Erythrocyte membrane phospholipids in psychotic patients

Thin layer chromatography and laser densitometry were used to examine erythrocyte membrane phospholipid composition in 13 medication-free patients with schizophrenia or schizoaffective disorder and 11 healthy control subjects. The patient group had significantly decreased levels of phosphatidylethanolamine and a significant increase in sphingomyelin. The patient group also showed a trend toward decreased levels of phosphatidylinositol bisphosphate. Levels of phosphatidylcholine and lysophosphatidylcholine did not significantly differ between groups. There were no significant effects of age, body mass index, or gender on any of the phospholipid parameters. The observed alterations may point to abnormalities in key membrane-related functions, including signal transduction and ion transport.

Frontal lobe metabolism and cerebral morphology in schizophrenia: 31P MRS and MRI studies

The relation between frontal lobe membrane phospholipid metabolism as measured by 31Phosphorus magnetic resonance spectroscopy (31P MRS) and cerebral morphology as measured on magnetic resonance images (MRI) was examined in nine first episode neuroleptic naive schizophrenic patients. Total corpus callosal area was significantly correlated with phosphodiester concentration. When examined separately, this relation was confined to the rostral quartile (genu) of the corpus callosum. The pathophysiological significance of this finding is discussed in relation to neurodevelopmental hypotheses of schizophrenia.

A preliminary 31P MRS study of autism: evidence for undersynthesis and increased degradation of brain membranes

In this pilot study, brain high energy phosphate and membrane phospholipid metabolism were investigated in the dorsal prefrontal cortex of 11 high-functioning autistic adolescent and young adult men (the age range is 12-36 years) and 11 age-, gender-, IQ, race- and socioeconomic status-matched normal controls using in vivo 31P nuclear magnetic resonance spectroscopy (MRS). The autistic group had decreased levels of phosphocreatine and esterified ends (alpha ATP + alpha ADP + dinucleotides + diphosphosugars) compared to the controls. When the metabolite levels were compared within each subject group with neuropsychologic and language test scores, a common pattern of correlations was observed across measures in the autistic group, but not in the control group. As test performance declined in the autistic subjects, levels of the most labile high energy phosphate compound and of membrane building blocks decreased, and levels of membrane breakdown products increased. No significant correlations were present with age in either group or with IQ in the control group, suggesting that these findings were not the consequence of age or IQ effects. This pilot study provides tentative evidence of alterations in brain energy and phospholipid metabolism in autism that correlate with the neuropsychologic and language deficits. The findings are consistent with a hypermetabolic energy state and undersynthesis of brain membranes and may relate to the neurophysiologic and neuropathologic abnormalities in autism.

Alterations of selected enzymes of phospholipid metabolism in Alzheimer's disease brain tissue as compared to non-Alzheimer's demented controls

Previous studies have demonstrated elevated brain levels of phosphomonoesters in early stages of Alzheimer's disease and elevations of phosphodiesters later in the disease. In addition, preliminary quantitative analyses of the phospholipids of Alzheimer's brain reveals either decreases in some phospholipids or elevations followed by decreases in others. This study quantitated the activities of selected enzymes involved in phospholipid and choline metabolism and demonstrated elevated glycerol-3-phosphorylcholine phosphodiesterase and decreased choline kinase activities in Alzheimer's disease brain. The former could provide an enzymatic mechanism for the increased phosphorylcholine found in Alzheimer's disease brain.

31P nuclear magnetic resonance spectroscopy: neurodevelopment and schizophrenia

A number of studies have demonstrated alterations in the structure and function of the frontolimbic system in some schizophrenic patients. Recent in vivo phosphorus-31 nuclear magnetic resonance studies of the dorsal prefrontal cortex in neuroleptic-naive, first-episode schizophrenic patients and matched controls have shown evidence of alterations in membrane phospholipid and energy metabolism. The membrane alterations observed in the schizophrenic patients are compatible with either premature aging or altered timing and exaggeration of regressive events occurring during normal brain development. These molecular changes may precede onset of clinical symptoms and brain structural changes in schizophrenia and suggest fresh approaches to the pathogenesis and treatment of this illness.

