerebrospinal fluid by liquid chromatography/tandem mass spectrometry

Mutations in the GBA gene, encoding glucocerebrosidase (GCase), are linked to Gaucher disease (GD) and are the most common risk factors for Parkinson’s disease (PD). The glucosylsphingosine (GlcSph) in cerebrospinal fluid (CSF) is used as a pharmacodynamic marker for GCase functionalizing therapy in GD patients. Its isobaric structural isomer, galactosylsphingosine (GalSph, psychosine), is also used as a diagnostic blood marker in Krabbe disease (KD) which is caused by a deficiency in β-galactocerebrosidase (GALC). However, there are no reports of GlcSph quantification in the CSF of GBA-PD patients and normal healthy humans due to low concentrations. In this study, we successfully quantified GlcSph in healthy human CSF using a highly sensitive LC-MS/MS method with separation of GalSph. The lower limit of quantitation (LLOQ) was 0.1 pg/mL. Additionally, GlcSph and GalSph concentrations in the plasma and brain were determined using different LC-MS/MS methods. The mean concentrations of GlcSph and GalSph in normal human CSF were 1.07 and 9.44 pg/mL, respectively. The GalSph level in the CSF and brain was higher than that of GlcSph, whereas plasma GalSph was lower than GlcSph. Because GCase and GALC are expressed in the brain and the peripheral tissues, GlcSph and GalSph in CSF would be a good surrogate of concentration change in the brain by targeted therapies. This method measures normal levels of GlcSph and GalSph in healthy human CSF without accumulation of sphingolipids, and confirms whether abnormal CSF concentrations can be reduced to normal levels by therapy.

Identification of Missing Proteins in Normal Human Cerebrospinal Fluid.

The cerebrospinal fluid (CSF) proteome data set presented herein was obtained after immunodepletion of abundant proteins and off-gel electrophoresis fractionation of a commercial pool of normal human CSF; liquid chromatography tandem mass spectrometry analysis was performed with a linear ion trap-Orbitrap Elite. We report the identification of 12 344 peptides mapping on 2281 proteins. In the context of the Chromosome-centric Human Proteome Project (C-HPP), the existence of seven missing proteins is proposed to be validated. This data set is available to the ProteomeXchange Consortium ( http://www.proteomexchange.org/ ) with the data set identifier PXD008029.

Data for a comprehensive map and functional annotation of the human cerebrospinal fluid proteome.

Knowledge about the normal human cerebrospinal fluid (CSF) proteome serves as a baseline reference for CSF biomarker discovery and provides insight into CSF physiology. In this study, high-pH reverse-phase liquid chromatography (hp-RPLC) was first integrated with a TripleTOF 5600 mass spectrometer to comprehensively profile the normal CSF proteome. A total of 49,836 unique peptides and 3256 non-redundant proteins were identified. To obtain high-confidence results, 2513 proteins with at least 2 unique peptides were further selected as bona fide CSF proteins. Nearly 30% of the identified CSF proteins have not been previously reported in the normal CSF proteome. More than 25% of the CSF proteins were components of CNS cell microenvironments, and network analyses indicated their roles in the pathogenesis of neurological diseases. The top canonical pathway in which the CSF proteins participated was axon guidance signaling. More than one-third of the CSF proteins (788 proteins) were related to neurological diseases, and these proteins constitute potential CSF biomarker candidates. The mapping results can be freely downloaded at http://122.70.220.102:8088/csf/, which can be used to navigate the CSF proteome. For more information about the data, please refer to the related original article [1], which has been recently accepted by Journal of Proteomics.

Establishing the Proteome of Normal Human Cerebrospinal Fluid

BackgroundKnowledge of the entire protein content, the proteome, of normal human cerebrospinal fluid (CSF) would enable insights into neurologic and psychiatric disorders. Until now technologic hurdles and access to true normal samples hindered attaining this goal.Methods and Principal FindingsWe applied immunoaffinity separation and high sensitivity and resolution liquid chromatography-mass spectrometry to examine CSF from healthy normal individuals. 2630 proteins in CSF from normal subjects were identified, of which 56% were CSF-specific, not found in the much larger set of 3654 proteins we have identified in plasma. We also examined CSF from groups of subjects previously examined by others as surrogates for normals where neurologic symptoms warranted a lumbar puncture but where clinical laboratory were reported as normal. We found statistically significant differences between their CSF proteins and our non-neurological normals. We also examined CSF from 10 volunteer subjects who had lumbar punctures at least 4 weeks apart and found that there was little variability in CSF proteins in an individual as compared to subject to subject.ConclusionsOur results represent the most comprehensive characterization of true normal CSF to date. This normal CSF proteome establishes a comparative standard and basis for investigations into a variety of diseases with neurological and psychiatric features.

