Search

PIGF is a protein involved in the ethanolamine phosphate (EtNP) transfer steps of glycosylphosphatidylinositol (GPI) biosynthesis. PIGF forms a heterodimer with either PIGG or PIGO, two enzymes that transfer an EtNP to the second or third mannoses of GPI respectively. Heterodimer formation is essential for stable and regulated expression of PIGO and PIGG, but the functional significance of PIGF remains obscure. In the present study, we show that PIGF binds to PIGO and PIGG through distinct molecular domains. Strikingly, C-terminal half of PIGF was sufficient for its binding to PIGO and PIGG and yet this truncation mutant could not complement the PIGF defective mutant cells, suggesting that heterodimer formation is not sufficient for PIGF function. Furthermore, we identified a highly conserved motif in PIGF and demonstrated that the motif is not involved in binding to PIGO or PIGG, but critical for its function. Finally, we identified a PIGF homologue from Trypanosoma brucei and showed that it binds specifically to the T. brucei PIGO homologue. These data together support the notion that PIGF plays a critical and evolutionary conserved role in the ethanolamine-phosphate transfer-step, which cannot be explained by its previously ascribed binding/stabilizing function. Potential roles of PIGF in GPI biosynthesis are discussed.

The proliferation of the malaria-causing parasite Plasmodium falciparum within the erythrocyte is concomitant with massive phosphatidylcholine and phosphatidylethanolamine biosynthesis. Based on pharmacological and genetic data, de novo biosynthesis pathways of both phospholipids appear to be essential for parasite survival. The present study characterizes PfCK (P. falciparum choline kinase) and PfEK (P. falciparum ethanolamine kinase), which catalyse the first enzymatic steps of these essential metabolic pathways. Recombinant PfCK and PfEK were expressed as His6-tagged fusion proteins from overexpressing Escherichia coli strains, then purified to homogeneity and characterized. Using murine polyclonal antibodies against recombinant kinases, PfCK and PfEK were shown to be localized within the parasite cytoplasm. Protein expression levels increased during erythrocytic development. PfCK and PfEK appeared to be specific to their respective substrates and followed Michaelis-Menten kinetics. The Km value of PfCK for choline was 135.3+/-15.5 microM. PfCK was also able to phosphorylate ethanolamine with a very low affinity. PfEK was found to be an ethanolamine-specific kinase (Km=475.7+/-80.2 microM for ethanolamine). The quaternary ammonium compound hemicholinium-3 and an ethanolamine analogue, 2-amino-1-butanol, selectively inhibited PfCK or PfEK. In contrast, the bis-thiazolium compound T3, which was designed as a choline analogue and is currently in clinical trials for antimalarial treatment, affected PfCK and PfEK activities similarly. Inhibition exerted by T3 was competitive for both PfCK and PfEK and correlated with the impairment of cellular phosphatidylcholine biosynthesis. Comparative analyses of sequences and structures for both kinase types gave insights into their specific inhibition profiles and into the dual capacity of T3 to inhibit both PfCK and PfEK.

ADPRibase-Mn (Mn2+-dependent ADP-ribose/CDP-alcohol pyrophosphatase) was earlier isolated from rat liver supernatants after separation from ADPRibase-I and ADPRibase-II (Mg2+-activated ADP-ribose pyrophosphatases devoid of CDP-alcohol pyrophosphatase activity). The last mentioned are putative Nudix hydrolases, whereas the molecular identity of ADPRibase-Mn is unknown. MALDI (matrix-assisted laser-desorption ionization) MS data from rat ADPRibase-Mn pointed to a hypothetical protein that was cloned and expressed and showed the expected specificity. It is encoded by the RGD1309906 rat gene, which so far has been annotated simply as 'hydrolase'. ADPRibase-Mn is not a Nudix hydrolase, but it shows the sequence and structural features typical of the metallophosphoesterase superfamily. It may constitute a protein family of its own, the members of which appear to be specific to vertebrates, plants and algae. ADP-ribose was successfully docked to a model of rat ADPRibase-Mn, revealing its putative active centre. Microarray data from the GEO (Gene Expression Omnibus) database indicated that the mouse gene 2310004I24Rik, an orthologue of RGD1309906, is preferentially expressed in immune cells. This was confirmed by Northern-blot and activity assay of ADPRibase-Mn in rat tissues. A possible role of ADPRibase-Mn in immune cell signalling is suggested by the second-messenger role of ADP-ribose, which activates TRPM2 (transient receptor potential melastatin channel-2) ion channels as a mediator of oxidative/nitrosative stress, and by the signalling function assigned to many of the microarray profile neighbours of 2310004I24Rik. Furthermore, the influence of ADPRibase-Mn on the CDP-choline or CDP-ethanolamine pathways of phospholipid biosynthesis cannot be discounted.

Human PHOSPHO1 is a phosphatase enzyme for which expression is upregulated in mineralizing cells. This enzyme has been implicated in the generation of P(i) for matrix mineralization, a process central to skeletal development. PHOSPHO1 is a member of the haloacid dehalogenase (HAD) superfamily of Mg2+-dependent hydrolases. However, substrates for PHOSPHO1 are, as yet, unidentified and little is known about its activity. We show here that PHOSPHO1 exhibits high specific activities toward phosphoethanolamine (PEA) and phosphocholine (PCho). Optimal enzymic activity was observed at approx. pH 6.7. The enzyme shows a high specific Mg2+-dependence, with apparent K(m) values of 3.0 microM for PEA and 11.4 microM for PCho. These results provide a novel mechanism for the generation of P(i) in mineralizing cells from PEA and PCho.

Extracts of frozen rat liver were found to catalyse the formation of 3H2O from DL-2-hydroxy[2-3H]glutarate. Three peaks of enzyme activities were observed on separation by chromatography on DEAE-Sepharose. The first and second peaks corresponded to an enzyme acting on L-2-hydroxyglutarate and the third peak corresponded to an enzyme acting on D-2-hydroxyglutarate, as indicated by competitive inhibition of the detritiation of the racemic radioactive compound by the unlabelled L- and D-isomers respectively. The enzyme acting on the D-form was further characterized. It was independent of NAD or NADP and it converted D-2-hydroxyglutarate into a-ketoglutarate, transferring electrons to artificial electron acceptors. It also oxidized D-lactate, D-malate and meso-tartrate and was stimulated by Zn2+, Co2+ and Mn2+, but not by Mg2+ or Ca2+. Subcellular fractionation indicated that it was present in the mitochondrial fraction. The enzyme was further purified by chromatography on Blue Trisacryl and phenyl-Sepharose, up to a stage where only a few bands were still visible by SDS/PAGE. Among the four candidate polypeptides that were identified by MS, one corresponded to a predicted mitochondrial protein homologous with FAD-dependent D-lactate dehydrogenase. The corresponding human protein was expressed in HEK-293 cells and it was shown to catalyse the detritiation of DL-2-hydroxy[2-3H]glutarate with similar properties as the purified rat enzyme.