N-acetyl-L-aspartate and other amino acid metabolites in Alzheimer's disease brain: A preliminary proton nuclear magnetic resonance study

We used proton nuclear magnetic resonance spectroscopy in this preliminary study of perchloric acid extracts of 12 Alzheimer's disease (AD) and five control brain samples to measure the relative levels of taurine, aspartate, glutamine, glutamate, gamma-aminobutyric acid (GABA), and the putative neuronal marker, N-acetyl-L-aspartate (NAA). We found no significant changes in taurine, aspartate, or glutamine. NAA was lower in AD compared with control, and this decrease correlated with the number of senile plaques and neurofibrillary tangles in adjacent tissue sections. GABA levels also were lower in AD brain. Glutamate levels were greater in AD than control and showed a close, inverse correlation with NAA levels. These findings suggest that the decrease in NAA reflects neuronal loss and that remaining neurons could be exposed to a relative excess of glutamate and a relative lack of GABA. If present in the neurotransmitter pool, this imbalance could result in neurotoxic cell damage. This hypothesis is further supported by in vitro and in vivo phosphorus 31 nuclear magnetic resonance findings.

Molecular insights into schizophrenia

A number of studies have demonstrated alterations in the structure and function of the frontal cortex in some schizophrenic patients. The possible etiology and pathogenesis of these abnormalities are unknown, but genetic and developmental causes are frequently mentioned. Recent in vivo 31P NMR studies of the dorsal prefrontal cortex have been conducted in eleven neuroleptic naive, first episode schizophrenic patients and compared with normal controls of comparable age, educational level and parental educational level. The findings in the schizophrenic patients are different from those of normal IQ adult autistic patients of comparable age and Alzheimer's patients but similar to normal elderly controls. These studies show decreased frontal lobe utilization of adenosine triphosphate in the schizophrenic patients which suggests a hypoactive dorsal prefrontal cortex. In addition, indices of membrane phospholipid metabolism are altered in the schizophrenic patients. However, the findings in the schizophrenic patients are quite similar to those observed in normal elderly controls and to those that normally occur to a lesser degree during adolescence. The phospholipid alterations observed in the schizophrenic patients are compatible with either premature aging or altered timing and exaggeration of the regressive events which occur during normal brain development. The changes in high-energy phosphate metabolism observed in the schizophrenic patients may prove to be state dependent, but the changes in membrane phospholipid metabolism could be related to molecular changes that precede the onset of clinical symptoms and brain structural changes in schizophrenia. These findings suggest new avenues of thinking about the pathogenesis and treatment of schizophrenia.

Actions of phosphomonoesters on CA1 hippocampal neurons as revealed by a combined electrophysiological and nuclear magnetic resonance study

Phosphomonoesters (PMEs), precursors of membrane phospholipids, are found in high levels in the developing brain and Alzheimer's disease brain. The present study details the neurophysiological and metabolic effects of acute PME elevation on the Fisher 344 rat in vitro hippocampal slice. Two abundant PMEs, phosphoethanolamine (PE) and L-phosphoserine (PS), reliably altered properties of synaptic transmission at the Schaffer collateral/commissural-CA1 cell synapse. Specifically, PE reversibly depressed the amplitude of population EPSPs at millimolar concentrations but had no effect at micromolar concentrations. PS had biphasic effects on population EPSPs, inducing first a reduction followed by an enhancement of response amplitude. In contrast to PE, the effects of PS were not reversible; population EPSPs were augmented during the wash of PS, and the CA3 region generated evoked (but not spontaneous) epileptiform discharges. 31P nuclear magnetic resonance spectroscopy revealed enhanced slice uptake of PS compared to PE. There was no significant effect of PE on slice high-energy phosphates but incubation with PS significantly lowered slice phosphocreatine (PCr) and ATP concentrations. These observations indicate that the slice uptake of PS could be energy requiring and the enhanced response amplitude observed at 5 mM PS also could produce a drain on high-energy phosphates. Possible modes of PME action on hippocampal physiology are discussed.