Perfusion fluids used in neurosurgery affect cerebrospinal fluid and surrounding brain parenchyma in the rat ventriculocisternal perfusion model

ARTCEREB, an irrigation and perfusion solution (Artcereb), is a preparation intended for the irrigation and perfusion of the cerebral ventricles, and it is therefore important to evaluate its effects on cerebrospinal fluid (CSF) and on the surrounding cerebrospinal parenchyma. To confirm the kinetics of the perfusion fluid component, we performed whole body autoradiography and examined glucose balance during ventriculocisternal perfusion with (14)C-glucose labeled Artcereb in the rat model, which simulates ventricular irrigation or ventriculocisternal perfusion in clinical neurosurgery. We also performed ventriculocisternal perfusion with Artcereb, lactated Ringer's solution, or normal saline, and observed the effect of these solutions on animal condition and on brain tissue morphology. In the kinetic study, diffusion of (14)C-glucose from the perfused Artcereb to the cerebrospinal tissue was seen on whole body autoradiography, and almost 90% of the glucose in the perfusion fluid was distributed to the cerebrospinal tissue and the systemic circulation. These data indicated that the perfusion fluid interacted actively with the CSF, surrounding parenchyma and the systemic circulation, and suggested that the formation of perfusion fluid affected CSF composition and cerebrospinal tissue functions. Animals perfused with normal saline were associated with serious symptoms including tonic convulsions and death, and exhibited neuronal death in the cerebrum. However, these severe changes were not observed in animals perfused with Artcereb or lactated Ringer's solution. We therefore propose that during neurosurgery, it is extremely important to use a physiological solution like Artcereb which closely resembles normal human CSF, in order to maintain cerebrospinal function and to alleviate postoperative adverse events.

Solid-phase extraction enhances detection of beta-amyloid peptides in plasma and enables Aβ quantification following passive immunization with Aβ antibodies

We have previously developed a solid-phase extraction (SPE) procedure to enable the detection of beta-amyloid (Aβ) peptides in brain tissue from non-transgenic animals. We have now adapted these methods to enrich the Aβ fraction in cerebrospinal fluid (CSF) and plasma. Human CSF and plasma and Tg2576 mouse plasma were subjected to guanidine denaturation followed by SPE in 96-well cassettes. The resulting eluates could be concentrated significantly to enhance detection of low-abundance Aβ peptides by immunoassay. The concentrated eluates diluted in a linear fashion with consistent recovery between SPE columns. This technique was therefore used to facilitate quantification of Aβ1-X, 1-40, 1-42, and 1-38 peptides in normal human CSF and plasma samples. SPE sample preparation was also applied to the plasma of mice dosed peripherally with a monoclonal antibody raised against Aβ. When such samples were assayed directly, the presence of the systemically administered antibody interfered with the subsequent immunoassay, by preventing detection of antibody-bound Aβ. After subjecting plasma from antibody-treated animals to denaturation and SPE, the antibody–antigen complex was disrupted, and the Aβ fraction could be isolated from the antibody-containing fraction. Application of this method allowed for detection of a 100-fold increase in plasma Aβ1-40 following treatment of Tg2576 mice or wild type littermate control mice with Aβ40-specific monoclonal antibody 9TL. Given the availability of a variety of SPE matrices, we hypothesize that these methods could facilitate plasma antigen retrieval using multiple therapeutic antibody approaches.

Candidate gene analysis of SPARCL1 gene in patients with multiple sclerosis

Recently, proteomic analysis in cerebrospinal fluid (CSF) from patients with MS identified four proteins which are present in MS but not in normal human CSF, including SPARCL1, an extracellular matrix-associated protein member of the SPARC family. One hundred eighty-six patients with MS and 185 age-matched controls were genotyped for A/G single nucleotide polymorphism (SNP) in exon 1 (rs1049539), C/G SNP in exon 4 (rs1049544), resulting in a substitution of an aspartate with an histidine, and A/G substitution in the exon 5 (rs1130643), leading to the substitution of alanine with threonine. No significant differences in either allelic or genotypic frequency of the three SNPs were found (P>0.05), even in stratifying MS patients according to the course of the disease. Stratifying according to gender, a trend towards a decreased frequency of the C/C genotype of the rs1049544 was observed in male patients as compared with male controls (30.2% versus 44.0%; P=0.217). Despite proteomic studies in CSF from MS patients suggested an important role for SPARCL1 in the development of the disease, SPARCL1 gene does not appear to act as susceptibility factor for MS in the population investigated here. However, the frequency of the C/C genotype of rs1049544 was decreased in male patients, possibly conferring a lower risk of developing MS in male population. Further studies are needed to clarify this issue.