Anandamide (an endocannabinoid) and other bioactive long-chain NAEs (N-acylethanolamines) are formed by direct release from N-acyl-PE (N-acyl-phosphatidylethanolamine) by a PLD (phospholipase D). However, the possible presence of a two-step pathway from N-acyl-PE has also been suggested previously, which comprises (1) the hydrolysis of N-acyl-PE to N-acyl-lysoPE by PLA1/PLA2 enzyme(s) and (2) the release of NAEs from N-acyllysoPE by lysoPLD (lysophospholipase D) enzyme(s). In the present study we report for the first time the characterization of enzymes responsible for this pathway. The PLA1/PLA2 activity for N-palmitoyl-PE was found in various rat tissues, with the highest activity in the stomach. This stomach enzyme was identified as group IB sPLA2 (secretory PLA2), and its product was determined as N-acyl-1-acyl-lysoPE. Recombinant group IB, IIA and V of sPLA2s were also active with N-palmitoyl-PE, whereas group X sPLA2 and cytosolic PLA2a were inactive. In addition, we found wide distribution of lysoPLD activity generating N-palmitoylethanolamine from N-palmitoyl-lysoPE in rat tissues, with higher activities in the brain and testis. Based on several lines of enzymological evidence, the lysoPLD enzyme could be distinct from the known N-acyl-PE-hydrolysing PLD. sPLA2-IB dose dependently enhanced the production of N-palmitoylethanolamine from N-palmitoyl-PE in the brain homogenate showing the lysoPLD activity. N-Arachidonoyl-PE and N-arachidonoyl-lysoPE as anandamide precursors were also good substrates of sPLA2-IB and the lysoPLD respectively. These results suggest that the sequential actions of PLA2 and lysoPLD may constitute another biosynthetic pathway for NAEs, including anandamide.

For an insight regarding the control of PtdEtn (phosphatidylethanolamine) synthesis via the CDPethanolamine pathway, rat liver cDNA encoding ECT (CTP:phosphoethanolamine cytidylyltransferase) was transiently or stably transfected in Chinese-hamster ovary cells and a rat liver-derived cell line (McA-RH7777), resulting in a maximum of 26- and 4-fold increase in specific activity of ECT respectively. However, no effect of ECT overexpression on the rate of [3H]ethanolamine incorporation into PtdEtn was detected in both cell lines. This was explored further in cells overexpressing four times ECT activity (McA-ECT1). The rate of PtdEtn breakdown and PtdEtn mass were not changed in McA-ECT1 cells in comparison with control-transfected cells. Instead, an accumulation of CDPethanolamine (label and mass) was observed, suggesting that in McA-ECT1 cells the ethanolaminephosphotransferase-catalysed reaction became rate-limiting. However, overexpression of the human choline/ethanolaminephosphotransferase in McA-ECT1 and control-transfected cells had no effect on PtdEtn synthesis. To investigate whether the availability of DAG (diacylglycerol) limited PtdEtn synthesis in these cells, intracellular DAG levels were increased using PMA or phospholipase C. Exposure of cells to PMA or phospholipase C stimulated PtdEtn synthesis and this effect was much more pronounced in McA-ECT1 than in control-transfected cells. In line with this, the DAG produced after PMA exposure was consumed more rapidly in McA-ECT1 cells and the CDPethanolamine level decreased accordingly. In conclusion, our results suggest that the supply of CDPethanolamine, via the expression level of ECT, is an important factor governing the rate of PtdEtn biosynthesis in mammalian cells, under the condition that the amount of DAG is not limiting.

Anandamide and other polyunsaturated N-acylethanolamines (NAEs) exert biological activity by binding to cannabinoid receptors. These receptors are linked to G(i/o) proteins and their activation leads to extracellular-signal-regulated protein kinase (ERK) and c-Jun N-terminal kinase (JNK) mitogen-activated protein kinase (MAP kinase) activation, inhibition of cAMP-dependent signalling and complex changes in the expression of various genes. Saturated and monounsaturated NAEs cannot bind to cannabinoid receptors and may thus mediate cell signalling through other targets. Here we report that both saturated/monounsaturated NAEs and anandamide (20:4(n-6) NAE) stimulate cannabinoid-receptor-independent ERK phosphorylation and activator protein-1 (AP-1)-dependent transcriptional activity in mouse epidermal JB6 cells. Using a clone of JB6 P(+) cells with an AP-1 collagen-luciferase reporter construct, we found that 16:0, 18:1(n-9), 18:1(n-7), 18:2(n-6) and 20:4(n-6) NAEs stimulated AP-1-dependent transcriptional activity up to 2-fold, with maximal stimulation at approx. 10-15 microM. Higher NAE concentrations had toxic effects mediated by alterations in mitochondrial energy metabolism. The AP-1 stimulation appeared to be mediated by ERK but not JNK or p38 signalling pathways, because all NAEs stimulated ERK1/ERK2 phosphorylation without having any effect on JNK or p38 kinases. Also, overexpression of dominant negative ERK1/ERK2 kinases completely abolished NAE-induced AP-1 activation. In contrast with 18:1(n-9) NAE and anandamide, the cannabinoid receptor agonist WIN 55,212-2 did not stimulate AP-1 activity and inhibited ERK phosphorylation. The NAE-mediated effects were not attenuated by pertussis toxin and appeared to be NAE-specific, as a close structural analogue, oleyl alcohol, failed to induce ERK phosphorylation. The data support our hypothesis that the major saturated and monounsaturated NAEs are signalling molecules acting through intracellular targets without participation of cannabinoid receptors.