Possible roles of L-phosphoserine in the pathogenesis of Alzheimer's disease

L-Phosphoserine is a membrane metabolite that is elevated in Alzheimer's disease brain. This compound has close structural similarity to L-glutamate. Electrophysiological studies indicate that L-phosphoserine has an acute inhibitory effect, but a delayed excitatory action. A hypothesis is developed based on pharmacological and electrophysiological studies that suggest that the inhibition may be mediated through presynaptic inhibition of L-glutamate release or perhaps antagonism of postsynaptic kainic acid receptors. The mechanism of the delayed excitation may lie in the tendency of L-phosphoserine to mimic the action of L-2-amino-4-phosphonobutyric acid, a blocker of chloride- and calcium-sensitive L-glutamate transport. L-Phosphoserine has also been found to be a competitive antagonist at the N-methyl-D-aspartate recognition site and an antagonist of metabotropic receptor-mediated hydrolysis of inositol phospholipids. Because of these actions, there are several potentially important implications for the elevation of L-phosphoserine in Alzheimer's disease, including production memory impairment through presynaptic inhibition of L-glutamate release or blockade of postsynaptic N-methyl-D-aspartate receptors and/or blockade of certain L-glutamate transport sites resulting in increased L-glutamate levels in the synaptic cleft.

Magnetic resonance spectroscopy in psychiatry: potential, pitfalls, and promise

Magnetic resonance spectroscopy (MRS) is a novel noninvasive approach to measuring important metabolites in living tissue. Its application to psychiatry is just beginning. In vivo MRS with 31P provides important information on brain phospholipid metabolism and energy production. In vivo 13C and 1H MRS can reveal information about carbohydrate, protein, and amino acid metabolism. In vivo 7Li and 19F MRS can be used to study the pharmacology of lithium and fluorinated psychopharmacological agents. MRS with 23Na can yield information about electrolyte balance. The limitations of in vivo MRS include poor sensitivity, poor resolution, and the fact that only highly mobile atomic nuclei can be detected. Future clinical application of MRS will benefit from improvements in the technology of localization, use of spectroscopy contrast agents, stronger magnets, and the merging of MRS and imaging technology.

L-phosphoserine, a metabolite elevated in Alzheimer's disease, interacts with specific L-glutamate receptor subtypes

L-Phosphoserine is one of the phosphomonoesters elevated in Alzheimer's disease brain and has close structural similarity to L-glutamate. This study attempts to define precisely the actions of L-phosphoserine at L-glutamate receptor subtypes. L-Phosphoserine is shown to bind to N-methyl-D-aspartate and kainic acid receptor subtypes, but not to the quisqualic acid subtype. Studies of [3H]MK-801 binding in the presence and absence of L-glutamate and glycine show L-phosphoserine to be a competitive N-methyl-D-aspartate antagonist. The IC50 of L-phosphoserine in these studies varies from 373 to 721 microM. This may indicate a physiologically relevant action of L-phosphoserine in Alzheimer's disease brain because whole brain concentrations may reach over 1 mM.

Al-ATP as an intracellular carrier of Al(III) ion

1. Using 27Al and 31P NMR spectroscopy in conjunction with an Al lactate aqueous reagent at pH 7.2, Al complexes of ATP and of phospholipids were characterized in synthetic-aqueous and organic-phospholipid chemical systems and in the intact human red blood cell. 2. The observed 31P NMR chemical shifts of the Al-ATP complex in aqueous laboratory preparations or the intact human red blood cell were, respectively, alpha phosphate, -11.53 delta; beta phosphate, -22.65 delta; and gamma phosphate, -10.95 delta. 3. The observed complexed 27Al chemical shift was -2.22 delta. 4. The relative affinities for Al of the phospholipids determined from 31P NMR spectroscopic titrations were PA much greater than Cl much greater than PS greater than PG approximately equal to PI greater than PE plus approximately equal to PE much greater than SPH greater than PC.

