Establishing the Proteome of Normal Human Cerebrospinal Fluid

Background

Knowledge 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 Findings

We 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.

Conclusions

Our 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.     

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 · 10^–3 (CV ± 35%, range: 2.1–8.3 · 10^–3) corresponding to a mean albumin quotient in this group of subjects of Q_Alb = 5.8 · 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.     

Purine and pyrimidine base and nucleoside concentrations in human cerebrospinal fluid and plasma

Purine and pyrimidine base and nucleoside levels were determined in adult human lumbar (CSF) and plasma by reversed-phase high performance liquid chromatography (HPLC). Guanine, thymine, cytosine and uracil were not detectable (<0.1 M) in human CSF or plasma. Adenine was detectable in plasma (0.3 M) but was not found in CSF (<0.2 M). Hypoxanthine and xanthine levels in CSF were each approximately 2.5 M. Plasma levels of hypoxanthine and xanthine were considerably lower (0.4–0.6 M). Purine and pyrimidine ribouncleosides in human CSF were less than or equal to 0.2 M with the exception of uridine which was present at concentrations of 2–3 M. Although low concentrations of thymidine and deoxyuridine (0.2 M) were present in human plasma, purine and pyrimidine deoxyribonucleosides were less than 0.1 M in human lumbar CSF.     

In vivo influences on cerebrospinal fluid amino acid levels

The concentration of 19 amino acids and ethanolamine in two independent groups (n = 36, n = 19) of normal human cerebrospinal fluid (CSF) samples were measured by high performance liquid chromatography. Age, sex, time of lumbar puncture, position during lumbar puncture, protein and glucose content of the CSF were monitored and the influence of these parameters on CSF amino acid levels was determined. Hypotheses formulated after observing measurements from the first group of CSF samples were tested against the second group of CSF samples using conservative statistics. The main finding was a positive correlation between CSF glucose and CSF glutamate levels.     

Physiological concentrations of purines and pyrimidines

The concentrations of bases, nucleosides, and nucleosides mono-, di- and tri-phosphate are compared for about 600 published values. The data are predominantly from mammalian cells and fluids. For the most important ribonucleotides average concentrations ±SD (M) are: ATP, 3,152±1,698; GTP, 468±224; UTP, 567±460 and CTP, 278±242. For deoxynucleosidestriphosphate (dNTP), the concentrations in dividing cells are: dATP, 24±22; dGTP, 5.2±4.5; dCTP, 29±19 and dTTP 37±30. By comparison, dUTP is usually about 0.2 M. For, the 4 dNTPs, tumor cells have concentrations of 6–11 fold over normal cells, and for the 4 NTPs, tumor cells also have concentrations 1.2–5 fold over the normal cells. By comparison, the concentrations of NTPs are significantly lower in various types of blood cells. The average concentration of bases and nucleosides in plasma and other extracellular fluids is generally in the range of 0.4–6 M; these values are usually lower than corresponding intracellular concentrations. For phosphate compounds, average cellular concentrations are: P_i, 4400; ribose-1-P, 55; ribose-5-P, 70 and P-ribose-PP, 9.0. The metal ion magnesium, important for coordinating phosphates in nucleotides, has values (mM) of: free Mg²⁺, 1.1; complexed-Mg, 8.0. Consideration of experiments on the intracellular compartmentation of nucleotides shows support for this process between the cytoplasm and mitochondria, but not between the cytoplasm and the nucleus.     

Elevated Serum and Cerebrospinal Fluid Free Fatty Acid Levels Are Associated with Unfavorable Functional Outcome in Subjects with Acute Ischemic Stroke