It has long been known that N-acylethanolamine phospholipids [N-acylphosphatidylethanolamine (N-acyl PE)] and N-acylethanolamines (NAEs) accumulate in mammalian tissues undergoing degenerative membrane changes associated with necrosis. Here we studied the effects of stress factors (UVB irradiation and serum deprivation) on the endogenous levels of N-acyl PE and NAE in mouse epidermal JB6 P(+) cells. We found that 16:0, 18:0, 18:1,n-9 and 18:1,n-7 are the predominant amide-linked fatty acids in both N-acyl PE and NAE in these cells. UVB irradiation and serum deprivation resulted in significantly increased levels of N-acyl PE and NAE, especially 18:1, n-9 N-acyl PE and NAE. UVB challenge increased the cellular content of anandamide (20:4,n-6 NAE), but this increase was the lowest of all NAEs measured. Serum deprivation resulted in a decreased cellular anandamide level, as well as a decrease in 20:4,n-6 N-acyl PE. Interestingly, the replacement of serum-free medium with medium containing 5% (v/v) fetal calf serum after 36 h of serum deprivation restored N-acyl PE and NAE levels almost completely within 4-8 h. These data suggest the involvement of N-acyl PE and NAE in cellular responses to stress.

Phosphatidylserine (PtdSer) is synthesized in mammalian cells by two base-exchange enzymes: PtdSer synthase (PSS)-1 primarily uses phosphatidylcholine as a substrate for exchange with serine, whereas PSS2 uses phosphatidylethanolamine (PtdEtn). We previously expressed murine PSS1 in McArdle hepatoma cells. The activity of PSS1 in vitro and the synthesis of PtdSer and PtdSer-derived PtdEtn were increased, whereas PtdEtn synthesis from the CDP-ethanolamine pathway was inhibited [Stone, Cui and Vance (1998) J. Biol. Chem. 273, 7293-7302]. We have now cloned and stably expressed a murine PSS2 cDNA in McArdle cells and M.9.1.1 cells [which are ethanolamine-requiring mutant Chinese hamster ovary (CHO) cells defective in PSS1]. Expression of the PSS2 in M.9.1.1 cells reversed the ethanolamine auxotrophy. However, the PtdEtn content was not normalized unless the culture medium was supplemented with ethanolamine. In both M.9.1.1 and hepatoma cells transfected with PSS2 cDNA the rate of synthesis of PtdSer and PtdSer-derived PtdEtn did not exceed that in parental CHO cells or control McArdle cells respectively, in contrast to cells expressing similar levels of murine PSS1. These observations suggest that PtdSer synthesis via murine PSS2, but not PSS1, is regulated by end-product inhibition. Moreover, expression of murine PSS2 in McArdle cells did not inhibit PtdEtn synthesis via the CDP-ethanolamine pathway, whereas expression of similar levels of PSS1 activity inhibited this pathway by approx. 50%. We conclude that murine PSS1 and PSS2, which are apparently derived from different genes, independently modulate phospholipid metabolism. In addition, mRNAs encoding the two synthases are differentially expressed in several murine tissues, supporting the idea that PSS1 and PSS2 might perform unique functions.

Cytochrome P-450 CYP2D6, human debrisoquine hydroxylase, metabolizes more than 30 prescribed drugs, the vast majority of which are small molecules containing a basic nitrogen atom. In contrast, the similar mouse protein Cyp2d-9 was first characterized as a testosterone 16alpha-hydroxylase. No common substrates have been reported for the two enzymes. Here we investigate the structural basis of this difference in substrate specificity. We have earlier used a combination of NMR data and homology modelling to generate a three-dimensional model of CYP2D6 [Modi, Paine, Sutcliffe, Lian, Primrose, Wolf, C.R. and Roberts (1996) Biochemistry 35, 4541-4550]. We have now generated a homology model of Cyp2d-9 and compared the two models to identify specific amino acid residues that we believe form the substrate-binding site in each protein and therefore influence catalytic selectivity. Although there are many similarities in active site structure, the most notable difference is a phenylalanine residue (Phe-483) in CYP2D6, which in the model is located such that the bulky phenyl ring is positioned across the channel mouth, thus limiting the size of substrate that can access the active site. In Cyp2d-9, the corresponding position is occupied by an isoleucine residue, which imposes fewer steric restraints on the size of substrate that can access the active site. To investigate whether the amino acid residue at this position does indeed influence the catalytic selectivity of these enzymes, site-directed mutagenesis was used to change Phe-483 in CYP2D6 to isoleucine and also to tryptophan. CYP2D6, Cyp2d-9 and both mutant CYP2D6 proteins were co-expressed with NADPH cytochrome P-450 reductase as a functional mono-oxygenase system in Escherichia coli and their relative catalytic activities towards bufuralol and testosterone were determined. All four proteins exhibited catalytic activity towards bufuralol but only Cyp2d-9 catalysed the formation of 16alpha-hydroxytesterone. Uniquely, the CYP2D6F483I mutant acquired the ability to metabolize testosterone to a novel product, which was identified by MS and proton NMR spectroscopy as 15alpha-hydroxytestosterone. NMR spin relaxation experiments were used to measure distances between the haem iron and protons of testosterone bound to the CYP2D6F483I mutant. These experiments demonstrate that very minor modifications to the active site structure of CYP2D6 can have a profound influence on the substrate specificity of the enzyme.

The tumour-promotion-sensitive (P+) and -resistant (P-) variants of mouse JB6 epidermis-derived cells have often been used to study the requirements for the tumour-promoting effect of PMA. As part of an effort to identify the defect(s) in JB6 P- cells that might prevent the promoting effect of PMA, stimulation of phospholipase D (PLD)-mediated hydrolysis of phosphatidylcholine (PtdCho) and phosphatidylethanolamine (PtdEtn) by PMA as well as the rate of phospholipid synthesis were compared in three P+ variants, two P- variants and a transformed variant of the JB6 cell line. PMA (5-100 nM) had significantly less stimulatory effect on PtdCho hydrolysis in P- cells than in P+ or transformed JB6 cells. The effects of PMA on PtdEtn hydrolysis in the P+ and P- cell lines were similar, whereas in transformed cells PMA had slightly less effect. Each JB6 cell line was found to contain similar amounts of PtdCho. In contrast, P- cells contained significantly less PtdEtn and a correspondingly higher level of ethanolamine phosphate compared with P+ and transformed cells. P- cells also secreted ethanolamine phosphate into the medium; this process was greatly enhanced by PMA. In the two P- variants the synthesis of PtdEtn from [14C]ethanolamine was reduced to various extents, whereas the rate of PtdCho synthesis was comparable in each JB6 cell line. The synthesis of PtdCho, but not PtdEtn, was greatly stimulated by PMA in both the P+ and P- clones. The results indicate that decreased synthesis/level of PtdEtn and suboptimal functioning of a PtdCho-specific PLD are common characteristics of the P- JB6 cells examined so far. The observed alterations in phospholipid metabolism may play a role in the resistance of P- cells to the tumour-promoting action of PMA.