Alzheimer's beta-amyloid protein is covalently modified when dissolved in formic acid

beta-Amyloid protein is a major protein component of neuritic plaques in the brain of Alzheimer's disease patients. A major advance in understanding the molecular biology of Alzheimer's disease came with the purification and sequencing of this protein. Because beta-amyloid protein is very insoluble, extreme conditions such as 88% formic acid were commonly used to dissolve its fibrils. We now report that 88% formic acid covalently modifies beta-amyloid protein fragments, probably by the formation of a formate ester to a serine in the protein. The t1/2 of the formylation is approximately 3.5 h, and the t1/2 for hydrolysis of the formylated peptide is much longer, being 9.9 h in water and 66 h in HPLC eluant. This suggests that if formic acid is used in the purification of beta-amyloid protein or peptide fragments of this protein, it is likely that some formylated peptide will be present in subsequent studies. Although unrecognized modification of a protein is inherently undesirable, it is uncertain what effects this formylation will have on ensuing studies. Certainly, investigations into the immunologic, physical, and physiologic properties of beta-amyloid protein could be influenced.

Changes in brain energy and phospholipid metabolism during development and aging in the Fischer 344 rat

The effects of brain development and aging on high-energy phosphate and membrane phospholipid metabolism were studied from birth to senescence in the Fischer 344 rat using 31P nuclear magnetic resonance spectroscopy. Marked developmental and smaller aging-related changes were observed in brain high-energy phosphates, phospholipid precursors and phospholipid breakdown products. The biochemical changes correlate with known histological and electrophysiological changes occurring in the brain during development (neuritic sprouting and onset of brain electrical activity) and aging (loss of dendritic processes). These findings provide a framework for interpreting the effects of physiological insults during different developmental and aging periods.

In vivo 31P NMR profiles of Alzheimer's disease and multiple subcortical infarct dementia

We used in vivo phosphorus 31 nuclear magnetic resonance (31P NMR) spectroscopy to study regional high-energy phosphate and phospholipid metabolism in brains of patients with dementia associated with probable Alzheimer's disease (AD) and multiple subcortical cerebral infarctions (MSID). The MSID patients demonstrated elevations of the phosphocreatine (PCr)/inorganic orthophosphate (Pi) ratio in both the temporoparietal and frontal regions. Phosphomonoesters (PME) and the ratio of PME to phosphodiesters were elevated in the temporoparietal region of AD. Pi was also elevated in the frontal and temporoparietal regions of AD. Findings from 31P NMR were accurate in distinguishing MSID from AD. Values of PCr/Pi accurately classified 100% of the MSID patients and 92% of AD. Pi and PME, considered jointly, also accurately classified all MSID and all but 1 AD. Findings from in vivo 31P NMR spectroscopy appear to yield metabolic profiles useful in distinguishing AD from MSID.

Quantitative evaluation of congo red binding to amyloid-like proteins with a beta-pleated sheet conformation

The binding of Congo red to several purified amyloid-like peptides having a beta-pleated sheet conformation was quantitatively examined. Congo red binds preferentially to the beta-pleated sheet conformation of both insulin fibrils and poly-L-lysine. Congo red does not bind nearly so well to poly-L-serine or polyglycine, despite the fact that these peptides also have a beta-pleated sheet conformation. Binding to insulin fibrils was saturable with an apparent Bmax of 2 moles of Congo red per mole of insulin fibrils and an apparent KD of 1.75 x 10(-7) M. Binding to beta-poly-L-lysine was similar but had a much higher apparent Bmax of 43. Binding of Congo red to beta-poly-L-lysine was pH dependent and appeared to be determined by the number of protonated lysine residues in the 250 amino acid peptide. We present a new hypothesis in which Congo red binds to amyloid-like proteins via bonds between the two negatively charged sulfonic acid groups of Congo red and two positively charged amino acid residues of two separate protein molecules which are properly oriented by virtue of the beta-pleated sheet conformation of the peptide backbone.

Two simple methods for quantifying low-affinity dye-substrate binding

Binding with low-affinity ligands, such as histological dyes, can be difficult to quantitate owing to the dissociation of bound ligand with washing or the retention of nonspecifically bound ligand because of incomplete washing. The present report describes two simple, rapid methods of discriminating bound from free ligand without the need for washing steps. One method is based on the spectral changes induced in a dye ligand, Congo red, on binding to the "receptor" insulin fibrils. This method discriminates spectrophotometrically between bound and free ligand without requiring any physical separation of the two forms. No radioactive ligands are necessary, and, by using disposable cuvettes, the entire binding assay can be done in a single container without the need for transfers. The second method employs a non-traditional filtration approach that avoids the need for a washing step by measuring the decrease in concentration of the dye ligand in the filtrate rather than by applying the usual approach of measuring the absolute amount of ligand bound to the precipitated "receptor." Both methods show saturation of binding sites and give similar values for the KD and Bmax.