An investigation into the lipid-binding properties of α-, β- and γ-synucleins in human brain and cerebrospinal fluid

Parkinson's disease (PD) and dementia with Lewy bodies (DLB) are both characterized by the formation and intraneuronal accumulation of fibrillar aggregates of α-synuclein (α-syn) protein in affected brain regions. α-Syn has biochemical properties and a structural motif characteristic of fatty acid binding proteins. Using the fatty acid binding resin Lipidex-1000, we investigated the capture of α-, β-, and γ-syn proteins as lipid-associated proteins from normal and DLB brain lysates, and from normal human cerebrospinal fluid (CSF). These were eluted from Lipidex-1000 and analyzed by SDS–NuPAGE followed by Western blotting. Using this methodology, we have been able to extract full-length and truncated forms of α-syn from brain lysates. We also extracted low levels of β-syn from DLB brains, but failed to extract any γ-syn. We were able to capture only full-length monomeric α-syn from normal human CSF. Our data confirm the fatty acid binding properties of α-syn, and to a lesser extent β-syn, but suggest that γ-syn does not share this same characteristic.

Proteomic analysis of multiple sclerosis cerebrospinal fluid.

Two-dimensional gel electrophoresis and peptide mass fingerprinting were used to identify proteins in cerebrospinal fluid (CSF) pooled from three patients with multiple sclerosis (MS) and in CSF pooled from three patients with non-MS inflammatory central nervous system (CNS) disorders. Resolution of CSF proteins on three pH gradients (3-10, 4-7 and 6-11) enabled identification of a total of 430 spots in the MS CSF proteome that represented 61 distinct proteins. The gels containing MS CSF revealed 103 protein spots that were not seen on control gels. All but four of these 103 spots were proteins known to be present in normal human CSF. The four exceptions were: CRTAC-IB (cartilage acidic protein), tetranectin (a plasminogen-binding protein), SPARC-like protein (a calcium binding cell signalling glycoprotein), and autotaxin t (a phosphodiesterase). It remains unknown whether these four proteins are related to the cause and pathogenesis of MS.

Remodelling of cerebrospinal fluid lipoproteins after subarachnoid hemorrhage

Lipoprotein particles (Lps) in normal human cerebrospinal fluid (CSF) are distinct from those found in plasma and include unique apolipoprotein E (apoE indicates protein; APOE, gene) containing lipoproteins rarely seen in human plasma. Less favourable neurological recovery after subarachnoid hemorrhage (SAH) has been observed in patients who posses the APOE ε4 allele raising the possibility that apoE influences neuronal survival after brain injury. We analysed Lps from control and SAH CSF testing the hypotheses that following brain injury CSF Lps undergo remodelling and apoE containing Lps are selectively depleted from brain injury CSF. Lipoproteins were fractionated using CSF from six control pools and six patients with SAH on a sepharose 6HR 10/30 size exclusion column. Fractions were assayed for total cholesterol (TC), free cholesterol (FC), phospholipid, triglyceride (TG), apoE, apolipoprotein B (apoB), and apolipoprotein AI (apoAI). Compared to control CSF there were significant ( P<0.05) increases in TC, FC, TG, and apoAI in SAH CSF. Plasma sized apoB-containing lipoproteins and a very small apoAI-containing Lps were identified in the SAH CSF, which were not present in controls. However, despite the release of plasma lipoproteins into the subarachnoid space, there was no significant increase in CSF apoE. These data provide novel indirect evidence suggesting that after SAH CSF Lps undergo remodelling and apoE containing Lps are selectively reduced in brain injury CSF. The remodelling of CSF Lps and selective reduction of apoE containing lipoproteins may reflect an important response of the human brain to injury.