The aim of this study was to evaluate the prognostic value of serum and cerebrospinal fluid (CSF) free fatty acid (FFA) levels in a cohort of patients with an acute ischemic stroke (AIS). In a prospective study, FFA levels were measured using an enzyme cycling method on admission in serum and CSF of 252 consecutive patients with AIS. The prognostic value of FFA to predict the functional outcome and mortality within 90-day was compared with the National Institutes of Health Stroke Scale score and with other known outcome predictors. Serum and CSF levels of FFA increased with increasing severity of stroke as defined by the NIHSS score (all P?<?0.001). Patients with an unfavorable outcomes and non-survivors had significantly increased FFA serum and CSF levels on admission (all P?<?0.0001). Multivariate logistic regression analysis adjusted for common risk factors showed that serum FFA ≥0.71 mmol/L (third quarters) was an independent predictor of functional outcome (odds ratios (OR)?=?4.86; 95 % confidence interval (CI) 2.26–10.48) and mortality (OR?=?7.72; 95 % CI 3.01–21.48). The area under the receiver operating characteristic curve of serum FFA was 0.79 (95 % CI, 0.72–0.86) for functional outcome and 0.86 (95 % CI, 0.78–0.94) for mortality. Similarly, CSF FFA level also was an indicator for predicting of functional outcome and mortality. FFA levels in serum and CSF may serve as independent biomarkers in addition of the traditional methods for assessing the functional outcome and mortality of AIS.     

Central and peripheral forms of C-type natriuretic peptide (CNP): evidence for differential regulation in plasma and cerebrospinal fluid

Aminoterminal proCNP (NTproCNP), a stable product of CNP gene expression and readily measured in human plasma, provides a new approach to studies of CNP which is rapidly degraded at source. CNP is detectable in human CSF but the presence and proportions of NTproCNP in CSF are unknown. Since CNP is widely expressed throughout the CNS, we hypothesized that the ratio of NTproCNP to CNP in CSF is greatly increased when compared to plasma and that CSF CNP peptides may contribute to their concentrations in the systemic circulation. Concurrent plasma and CSF concentrations of CNP forms were measured in 51 subjects undergoing spinal anesthesia for arranged orthopedic procedures. Elevated concentrations of NTproCNP (1045 ± 359 pmol/L), characterized by HPLC-RIA, were found in CSF and greatly exceeded those of CNP (7.9 ± 3.2 pmol/L). The ratio of NTproCNP to CNP in CSF (145 ± 55) was much higher than in plasma (31 ± 27). A significant inverse relation was found between plasma and CSF CNP concentrations (r = −0.29, p < 0.05). cGMP and neprilysin were unrelated to CNP levels in CSF. We conclude that CNP is differentially regulated across the brain in normal health. Despite markedly elevated levels of NTproCNP in CSF, it is unlikely that these contribute to systemic levels in healthy adults. Identifying NTproCNP as the dominant CNP form in CSF opens up the possibility of its use in future studies exploring CNP regulation within the CNS and possible applications in the diagnosis and monitoring of subjects with central neural disorders.     

A Technique for Serial Collection of Cerebrospinal Fluid from the Cisterna Magna in Mouse

Alzheimer''s disease (AD) is a progressive neurodegenerative disease that is pathologically characterized by extracellular deposition of β-amyloid peptide (Aβ) and intraneuronal accumulation of hyperphosphorylated tau protein. Because cerebrospinal fluid (CSF) is in direct contact with the extracellular space of the brain, it provides a reflection of the biochemical changes in the brain in response to pathological processes. CSF from AD patients shows a decrease in the 42 amino-acid form of Aβ (Aβ42), and increases in total tau and hyperphosphorylated tau, though the mechanisms responsible for these changes are still not fully understood. Transgenic (Tg) mouse models of AD provide an excellent opportunity to investigate how and why Aβ or tau levels in CSF change as the disease progresses. Here, we demonstrate a refined cisterna magna puncture technique for CSF sampling from the mouse. This extremely gentle sampling technique allows serial CSF samples to be obtained from the same mouse at 2-3 month intervals which greatly minimizes the confounding effect of between-mouse variability in Aβ or tau levels, making it possible to detect subtle alterations over time. In combination with Aβ and tau ELISA, this technique will be useful for studies designed to investigate the relationship between the levels of CSF Aβ42 and tau, and their metabolism in the brain in AD mouse models. Studies in Tg mice could provide important validation as to the potential of CSF Aβ or tau levels to be used as biological markers for monitoring disease progression, and to monitor the effect of therapeutic interventions. As the mice can be sacrificed and the brains can be examined for biochemical or histological changes, the mechanisms underlying the CSF changes can be better assessed. These data are likely to be informative for interpretation of human AD CSF changes.Open in a separate windowClick here to view.^{(49M, flv)}     