Chinese hamster ovary (CHO) cells express only a trace amount of phosphatidylethanolamine N-methyltransferase (PEMT) activity. CHO cells make their phosphatidylcholine (PC) via the CDP-choline pathway. We investigated whether or not overexpression of PEMT2, an isoform of PEMT, in these cells would down-regulate the activity of the CDP-choline pathway. Transfection of CHO cells with PEMT2 cDNA behind the cytomegalovirus promoter resulted in a series of cell lines that overexpressed PEMT2. Phospholipid metabolism was characterized in cell lines that expressed a medium (281 pmol/min per mg of protein) and a high (1300 pmol/min per mg of protein) level of PEMT activity. The expression of the regulated enzyme (CTP:phosphocholine cytidylyltransferase) in the CDP-choline pathway was increased, not decreased, in these cell lines as judged by immunoblot analysis and enzymic activity. Conversion of phosphatidylethanolamine to PC was enhanced in CHO cells that expressed PEMT2 activity. PC mass was not increased in the transfected compared with the control cells. The rate of PC catabolism made by either the CDP-choline or methylation pathways was unaffected by PEMT2 expression. We conclude that expression of PEMT2 in CHO cells does not down-regulate, but rather enhances, the expression of CTP:phosphocholine cytidylyltransferase.

Anandamide (arachidonoylethanolamide, AnNH) has been recently proposed as the endogenous ligand at the brain cannabinoid receptor CB1. Two alternative pathways have been suggested for the biosynthesis of this putative mediator in the central nervous system. Here we present data (1) substantiating further the mechanism by which AnNH is produced by phospholipase D (PLD)-catalysed hydrolysis of N-arachidonoylphosphatidylethanolamine in mouse neuroblastoma N18TG2 cells, and (2) suggesting for the first time that AnNH is biosynthesized via the same mechanism in a non-neuronal cell line, mouse J774 macrophages, together with other acylethanolamides and is possibly involved in the control of the immune/inflammatory response. Lipids from both neuroblastoma cells and J774 macrophages were shown to contain a family of N-acylphosphatidylethanolamines (N-aPEs), including the possible precursor of AnNH, N-arachidonoyl-PE. Treatment with exogenous PLD, but not with exogenous phospholipase A2 and ethanolamine, resulted in the production of a series of acylethanolamides (AEs), including AnNH, from both cell types. The formation of AEs was accompanied by a decrease in the levels of the corresponding N-aPEs. Enzymically active homogenates from either neuroblastoma cells or J774 macrophages were shown to convert synthetic N-[3H]arachidonoyl-PE into [3H]AnNH, thus suggesting that in both cells an enzyme is present which is capable of catalysing the hydrolysis of N-aPE(s) to the corresponding AE(s). Finally, as previously shown in central neurons, on stimulation with ionomycin, J774 macrophages also produced a mixture of AEs including AnNH and palmitoylethanolamide, which has been proposed as the preferential endogenous ligand at the peripheral cannabinoid receptor CB2 and, consequently, as a possible down-modulator of mast cells. On the basis of this as well as previous findings it is now possible to hypothesize for AnNH and palmitoylethanolamide, co-synthesized by macrophages, a role as peripheral mediators with multiple actions on blood cell function.

The generation of the lipid signalling molecules, diacylglycerol (DAG) and phosphatidic acid (PA), has been implicated in the transduction events essential for proliferation of murine B6SUt.EP stem cells responsive to erythropoietin (EPO). Some of the responses were rapid and transient while others were slower and sustained. In an attempt to better understand the biphasic nature of DAG and PA appearance in response to EPO, an analysis of the molecular species of DAG, phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylinositol (PI), and PA in control and EPO-treated B6SUt.EP cells was made by HPLC and TLC. Fifteen to eighteen species were identified, which were increased non-uniformly by 0.2 unit/ml EPO. Greater increases (x6) were observed in 16:0,20:4 and 18:0,20:4 DAGs than in other species. The molecular species profiles of the stimulated DAGs were compared with the profiles of molecular species contained in the phospholipids. Comparison of the increase in DAG species caused by EPO with the molecular species present in PC and PI showed both PI and PC as the source of the fast DAG accumulation and only PC as the source of the slow DAG accumulation. PE was involved in both phases. We found a consistent formation of ethanolamine over time, in larger amounts than choline, providing strong evidence that, in addition to PC, PE is a major substrate. In addition, changes in molecular species of PA in response to EPO suggest that PI cannot account for the mass of PA formed during the first 30 s incubation with EPO, nor for PA formed during 30 min with EPO. It is concluded that the majority of PA was formed by a direct action of phospholipase D on PC.

1. The comparative study of the effect of bradykinin (BK) in young and old IMR-90 human fibroblasts shows that old cells are characterized by a reduced increase in 1,2-diacylglycerol (1,2-DAG) generation upon stimulation after short-term treatment and a significant higher increase after long-term agonist treatment. BK-induced activation of phospholipase D (PLD), the major enzyme involved in sustained 1,2-DAG generation, was 2.5-fold higher in old cells, strongly suggesting that it is involved in the potentiated increase of 1,2-DAG formation. The increased activation of PLD by BK in old cells was specific, since in parallel experiments the effect of thrombin was not significantly different in young and old cells. PLD activity in old cells was only reduced by down-regulation of protein kinase C (PKC) activity, in contrast to what was observed in young cells where it was completely abolished. This indicates that the enzyme activity in old cells was partially PKC-independent. BK was also able to increase the release of [14C]ethanolamine, a water-soluble product of hydrolysis of phosphatidylethanolamine (PtdEtn), through PLD activation in young and old cells. The BK effect was significantly higher in old cells and, very likely, PKC-independent, since phorbol 12-myristate 13-acetate failed to induce PtdEtn hydrolysis. 2. The present results indicate that the PLD/1,2-DAG pathway is specifically potentiated by BK in old fibroblasts, demonstrating that the formation of positive effectors of PKC activation is not necessarily decreased in cellular senescence. It remains to be established whether the increased generation of DAG upon BK stimulation plays any role in the altered PKC signalling pathway which characterizes old fibroblasts.