Development of an ultrasensitive enzyme immunoassay for the determination of matrix metalloproteinase-9 (MMP-9) levels in normal human cerebrospinal fluid

Determination of matrix metalloproteinase-9 (MMP-9) in human cerebrospinal fluid (CSF) to study blood–brain barrier impairment and immune cell migration in inflammatory neurological diseases recently became a matter of major interest. Regularly, MMP-9 was determined qualitatively or semi-quantitatively by zymography (gelatin gel electrophoresis) or quantitatively by enzyme immunoassay (EIA). As yet, it was not possible by either method to detect MMP-9 in CSF of controls (patients without pathologically increased CSF parameters). We developed an ultrasensitive two-side enzyme-linked immunosorbent assay (ELISA) which allows for the first time to measure reliably MMP-9 concentrations in CSF of controls. This ELISA uses a monoclonal as capture and a polyclonal as detector antibody. The detection limit of the assay is below 10 pg/ml and the assay range is 15–2000 pg/ml. Intra-assay precision is 2.5% for low and 3.7% for high, inter-assay precision is 11% for low and 10.7% for high values, respectively. The determination of the MMP-9 concentration in 50 control CSF gave the following results: range, 22–146 pg/ml; median, 76 pg/ml. The measurement of native and recombinant MMP-9 was carried out with three commercially available ELISAs, most widely employed in MMP-9 research, and compared to the newly developed one. All ELISAs recognize recombinant MMP-9 by factors of 5–20 less sensitively than native MMP-9.

CONCENTRATIONS OF HEPATOCYTE GROWTH FACTOR IN CEREBROSPINAL FLUID UNDER NORMAL AND DIFFERENT PATHOLOGICAL CONDITIONS

Hepatocyte growth factor (HGF) and its specific receptor, MET, are expressed in the developing and adult mammalian brain. Recent studies have shown a neurotrophic activity of HGF in the nervous system. The present study focused on HGF concentrations in the cerebrospinal fluid (CSF) and serum in normal persons and in different central nervous system (CNS) diseases considering blood–CSF barrier (BCB) function. Concentrations of HGF were analyzed using an enzyme-linked immunosorbent assay (ELISA). HGF was present in normal human CSF (346±126pg/ml) representing approximately half of the HGF serum concentrations. The CSF HGF levels were not significantly changed in chronic CNS disease and in aseptic meningitis (419±71pg/ml), but significantly increased in patients with bacterial meningitis (6101± 5200pg/ml). The HGF levels in CSF were not influenced by increased serum concentrations in patients with normal or mildly affected BCB function. The results show that HGF is present in normal CSF and does not appear to cross the CSF barrier significantly unless it is severely disrupted. So far, strong increases of HGF concentration in CSF are only present in acute bacterial meningitis.

Quantification of free fatty acids in human cerebrospinal fluid.

Free fatty acids (FFA) in cerebrospinal fluid (CSF) are well-recognized markers of brain damage in animal studies. Information is limited regarding human CSF in both normal and pathological conditions. Samples of CSF from 73 patients, who had undergone lumbar puncture for medically indicated reasons, came from a core laboratory upon completion of ordered tests. Using high performance liquid chromatography, mean FFA concentrations (microg/L +/- SEM) were: arachidonic 26.14 +/- 3.44; docosahexaenoic 60.74 +/- 5.70; linoleic 105.07 +/- 10.98; myristic 160.38 +/- 16.17; oleic 127.91 +/- 10.13; and palmitic 638.34 +/- 37.27. No differences in FFA concentrations were seen with gender, race, age, and/or indication for lumbar puncture. This is the first study to document normal human CSF FFA concentrations in a large series. Further characterization of FFA in pathological conditions may provide markers for evaluating clinical treatments and assisting in prognostication of neurological disease.

Formation of Isoprostane-like Compounds (Neuroprostanes) in Vivo from Docosahexaenoic Acid