Effects of hyperthermia on enzymes and electrolytes in blood and cerebrospinal fluid in dogs

The effects of exposure to various degrees of heat stress on serum glutamate—oxaloacetate transaminase (SGOT), serum glutamate-pyruvate transaminase (SGPT), alkaline phosphatase (ALK-P-ase), calcium and chlorides have been studied on 75 dogs. Rectal temperature (T_re) was recorded before and after exposure to heat stress. These dogs were divided into 5 groups, according to the T_re level attained after exposure to heat stress. Rectal temperature was raised from normal to 39.45±0.47C in the first group, to 40.93±0.17C in the second group, to 41.87±0.22C in the third group, to 42.90 ± 0.21C in the fourth group and to 43.93±0.19C in the fifth group. The concentration of enzymes SGOT, SGPT and ALK-P-ase in blood and cerebrospinal fluid (CSF) increased significantly with hyperthermia. Calcium and chlorides concentrations in blood and in CSF tended to increase in hyperthermia. The integrity of the blood brain barrier for these enzymes and calcium is maintained under mild hyperthermia but it breaks down partially under influence of more severe hyperthermia. Core temperature above 41C results in damage to tissues and consequential rise of plasma enzymes. This degree of hyperthermia also seems to mark the beginning of injury to blood brain barrier. Critical core temperature tolerated by 50% of animals was 44C.     

Concentration Gradients of Free and Total γ-Aminobutyric Acid and Homocarnosine in Human CSF: Comparison of Suboccipital and Lumbar Sampling

Abstract: Concentrations of free and total γ-aminobutyric acid (GABA) and homocarnosine were determined in sequential aliquots of the first 30 ml of CSF obtained by lumbar puncture in five patients. Rostrocaudal gradients were calculated and compared to gradients estimated by determining concentrations of these substances in CSF obtained by simultaneous suboccipital and lumbar punctures in four more patients. In the lumbar fractions study, rostrocaudal mean gradients of 0.36, 36, and 21 pmol/ml for free GABA, total GABA, and homocarnosine, respectively, were calculated. In the suboccipital/lumbar study, gradients of 0.33, 30, and 24 pmol/ml for free GABA, total GABA, and homocarnosine, respectively, were estimated. These results indicate that valid comparison of CSF concentrations of these substances is restricted to similar fractions and suggest that in CSF the substances originate largely from brain rather than from peripheral sources.     

Serotonin in human lumbar cerebrospinal fluid: a reassessment

An inter-laboratory comparison study was carried out in order to ascertain mean levels of serotonin (5-HT) in human lumbar cerebrospinal fluid (CSF). Analyses were performed using high performance liquid chromatography (HPLC) coupled with either electrochemical (LC-EC) or fluorometric (LC-F) detection. With the detection limits obtained (7-8 pg/ml for LC-EC, 7-15 pg/ml for LC-F) 5-HT was not usually detected in human lumbar CSF. The findings indicate that the true mean concentration of CSF 5-HT is less than 10 pg/ml. This upper limit is substantially lower than all previous reports of 5-HT concentrations in normal human lumbar CSF. The extremely low concentrations of 5-HT present in CSF make it unlikely that CSF 5-HT will be of clinical utility in assessing central serotonergic function.     

Use of a Hemoglobin-Trapping Approach in the Determination of Nitric Oxide in In Vitro and In Vivo Systems