Correlation of phosphorus-31 magnetic resonance spectroscopy and morphologic findings in Alzheimer's disease

Senile plaques (SPs), especially, and neurofibrillary tangles are important pathologic markers for the diagnosis of Alzheimer's disease (AD), but neither is pathognomonic for AD. We hypothesize that elevations in levels of phosphomonoesters, precursors of membrane phospholipids, occur early in the pathogenesis of AD and precede the appearance of SPs. In contrast, elevations in levels of phosphodiesters, breakdown products of phospholipids, reflect degeneration of neural membranes and will correlate with the appearance of SPs. Correlative phosphorus-31 magnetic resonance spectroscopy and morphologic studies conducted to test this hypothesis disclosed that elevations in levels of phosphomonoesters had a negative correlation with the numbers of SPs, and elevations in levels of phosphodiesters had a positive correlation with the numbers of SPs. No correlations were observed for either membrane parameter and neurofibrillary tangles. These findings support our hypothesis and suggest that aberrations in the synthesis of membrane phospholipids are early metabolic events in the pathogenesis of AD.

31P nuclear magnetic resonance study of the brain in Alzheimer's disease

The histopathological hallmarks of Alzheimer's disease have long been considered to be neurofibrillary tangles (NFT) and neuritic (senile) plaques (SP). Neither of these structures, however, are unique to Alzheimer's disease, and both probably represent end-stage markers of the disorder. NFT have been demonstrated in many disorders; SP occur in small numbers with normal aging. Evidence is presented for elevation of phosphomonoesters (PME) in Alzheimer's brain compared to non-Alzheimer's diseased controls and normal controls. The PME detected by 31P nuclear magnetic resonance (NMR) spectroscopy of autopsy brain are predominantly anabolic precursors of membrane phospholipids. Elevated PME could be secondary to a metabolic block at the rate-limiting enzyme in membrane phospholipid synthesis, which is cytidine triphosphate (CTP): phosphocholine (or phosphoethanolamine) cytidyltransferase (EC 2.7.7.15). Elevated PME could also be secondary to decreased breakdown of PME by phospholipase D activity. Since CTP: phosphocholine cytidyltransferase is inactivated by phosphorylation and since there is independent evidence for hyperphosphorylation of tau and MAP-2 proteins in AD brain, enhanced protein kinase activity could be a common factor. Preliminary evidence suggests that PME could interact with N-methyl-D-aspartate receptors and potentially act as false neurotransmitters. Further studies will be needed to investigate these possibilities.

Considerations for brain pH assessment by 31P NMR

Recently in vivo NMR spectroscopy has been used to measure brain pH non-invasively. Both the inorganic orthophosphate (Pi) chemical shift (delta) and the difference between the chemical shifts of phosphocreatine (PCr) and Pi(delta delta PCr-Pi) have been proposed as indicators of brain pH. However, the precise delta of Pi may be difficult to determine under normoxic conditions as is the delta of PCr under hypoxic/ischemic conditions. Ideally one needs a NMR delta parameter that: (1) linearly changes between pH 6.0-8.0, (2) is either relatively unaffected or predictably affected by cations (e.g., Mg2+) other than H+, and that (3) comes from readily observable 31P NMR resonances whose delta's can be accurately assessed under all physiological conditions. Therefore, we undertook a systematic 31P NMR study of the pH and Mg2+ titration curves for 16 phosphorus-containing metabolites observed in brain by 31P NMR. On the basis of the titration curves, the delta delta's for PCr-Pi, phosphoethanolamine (PE)-Pi, and PCr-PE fulfill criteria (1) and (2), but not criterion (3). However, the delta delta of ATP gamma-alpha fulfills all three criteria and potentially provides information on the intracellular Mg2+ concentration.