F2-isoprostanes are prostaglandin F2-like compounds that are formed nonenzymatically by free radical-induced oxidation of arachidonic acid. We explored whether oxidation of docosahexaenoic acid (C22:6omega3), which is highly enriched in the brain, led to the formation of F2-isoprostane-like compounds, which we term F4-neuroprostanes. Oxidation of docosahexaenoic acid in vitro yielded a series of compounds that were structurally established to be F4-neuroprostanes using a number of mass spectrometric approaches. The amounts formed exceeded levels of F2-isoprostanes generated from arachidonic acid by 3.4-fold. F4-neuroprostanes were detected esterified in normal whole rat brain and newborn pig cortex at a level of 7.0 +/- 1.4 ng/g and 13.1 +/- 8 ng/g, respectively. Furthermore, F4-neuroprostanes could be detected in normal human cerebrospinal fluid and levels in patients with Alzheimer's disease (110 +/- 12 pg/ml) were significantly higher than age-matched controls (64 +/- 8 pg/ml) (p < 0.05). F4-neuroprostanes may provide a unique marker of oxidative injury to the brain and could potentially exert biological activity. Furthermore, the formation of F4-neuroprostane-containing aminophospholipids might adversely effect neuronal function as a result of alterations they induce in the biophysical properties of neuronal membranes.

Prothrombin in normal human cerebrospinal fluid originates from the blood.

In spite of the fact that prothrombin is produced by cells within the central nervous system, its presence in the cerebrospinal fluid (CSF) has not been investigated. We determined the concentration of prothrombin in CSF with reference to the concentration in plasma in paired samples from 18 "normal" control patients and 4 patients with relapsing-remitting type of multiple sclerosis (MS). The newly developed ELISA was very specific (no cross-reactivity with thrombin) and sensitive (detection limit--0.7 ng/ml) with an imprecision of CV = 8.3% (intraseries) and 7.0% (interassay). The mean prothrombin concentration in normal CSF was 0.55 mg/l (CV +/- 33%, range: 0.28-0.93 mg/l), in normal plasma 121.8 mg/l +/- 21%, resulting in a mean CSF/plasma concentration quotient (Q(Proth)--4.5 x 10(-3) (CV +/- 35%, range: 2.1-8.3 x 10(-3)) corresponding to a mean albumin quotient in this group of subjects of Q(Alb) = 5.8 x 10(-3). Due to the Q(Proth) and the molecular weight of prothrombin (72 kDa)--similar to that of albumin--we conclude that prothrombin in normal human CSF originates predominantly (>95%) from blood. The enzymatic activity in CSF is conserved. Comparable results obtained in MS patients with only few small MRI lesions suggest that local chronic inflammatory disease of the central nervous system does not influence prothrombin concentration in the CSF if the blood-CSF barrier function is normal.

ALZHEIMER'S SOLUBLE AMYLOID BETA PROTEIN IS SECRETED BY HEPG2 CELLS AS AN APOLIPOPROTEIN

Recently we reported that the soluble form of amyloid beta protein (sAβ) in normal human plasma and cerebrospinal fluid is associated with lipoprotein (LP) particles. In this paper we tested the sAβ secretion by cells in association with LP in the model of the human hepatoma HepG2 cell line. These cells secreted sAβ to the culture media and expressed intracellular sAβ immunoreactivity. Soluble Aβ in the cell supernatant was detected in 200–300kDa LP complexes in association with apoA-I, apoJ, transthyrethin and phospholipids, triglycerides and free and esterified cholesterol. This was assessed by size exclusion HPLC, immunoprecipitation with corresponding antibodies and by analysis of sAβ associated metabolically-labeled lipids, respectively. Our results suggest that sAβ to LP association represents a unique mechanism, governing the normal biology of sAβ.

Generation and characterization of an IGFBP-7 antibody: identification of 31kD IGFBP-7 in human biological fluids and Hs578T human breast cancer conditioned media.

The cDNA encoding mac25 (IGFBP-7) was firsr derived from mRNA isolated from leptomeningial and senescent human mammary epithelial cells (1,2). The open reading frame was shown to predict a protein with homology to the amino terminus of the IGF binding proteins, (IGFBP)1-6. Studies in our laboratory have shown that baculovirus generated mac25 binds IGF-I and-II in a specific manner, leading to the renaming of mac25 as IGFBP-7 (3). Further studies at the cellular level, to identify the involvement of IGFBP-7 in IGF regulation and cell growth, require a specific antibody against the protein, which has yet to be identified in either cultured cells or in vivo. We have now generated three polyclonal antibodies against the purified baculovirus peptide and, by western immunoblots and immunoprecipitation, demonstrated the existence of a specific 31,000 dalton protein. It is a secreted protein, and can be identified in the conditioned media of Hs578T breast cancer cells, as well as in normal human urine, cerebrospinal fluid and amniotic fluid. Subsequent studies with these antibodies should help elucidate the physiological role(s) of this protein.

Molecular characterization of beta-trace protein in human serum and urine: a potential diagnostic marker for renal diseases.