We describe methods for measuring the release of nitric oxide (NO) derived from organic nitrates in vitro, using triple wavelength and difference spectrophotometry in the presence and absence of concentric microdialysis probes. These methods are based on the ability of NO to oxidize oxyhemoglobin (OxyHb) to methemoglobin (MetHb) quantitatively in aqueous solution. Isosorbide dinitrate (ISDN), a thiol-dependent organic nitrate, increased MetHb concentration in 45 min from 2.47 ± 0.47 to 4.15 ± 0.12 M (p < 0.05) and decreased OxyHb concentration from 2.13 ± 0.35 to 0.33 ± 0.26 M (p < 0.05) at 37°C. At 27°C, the OxyHb concentration was not significantly altered (2.04 ± 0.23 to 1.60 ± 0.04 M) by ISDN, nor was the MetHb concentration (from 2.68 ± 0.50 to 2.59 ± 0.25 M). Sodium nitroprusside (SNP), a thiol-independent organic nitrate, increased MetHb concentrations in 30 min from 4.21 ± 0.26 to 6.00 ± 0.56 M (p < 0.05) at 37°C, and from 4.23 ± 0.39 to 5.90 ± 0.43 M (p < 0.01) at 27°C. SNP also decreased OxyHb concentrations in 30 min from 1.99 ± 0.32 to 0.13 ± 0.12 M (p < 0.01) at 37°C, and from 2.25 ± 0.31 to 0.13 ± 0.09 M (p < 0.01) at 27°C. Difference spectrophometry indicated that 0.25-5 mM SNP significantly increased NO production in a dose-dependent fashion. This hemoglobin-trapping technique was also useful in quantifying the concentrations of NO released from SNP in aqueous solution in vitro, using concentric microdialysis probes. The NO concentration following exposure to SNP was 530 ± 50 nM, as determined using the difference spectrophotometric technique. To demonstrate the applicability of this technique to in vivo microdialysis, we implanted concentric microdialysis probes into hippocampus and cerebellum of conscious and anesthetized rats. Baseline NO concentrations in hippocampus of conscious and anesthetized rats were 11 ± 2 nM and 23 ± 9 nM, respectively, while in the cerebellum NO concentrations were 28 ± 9 nM and 41 ± 20 nM, respectively. These results demonstrate that microdialysis using a novel hemoglobin-trapping technique possesses adequate sensitivity to measure the NO levels produced from organic nitrates in aqueous solutions, and further document the applicability of this approach to in vivo systems.     

Biometeorological norm as tolerance interval of man to weather stimuli

The paper discusses the problem of a biometeorological norm as the tolerance interval of the human organism to weather stimuli. The methodical approach is based on the normal distribution of variables. In this study thermal sensation (Effective Temperature) and daily numbers of traffic accidents were correlated. The results confirm the weak stimulus effect of meteorological conditions when they lie within the normal ±0.0 to ±1.5 standard deviation range.     

Fasting increases plasma membrane fatty acid-binding protein (FABPPM) in red skeletal muscle

The present study was designed to investigate the presence of the fatty acid-binding protein (FABPPM) in the plasma membranes of skeletal muscles with different oxidative capacities for free fatty acid (FFA) oxidation during conditions of normal (fed) or increased (fasted) FFA utilization in the rat. Female Sprague-Dawley rats were either fed or fasted for 12, 24, or 48 h and, plasma membranes (PM) fractions from red and white skeletal muscles were isolated. Short-term fasting significantly decreased body weight by 11% and blood glucose concentration by 42% (6.6 ± 0.2-3.8 ± 0.4 mmol/l) and increased plasma FFA concentration by 5-fold (133 ± 14-793 ± 81 µmol/l). Immunoblotting of PM fractions showed that FABPPM protein content was 83 ± 18% higher in red than in white skeletal muscle and correlated with oxidative capacity as measured by succinate dehydrogenase activity (r = 0.78, p < 0.05). Short-term fasting significantly increased FABPPM protein content by 60 ± 8% in red skeletal muscle but no change was measured in white skeletal muscle. These results show that FABPPM protein content in skeletal muscle is related to oxidative potential and can be increased during a physiological condition known to be associated with an increase in FFA utilization, suggesting that cellular expression of FABPPM may play a role in the regulation of FFA metabolism in skeletal muscle. (Mol Cell Biochem 166: 153-158, 1997)     

An ultramicro bioluminescence assay of enolase: Application to human cerebrospinal fluid

A highly sensitive method based on bioluminescence is described for the assay of enolase which can measure as little as 0.4×10^–6IU of activity. This corresponds to an amount of enzyme present in 1–2 l of normal human cerebrospinal fluid and is therefore easily applicable to clinical samples of CSF which can only be obtained in very small amounts. The reproducibility of the method is very high within a broad range of enzyme concentrations and the assay is linear from 0.4×10^–6 IU up to at least 50×10^–6IU of enzyme. This would permit application of the method to biological samples containing low as well as high enolase activities and especially for monitoring changes in enolase concentrations in the CSF and in the serum, as a function of pathological lesions in the central nervous system and other tissues.Part of this work was presented at the 2nd International Symposium on monoclonal antibodies and inborn errors of metabolism, Brugge, Belgium, October 1983.     