Evidence is presented for the operation of an ethanolamine-phosphatidylethanolamine (PtdEtn) cycle in Chinese hamster ovary cells. PtdEtn was labelled with [3H]ethanolamine and radioactivity was chased by incubation with 1 mM unlabelled ethanolamine. Radioactivity in [3H]PtdEtn gradually declined over a 23 h time period. In contrast, when the cells were incubated in medium lacking unlabelled ethanolamine, radioactivity in PtdEtn remained constant for at least 23 h. These observations suggest that the ethanolamine moiety is continuously released from PtdEtn and recycled back into PtdEtn. In cells incubated without unlabelled ethanolamine, labelled ethanolamine released from PtdEtn is re-incorporated into PtdEtn without significant dilution. In contrast, in cells incubated with unlabelled ethanolamine the specific radioactivity of the intracellular ethanolamine pool decreases as a result of dilution by the exogenous ethanolamine, hence radioactivity in PtdEtn gradually declines. Similar results were obtained for confluent and non-confluent cells. Our data also demonstrate that when PtdEtn is derived from phosphatidylserine decarboxylation, the ethanolamine cycle operates only in actively dividing, and not in confluent, cells, implying that PtdEtn derived from different biosynthetic origins [i.e. from decarboxylation of phosphatidylserine or from ethanolamine (most likely via the CDP-ethanolamine pathway)] is metabolized differently.

Olive (Olea europaea L.) callus cultures were incubated with [2-14C]ethanolamine and [Me-14C]choline in order to study phospholipid synthesis. Radioactivity from [Me-14C]choline was shown to be incorporated into the phosphatidylcholine via the CDP-base pathway. [2-14C]Ethanolamine was primarily incorporated into phosphatidylethanolamine, but significant radio-activity was also detected in phosphatidylcholine, indicating the operation of a methylation route. Incubation with [2-14C]ethanolamine indicated that phosphatidylcholine and phosphatidylethanolamine incorporated radiolabel over a similar time course. This led us to investigate the possibility that phosphatidylcholine was being synthesized by a methylation pathway distinct from the direct methylation of phosphatidylethanolamine. There was extensive incorporation of [2-14C]ethanolamine into different components of the aqueous phase of the incubations, within which phospho-base derivatives of ethanolamine were prominent. These intermediates were identified and provided evidence for the operation of an alternative methylation pathway via phosphodimethylethanolamine for the biosynthesis of phosphatidylcholine in olives.

We have investigated whether the growth requirement of keratinocytes for ethanolamine is due to defective synthesis of ethanolamine phosphoacylglycerols (EPG) via decarboxylation of serine phosphoacylglycerols. Proliferating keratinocytes readily incorporated [3H]ethanolamine into phosphatidylethanolamine (PE) and [3H]serine into phosphatidylserine (PS) and PE. Non-proliferating keratinocytes in ethanolamine-free medium incorporated [3H]glycerol into phosphatidylcholine (PC), PS and PE in decreasing order of label incorporated. The order of decreasing incorporation of glycerol after addition of ethanolamine to the medium was PC > PE > PS. Incubation of non-proliferating keratinocytes with [3H]serine resulted in incorporation of label into PS and PE. The extent of incorporation of [3H]serine into PS in non-proliferating keratinocytes was not less than that in proliferating cells. Addition of ethanolamine to the medium of non-proliferating keratinocytes did not change the quantity of label incorporated into PS, but resulted in a decrease of label incorporated into PE. When cells were prelabelled overnight with [3H]serine and subsequently incubated in medium containing ethanolamine, the loss of label from PS was inhibited relative to that of control cells incubated in medium without ethanolamine. The activity of PS decarboxylase activity in keratinocyte mitochondria was inhibited by phosphoethanolamine and PE, but not by ethanolamine or CDP-ethanolamine. Both proliferating and non-proliferating keratinocytes incorporated [3H]serine into ether-linked ethanolamine phospholipids. Taken together, the above results suggest that (1) both proliferating and non-proliferating keratinocytes are able to synthesize PE and ether-linked ethanolamine phospholipids from serine, and therefore the ethanolamine-requirement of the cells is not due to a defective decarboxylase pathway; (2) any inability of the decarboxylase pathway to meet cellular EPG requirement is not due to decreased synthesis of serine phospholipids; (3) synthesis of PE via decarboxylation, the major route in nonproliferating keratinocytes, appears to decrease when ethanolamine is made available and the CDP-ethanolamine pathway is functioning; (4) phosphoethanolamine and increased PE produced from the CDP-ethanolamine pathway may inhibit PS decarboxylase activity in the cells and provide a means of coordinating the synthesis of PE by the two pathways to prevent excess production.

Hepatic phosphatidylcholine (PC) from the immature fetal guinea pig at day 55 of gestation comprised mainly unsaturated molecular species containing C18:2(n-6) and C22:6(n-3) at the sn-2 position, reflecting placental permeability to essential fatty acids. At both day 55 and term (day 68), [Me-14C]choline was incorporated in utero over 3 h largely into sn-1-C16:0 PC species, with incorporation into sn-1-C18:0 PC species increasing by 18 h of incubation. Comparison of specific radioactivities after 3 h and 18 h suggests PC acyl remodelling by phospholipase A1. No incorporation into C20:4(n-6)-containing PC species could be detected of either [Me-14C]choline in vivo or CDP-[Me-14C]choline in isolated microsomes. The major phosphatidylethanolamine (PE) species were 16:0/22:6 and 18:0/22:6. Although [14C]ethanolamine was initially incorporated mainly into sn-1-C16:0 species, specific-radioactivity analysis suggested differential turnover rather than acyl remodelling. [1,2-14C]Ethanolamine and [Me-14C]methionine incorporation into PC molecular species indicated that both newly synthesized and total PE pools were available for N-methylation. Since the PC pool synthesized from PE included C20:4- and C22:6-containing species, N-methylation may provide a mechanism for supplying essential long-chain fatty acids to developing tissues that can be regulated independently from bulk PC synthesis.