We have isolated beta-trace protein from cerebrospinal fluid, serum, plasma, and urine samples of normal volunteers and sera and hemofiltrate of patients with chronic renal failure. Blood-derived and urinary beta-trace have significantly higher molecular weights than their cerebrospinal fluid counterpart, the amino acid sequences being identical. Oligosaccharide structural analysis revealed these molecular weight differences to be due to different N-glycosylation. beta-Trace from hemofiltrate and urine has larger sugar chains and concurrently significantly higher sialylation than cerebrospinal fluid-beta-trace which bears truncated "brain-type" oligosaccharide chains (published previously). beta-Trace concentrations were about 40 ng/ml for normal sera and plasma. 2000-6000 ng/ml were measured in sera of dialysis patients whereas in normal human cerebrospinal fluid, beta-trace concentration was about 8000 ng/ml. A reduced amount of 900 ng/ml was found in a single case of hydrocephalus cerebri. The sialylated glycoforms of beta-trace detected in the blood are presumably derived from resorbed cerebrospinal fluid protein whereas beta-TP-molecules bearing asialo-oligosaccharides are absent due to their hepatic clearance. The residual, sialylated beta-TP-species are probably eliminated from the blood via the kidney. This physiological clearance mechanism for the sialylated glycoforms is disturbed in hemodialysis patients resulting in about 100-fold elevated serum concentrations. These results let us suggest beta-trace may become a useful novel diagnostic protein in renal diseases.

Regional differences in in vitro growth of neural cell processes during development

Primary cell cultures from cerebral cortex, striatum and ventral mesencephalon obtained from rat fetal (embryonic day 17, E17) or postnatal (day 2, PN2) donors were grown either in media conditioned by subcultured astroglia from the same regions, an artificial trophic medium, normal human amniotic fluid, or in normal human cerebrospinal fluid. To estimate the presence of neuronal-like and non-neuronal cells, cell morphology and immunocytochemistry against microtubule-associated proteins and β-tubulin were taken into consideration. The percentage of emitting neural cells and length of cell processes were determined after 24 hr in culture. Growth of cell processes in neuronal and non-neuronal cells from prenatal striatum was minimal compared with that in cerebral cortex and ventral mesencephalon, regardless of the culture condition. Nerve growth factor, basic fibroblast growth factor or epidermal growth factor did not significantly modify cell growth in E17 cultures, except for epidermal growth factor, which reduced the number of emitting cells in striatal cultures and increased it in cerebral cortex ones. Cultures derived from postnatal striatum showed a significant increase in neurite length when grown in an astroglial conditioned medium as compared to cultures derived from prenatal (E17) striatum. Results suggest significant regional differences in the brain regarding growth of cell processes at age E17, and reversal of striatal ability to grow cell processes by postnatal day 2. Reduced growth of cell processes showed by E17 striatum cultures was rather independent of the culture media. This fact could suggest that such early regional differences would depend on characteristics of sublineages present at this developmental stage, which would modulate the organization of regional neuropils. The restricted growth of cell processes in cultures from E17 striatum, no longer present in postnatal striatum, suggests that inputs to the striatum may modify expression of cell lineages at later stages of development.

Biochemical Characterization of Alzheimer's Soluble Amyloid Beta Protein in Human Cerebrospinal Fluid: Association with High Density Lipoproteins

The soluble form of Alzheimer's amyloid β protein (sAβ) is associated with high density lipoproteins (HDL) in normal human plasma (BBRC,1994,205,1164–1171). Since sAβ is also present in cerebrospinal fluid (CSF) and the lipoprotein pattern of CSF is different from that of plasma, it was of interest to ascertain whether the interaction of sAβ with HDL also occurs in CSF. Normal human CSF lipoproteins were obtained by sequential flotation ultracentrifugation and analyzed for the presence of sAβ via immunoblot, size-exclusion chromatography, immunoelectron microscopy, N-terminal sequence and mass-spectrometry analyses. Soluble Aβ was associated with CSF-HDL particles of 16.8 ± 3.2 nm in diameter and approximately 200 kDa of relative molecular mass. A ∼4.3 kDa component purified by HPLC was immunoreactive with anti-Aβ antibodies and exhibited an N-terminal sequence identical to the Aβ peptide with a mass of 4325.1 Da, indicating that the main sAβ specie associated with CSF-HDL is Aβ1-40.