Effects of acute tryptophan depletion on plasma and cerebrospinal fluid tryptophan and 5-hydroxyindoleacetic acid in normal volunteers

Brain serotonin synthesis and metabolism (turnover), as indicated by CSF concentrations of 5-hydroxyindoleacetic acid (5-HIAA), may depend on plasma concentrations of the essential amino acid L-tryptophan (TRP). We investigated the biochemical effects of acute plasma TRP depletion (ATD) in normal volunteers undergoing a 36-h CSF collection via lumbar drain. Six subjects who were in good health were put on a low-TRP diet (160 mg/day) 24 h before lumbar puncture; this diet was continued for the first 22 h of the CSF collection. At hour 22, subjects ingested a TRP-deficient 15-amino acid drink shown previously to deplete plasma TRP. Total plasma TRP, free plasma TRP, and CSF TRP subsequently decreased 86.3, 86.5, and 92.3%, respectively. CSF 5-HIAA decreased by 32.8%. Plasma total and free TRP concentrations were both decreased at approximately 2 h following ingestion of the TRP-free amino acid drink and were lowest approximately 6 h after ATD; CSF TRP and 5-HIAA were decreased at 2.5 h and approximately 4 h after ATD, respectively. CSF TRP was lowest 8.0 h later. CSF 5-HIAA continued to decrease 14 h after the TRP-deficient amino acid drink was given.     

Assessment of Amyloid β Protein in Cerebrospinal Fluid as an Aid in the Diagnosis of Alzheimer's Disease

Abstract: The principal constituent of amyloid plaques found in the brains of individuals with Alzheimer's disease (AD) is a 39–42-amino-acid protein, amyloid β protein (Aβ). This study examined whether the measurement of Aβ levels in CSF has diagnostic value. There were 108 subjects enrolled in this prospective study: AD (n = 39), non-AD controls (dementing diseases/syndromes; n = 20), and other (n = 49). CSF was obtained by lumbar puncture, and Aβ concentrations were determined using a dual monoclonal antibody immunoradiometric sandwich assay. The mean Aβ value for the AD group (15.9 ± 6.8 ng/ml) was not significantly different from that for the non-AD control group (13.0 ± 7.1 ng/ml; p = 0.07), and substantial overlap in results were observed. Aβ values did not correlate with age ( r = −0.05, p = 0.59), severity of cognitive impairment ( r = 0.22, p = 0.21), or duration of AD symptoms ( r = 0.14, p = 0.45). These findings are in conflict with other reports in the literature; discrepant results could be due to the instability of Aβ in CSF. Aβ immunoreactivity decays rapidly under certain conditions, particularly multiple freeze/thaw cycles. Use of a stabilizing sample treatment buffer at the time of lumbar puncture allows storage of CSF without loss of Aβ reactivity. In conclusion, the total CSF Aβ level is not a useful marker for current diagnosis of AD.     

Characterization of the inhibitory effects of erythromycin and clarithromycin on the HERG potassium channel

Both erythromycin and clarithromycin have been reported to cause QT prolongation and the cardiac arrhythmia torsade de pointes in humans, however direct evidence documenting that these drugs produce this effect by blocking human cardiac ion channels is lacking. The goal of this study was to test the hypothesis that these macrolide antibiotics significantly block the delayed rectifier current (I_Kr) encoded by HERG (the human ether-a-go-go-related gene) at drug concentrations, temperature and ionic conditions mimicking those occurring in human subjects. Potassium currents in HEK 293 cells stably transfected with HERG were recorded using a whole cell voltage clamp method. Exposure of cells to erythromycin reduced the HERG encoded potassium current in a concentration dependent manner with an IC₅₀ of 38.9 ± 1.2 M and Hill Slope factor of 0.4 ± 0.1. Clarithromycin produced a similar concentration-dependent block with an IC₅₀ of 45.7 ± 1.1 M and Hill Slope factor of 1.0 ± 0.1. Erythromycin (25–250 M) and clarithromycin (5 or 25 M) also produced a significant decrease in the integral of the current evoked by an action potential shaped voltage clamp protocol. The results of this study document that both erythromycin and clarithromycin significantly inhibit the HERG potassium current at clinically relevant concentrations.     