The presence of cytosolic S-adenosylmethionine-dependent N-methyltransferase(s) activity(ies) capable of converting phosphoethanolamine into phosphocholine has been recently demonstrated in the rat brain. At least two enzymes are involved in the methylation of phosphoethanolamine to phosphocholine and these are separable by ammonium sulphate fractionation. The enzyme catalysing the last step of this methylation process is present in the 50-80% ammonium sulphate fraction. A 220-fold purified enzyme has been obtained with sequentially employed Q-Sepharose fast flow and octyl-Sepharose CL4B column chromatography. The maximum enzyme activity was at pH 9.5. The Km values for S-adenosylmethionine, the methyl donor, and phosphodimethylethanolamine, the methyl acceptor, were 125 microM and 750 microM respectively. This phosphodimethylethanolamine N-methyltransferase was found to be calcium-dependent, with a 4-fold increase in activity at 0.5 mM-CaCl2. S-Adenosylhomocysteine at 0.5 mM caused 100% inhibition of the activity. The effects of various structural analogues on the phosphodimethylethanolamine N-methyltransferase activity were also investigated and these results suggest that the enzyme is specific to the substrate. These results provide evidence for the existence of the pathway for the methylation of phosphoethanolamine to phosphocholine in rat brain cytosol.

The effect of partial (70%) hepatectomy on phosphatidylethanolamine (PE) synthesis was studied in rat liver during the first 4 post-operative days. Between 4 and 96 h after partial hepatectomy, the mass of PE increased from 30% to 80% of sham-operation values. In line with the increase in PE mass, the rate of PE synthesis in vivo from [14C]ethanolamine was stimulated 1.6- and 1.3-fold at 22 and 48 h after partial hepatectomy respectively. Surprisingly, the activity of CTP:phosphoethanolamine cytidylyltransferase (EC 2.7.7.14) was virtually unchanged after partial hepatectomy. In addition, neither ethanolamine kinase (EC 2.7.1.82) nor ethanolaminephosphotransferase (EC 2.7.8.1) showed any changes in activity over the time period studied. Hepatic levels of ethanolamine and phosphoethanolamine were drastically increased after partial hepatectomy, as compared with sham operation, whereas levels of CDP-ethanolamine and microsomal diacylglycerol were not affected. Interestingly, partial hepatectomy caused the concentration of free ethanolamine in serum to increase from 29 microM to approx. 50 microM during the first day after surgery. In hepatocytes isolated from non-operated animals, incorporation of [3H]ethanolamine into PE was stimulated by increasing the ethanolamine concentration from 10 up to 50 microM, whereas the radioactivity associated with phosphoethanolamine only increased at ethanolamine concentrations higher than 30 microM. Taken together, our results indicate that the observed increase in serum ethanolamine concentration after partial hepatectomy is probably responsible for both the increase in PE biosynthesis and the accumulation of ethanolamine and phosphoethanolamine in regenerating liver.

Quinine, 4-aminopyridine and tetraethylammonium, three compounds generally used as effectors of K+ channels, strongly modify phospholipid metabolism. In the human monocytic cell line THP1, the three drugs enhanced the incorporation of [3H]serine into phosphatidylserine and that of [3H]inositol into phosphatidylinositol in the absence of significant changes in the uptake of the 3H labels. On the contrary, the biosynthesis of both phosphatidylcholine and phosphatidylethanolamine was strongly inhibited. This inhibition appeared to be mainly due to the inhibition of both [3H]choline and [3H]ethanolamine uptake by the cells, by impairment of choline transport in a competitive mode.

Studies with mammalian cell lines have led to suggestions that mammalian tissues may derive all of their phosphatidylethanolamine (PE) from the decarboxylation of phosphatidylserine (PS), and also that the physiological significance of the CDP-ethanolamine pathway was the synthesis of ethanolamine plasmalogen. We have therefore investigated the biosynthesis of PE and ethanolamine plasmalogen via the CDP-ethanolamine and decarboxylation pathways in vivo in three rat tissues (heart, kidney and liver), which differ in ethanolamine plasmalogen content. In all three tissues [14C]ethanolamine was incorporated into both PE and ethanolamine plasmalogen, whereas [3H]serine was incorporated into only PS and PE fractions. When [14C]ethanolamine was introduced into the animals, the specific radioactivity of ethanolamine plasmalogen in the kidney was always greater than that of the PE fraction; in the heart the specific radioactivity of the ethanolamine plasmalogen fraction was similar to that of the PE fraction, whereas in the liver the specific radioactivity of the PE fraction was always greater than that of the ethanolamine plasmalogen fraction. The results obtained in this study indicate that: (1) the CDP-ethanolamine pathway is utilized for the synthesis of both PE and ethanolamine plasmalogen in all three tissues; (2) the decarboxylation pathway is utilized solely for the synthesis of PE; (3) serine plasmalogens are not formed by base-exchange reactions; (4) the relative utilization of the CDP-ethanolamine pathway for the synthesis of PE and ethanolamine plasmalogen varies among tissues. Our studies also revealed that the hypolipidaemic drug MDL 29350 is a potent inhibitor of PE N-methyltransferase activity in vitro and in vivo.

The specificity of the phospholipid head-group for feedback regulation of CTP: phosphocholine cytidylyltransferase was examined in rat hepatocytes. In choline-deficient cells there is a 2-fold increase in binding of cytidylyltransferase to cellular membranes, compared with choline-supplemented cells. Supplementation of choline-deficient cells with choline, dimethylethanolamine, monomethylethanolamine or ethanolamine resulted in an increase in the concentration of the corresponding phospholipid. Release of cytidylyltransferase into cytosol was only observed in hepatocytes supplemented with choline or dimethylethanolamine. The apparent EC50 values (concn. giving half of maximal effect) for cytidylyltransferase translocation were similar for choline and dimethylethanolamine (25 and 27 microM respectively). The maximum amount of cytidylyltransferase released into cytosol with choline supplementation (1.13 m-units/mg membrane protein) was twice that (0.62) observed with dimethylethanolamine. Supplementation of choline-deficient hepatocytes with NN'-diethylethanolamine, N-ethylethanolamine or 3-aminopropanol also did not cause release of cytidylyltransferase from cellular membranes. The translocation of cytidylyltransferase appeared to be mediated by the concentration of phosphatidylcholine in the membranes and not the ratio of phosphatidylcholine to phosphatidylethanolamine. The results provide further evidence for feedback regulation of phosphatidylcholine biosynthesis by phosphatidylcholine.