Cerebrospinal fluid gradients of acetylcholinesterase and butyrylcholinesterase activity in healthy aging

Acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) activities were measured in 13 sequential 2 ml aliquots of cerebrospinal fluid (CSF) obtained by lumbar puncture from 7 young and 7 elderly healthy normal subjects. The slopes of the rostrocaudal gradients of AChE and BChE were calculated and compared to those of total protein concentration and the major dopaminergic metabolite homovanillic acid (HVA), for which a pronounced rostrocaudal gradient (with highest concentrations of HVA in more rostral CSF) is consistent with HVA originating primarily from the brain. AChE activity was higher in more caudal fractions of young, but not elderly subjects and there was a significant difference between the mean AChE gradient slopes in the young and old groups. These results suggest that the spinal cord makes an important contribution to AChE activity in lumbar CSF. Furthermore, the absence of a negative AChE gradient in elderly subjects may be the result of a greater rate of entry of cerebral AChE into CSF, possibly as a consequence of an increased ventricular surface area and shorter diffusion distances in atrophic elderly brains. In contrast to AChE, BChE activity and total protein concentrations were higher in more caudal CSF fractions of not only young but also old subjects. In addition, there was a significant correlation between the gradient slopes of BChE activity and total protein concentrations, suggesting that the majority of BChE activity in lumbar CSF derives from the same source as the majority of total protein, namely plasma. The diffuse (i.e. brain and spinal cord) origin of AChE in lumbar CSF would explain the relatively modest changes in lumbar CSF AChE activity in diseases involving certain central cholinergic systems, most notably Alzheimer's disease.     

Mechanisms of cellular uptake of long chain free fatty acids

Cells take up long chain free fatty acids (FFA) in vivo from the non-protein bound ligand pools in extracellular fluid and plasma, which contain ~100 and 600 M albumin, respectively. The physiologic range of unbound FFA concentrations in such fluids has traditionally been calculated at < 1 M. Studies of [3H]-oleate uptake by hepatocytes, adipocytes, cardiac myocytes and other cell types demonstrate that FFA uptake within this range is saturable, and exhibits many other kinetic properties indicative of facilitated transport. Within this range, the uptake kinetics of the acidic (pKa = 0.5) FFA analog 2,2,3- heptafluorostearate are similar to those of stearate. Thus, uptake of physiologic concentrations of FFA involves facilitated transport of the FFA anion (FA-). Over a much wider range of unbound FFA concentrations hepatocellular [3H]-oleate uptake exhibits both saturable and non-saturable components. Oleate binding to liver plasma membranes (LPM) also demonstrates such components. Comparing the two components of FFA uptake to the corresponding components of binding permits estimates of trans-membrane transport rates. T1/2 for saturable uptake (~ 1 sec) is less than for non-saturable uptake (~ 14 sec). Others have determined the flip-flop rates of protonated FFA (FAH) across small and large unilamellar vesicles (SUV, LUV) and across cellular plasma membranes. These reported flip-flop rates, measured by the decrease in pH resulting from the accompanying proton flux, exhibit a highly significant inverse correlation with cell and vesicle diameter (r = 0.99). Although T1/2's in vesicles are in the msec range, those in cells are > 10 sec, and thus comparable to the rates of non-saturable uptake we determined. Thus, under physiologic conditions, the predominant mechanism of cellular FFA uptake is facilitated transport of FA-; at much higher, non- physiologic FFA concentrations, passive flip-flop of FAH predominates. Several plasma membrane proteins have been identified as potential mediators of facilitated FFA transport. Studies in animal models of obesity and non-insulin dependent diabetes mellitus demonstrate that tissue-specific regulation of facilitated FFA transport has important pathophysiologic consequences.