1. The anaerobic rumen protozoon Entodinium caudatum was incubated either intact or with various radioactive precursors of phospholipids after ultrasonication. 2. Pulse-chase experiments showed a rapid turnover of phosphatidylinositol and much slower turnovers of phosphatidylethanolamine and phosphatidylcholine. 3. E. caudatum imbibed choline very rapidly; this was immediately and exclusively converted into phosphatidylcholine which was shown by radioautography after 10 min to be distributed throughout the cell membranes. 4. Phosphatidylcholine was synthesized through a phosphorylcholine-CDP-choline pathway, the methylation or base-exchange pathways not being present. 5. Under suitable conditions [Me-14C]choline can be substantially (50-60%) converted into CDP-choline by sonicated E. caudatum and this provides an excellent method of preparing this biosynthetic intermediary. 6. [2-14C]Ethanolamine was taken up much less readily than choline. The former was incorporated into phosphatidylethanolamine by the CDP-ethanolamine pathway. 7. Doubly labelled [32P]phosphatidyl[2-3H]ethanolamine was converted into ceramide phosphorylethanolamine and N-(1-carboxyethyl)phosphatidyl-ethanolamine, without change in the isotopic ratio. Ceramide phosphoryl [2-14C]-ethanolamine was converted into phsophatidylethanolamine. 8. Palmitic acid, oleic acid and linoleic acid were taken by E. caudatum cells and incorporated into phospholipids. By contrast, although stearic acid was taken up it was hardly incorporated into phospholipids.

Biosynthetic labelling experiments with cercariae and schistosomula of the multicellular parasitic trematode Schistosoma mansoni were performed to determine whether [3H]palmitate or [3H]ethanolamine was incorporated into proteins. Parasites incorporated [3H]palmitate into numerous proteins, as judged by SDS/polyacrylamide-gel electrophoresis and fluorography. The radiolabel was resistant to extraction with chloroform, but sensitive to alkaline hydrolysis, indicating the presence of an ester bond. Further investigation of the major 22 kDa [3H]palmitate-labelled species showed that the label could be recovered in a Pronase fragment which bound detergent and had an apparent molecular mass of 1200 Da as determined by gel filtration on Sephadex LH-20. Schistosomula incubated with [3H]ethanolamine for up to 24 h incorporated this precursor into several proteins; labelled Pronase fragments recovered from the three most intensely labelled proteins were hydrophilic and had a molecular mass of approx. 200 Da. Furthermore, reductive methylation of such fragments showed that the [3H]ethanolamine bears a free amino group, indicating the lack of an amide linkage. We also evaluated the effect of phosphatidylinositol-specific phospholipase C from Staphylococcus aureus: [3H]palmitate-labelled proteins of schistosomula and surface-iodinated proteins were resistant to hydrolysis with this enzyme. In conclusion, [3H]palmitate and [3H]ethanolamine are incorporated into distinct proteins of cercariae and schistosomula which do not bear glycophospholipid anchors. The [3H]ethanolamine-labelled proteins represent a novel variety of protein modification.

1. Kinetic studies of ethanolamine ammonia-lyase formation by Escherichia coli suggested that coenzyme B12 (5'-deoxyadenosylcobalamin), with ethanolamine, is a co-inducer. 2. Enzymic and immunological tests failed to show the formation of complementary enzyme components induced separately by ethanolamine and cobalamin respectively. 3. Although specific for ethanolamine as the substrate, enzyme formation was induced by certain analogues, e.g. 2-aminopropan-1-ol. 4. Experiments with cyano[57Co]-cobalamin suggested that neither coenzyme B12 nor some more tightly bound coenzymically inactive cobamide was necessary for enzyme stability in vitro. 5. Mutants of E. coli were obtained which formed ethanolamine ammonia-lyase apoenzyme constitutively, showing that neither ethanolamine nor cobalamin was required for assembly or post-transcriptional stability of the enzyme in vivo. Constitutive enzyme formation was subject to catabolite repression, particularly by glucose. 6. It appears likely that ethanolamine and coenzyme B12, acting in concert, induce ethanolamine ammonia-lyase formation. The term 'concerted' induction is proposed for this phenomenon.

The effects of the potent tumour-promoting phorbol ester 12-O-tetradecanoylphorbol 13-acetate (TPA) on phosphatidylcholine (PtdCho) metabolism were investigated in the neuroblastoma X glioma hybrid cell line NG108-15. TPA (100 nM) stimulated by 150-200% the release into the medium of 3H radioactivity from cells that had been pre-labelled with [3H]choline. H.p.l.c. analysis of the medium revealed that TPA stimulated the release of only free [3H]choline (212 +/- 11% of control), without affecting such other labelled metabolites as [3H]phosphocholine and [3H]glycerophosphocholine. This effect was concentration-dependent, with a half-maximal effect obtained at 27.5 +/- 6.8 nM, and was observable as early as 5-10 min after exposure to TPA. The TPA-induced release of [3H]choline into the medium was accompanied by a small and variable decrease in cellular [3H]PtdCho (to 93 +/- 4% of control). However, the radioactivity associated with water-soluble cellular choline metabolites (mainly [3H]phosphocholine and [3H]glycerophosphocholine) remained unchanged. TPA also stimulated the release of [3H]choline derived from [3H]PtdCho that had been produced via the methylation pathway from [3H]methionine. These data suggest that phosphatidylcholine may serve as the source of free choline released from the cells in response to TPA. The possible enzymic mechanisms underlying this response are discussed.

1. The addition of mitochondria to an incubation system containing the soluble and microsomal fractions of rat liver enhances severalfold the incorporation of each of ethanolamine, phosphorylethanolamine and CDP-ethanolamine into phosphatidylethanolamine. 2. In the presence of microsomal, mitochondrial and soluble fractions, CDP-ethanolamine exhibits the greatest initial rate of incorporation (approx. 6nmol/h per mg of protein), being slightly faster than that of phosphorylethanolamine (approx. 5nmol/h per mg of protein). Incorporation of ethanolamine proceeds very slowly for the first 20min and only after 30min gives rates approaching those of the other two precursors. 3. By using a substrate ;dilution' technique it was shown that in the reconstituted system the affinity of each of the enzymes for their respective substrates is very high: 10mum for ethanolamine, 25mum for phosphorylethanolamine and 5mum for CDP-ethanolamine. 4. Isolation of the mitochondrial and microsomal fractions from the medium after incubation together with phosphorylethanolamine showed that about 70% of the total radioactivity was present in the microsomal fraction and about 30% in the mitochondria after only 20min. Similar experiments with ethanolamine as precursor revealed that after 20min only about 15% of the total radioactivity was present in the mitochondria but that after 40min about 30% was present in this fraction. 5. Heating and phospholipase treatment of mitochondria, but not freeze-thawing, eliminated the stimulatory effect of mitochondria on phospholipid synthesis. 6. The reconstituted system exhibits an absolute requirement for Mg(2+) (2mm gave maximal rates) and is inhibited by very low concentrations of Ca(2+) (100mum-Ca(2+) produced half-maximal inhibition with 3mm-Mg(2+)). Further addition of Mg(2+) overcame the Ca(2+) inhibition, suggesting that the inhibitory effect is readily reversible. 7. The concept that modification of the Mg(2+)/Ca(2+) ratio is a means of controlling the rate of cellular phospholipid synthesis is introduced.

1. Analogues of ethanolamine and choline were incubated with different labelled precursors of phospholipids and isolated hepatocytes and the effects on phospholipid synthesis were studied. 2. 2-Aminopropan-1-ol and 2-aminobutan-1-ol were the most efficient inhibitors of [(14)C]ethanolamine incorporation into phospholipids, whereas the incorporation of [(3)H]choline was inhibited most extensively by NN-diethylethanolamine and NN-dimethylethanolamine. 3. When the analogues were incubated with [(3)H]glycerol and hepatocytes, the appearance of (3)H in unnatural phospholipids indicated that they were incorporated, at least in part, via CDP-derivatives. The distribution of [(3)H]glycerol among molecular species of phospholipids containing 2-aminopropan-1-ol and 1-aminopropan-2-ol was the same as in phosphatidylethanolamine. In other phospholipid analogues the distribution of (3)H was more similar to that in phosphatidylcholine. 4. NN-Diethylethanolamine stimulated both the conversion of phosphatidylethanolamine into phosphatidylcholine and the incorporation of [Me-(14)C]methionine into phospholipids. Other N-alkyl- or NN-dialkyl-ethanolamines also stimulated [(14)C]methionine incorporation, but inhibited the conversion of phosphatidylethanolamine into phosphatidylcholine. This indicates that phosphatidyl-NN-diethylethanolamine is a poor methyl acceptor, in contrast with other N-alkylated phosphatidylethanolamines. 5. These results on the regulation of phospholipid metabolism in intact cells are discussed with respect to the possible control points. They also provide guidelines for future experiments on the manipulation of phospholipid polar-headgroup composition in primary cultures of hepatocytes.

Sphingomyelin synthesis was studied in slices of rat heart by using [Me-14C]choline, [1,2-14C]ethanolamine, S-adenosyl-L-[14C]methionine and [32P]Pi as as precursors. In the presence of both [Me-14C]choline and [32P]Pi the ratio of the specific radioactivities of 14C and 32P in phosphatidylcholine was greater than in sphingomyelin at all the times studied. This suggested that synthesis of phosphatidylcholine and sphingomyelin de novo did not involve the utilization of a common pool of cytidine diphosphate choline. In addition, studies with [1,2-14C]ethanolamine and S-adenosyl-L-[14C]methionine indicated that a quantitatively significant pool of choline, derived from these precursors, was selectively utilized for sphingomyelin formation. This pool was not represented by phosphatidylcholine formed by methylation of phosphatidylethanolamine or by other pathways.

The incubation of neurons from chick embryos in primary culture with [3H]ethanolamine revealed the conversion of this base into monomethyl, dimethyl and choline derivatives, including the corresponding free bases. Labelling with [methyl-3H]monomethylethanolamine and [methyl-3H]dimethylethanolamine supported the conclusion that in chick neuron cultures, phosphoethanolamine appears to be the preferential substrate for methylation, rather than ethanolamine or phosphatidylethanolamine. The methylation of the latter two compounds, in particular that of phosphatidylethanolamine, was seemingly stopped at the level of their monomethyl derivatives. Fetal rat neurons in primary culture incubated with [3H]ethanolamine showed similar results to those observed with chick neurones. However, phosphoethanolamine and phosphatidylethanolamine and, to a lesser extent, free ethanolamine, appeared to be possible substrates for methylation reactions. The methylation of water-soluble ethanolamine compounds de novo was further confirmed by experiments performed in vivo by intraventricular injection of [3H]ethanolamine. Phosphocholine and the monomethyl and dimethyl derivatives of ethanolamine were detected in the brain 15 min after injection.

1. The kinetics of phosphatidylcholine and phosphatidylethanolamine synthesis in rat liver were followed 5-60 min after the intraportal injection of [14-C]choline and [3-H]-ethanolamine. 2. At all time-intervals the specific radioactivity of CDP-choline was only about half that of phosphorylcholine. This indicated that CDP-choline was formed at a similar rate from phosphorylcholine and phosphatidylcholines, the latter probably through the reverse reaction of cholinephosphotransferase (EC 2.7.8.2.). In view of recent data obtained from experiments in vitro this implies a significant role for the cholinephosphotransferase reaction in the turnover of molecular species of phosphatidylcholine. 3. The specific radioactivity of CDP-ethanolamine was about twice that of phosphorylethanolamine at all time-intervals studied. This supports a previous suggestion that the liver phosphorylethanolamine pool is subject to compartmentation and shows that there is no rapid equilibration between different pools. In contrast with a recent study, no evidence was found for any significant methylation of phosphoryl-or CDP-ethanolamine to the corresponding choline derivative. 4. Quantitative data on the biosynthesis of molecular species of phosphoLIPIDS via CDP derivatives were calculated according to simple kinetic models. They were in the same range as those calculated from earlier data on precusors incorporated via diacylglycerols. 5. The proportion of radioactive phosphatidylethanolamines appearing in the plasma was approximately ten times lower than that for phosphatidylcholines. No selectivity was observed in the transfer into plasma of different molecular species of phosphatidylethanolamine.

1. The 120-fold purification of ethanolamine ammonia-lyase from Escherichia coli extracts, to apparent homogeneity, is described. Ethanolamine, dithiothreitol, glycerol and KCl protected the apoenzyme from inactivation. 2. At the optimum pH7.5, K(m) values for ethanolamine and coenzyme B(12) were 44mum and 0.42mum respectively. The K(m) for ethanolamine was markedly affected by pH, transitions occurring at pH7.0 and 8.35. 3. The enzyme was specific for ethanolamine as substrate, none of the 18 analogues tested being active. l-2-Aminopropan-l-ol (K(i) 0.86mum), dl-1-aminopropan-2-ol (K(i) 2.2mum) and dl-1,3-diaminopropan-2-ol (K(i) 88.0mum) inhibited competitively. 4. Enzyme activity was inhibited, irreversibly and non-competitively, by the coenzyme analogues methylcobalamin (K(i) 1.4nm), hydroxocobalamin (K(i) 2.1nm) and cyanocobalamin (K(i) 4.8nm). 5. Iodoacetamide inhibited in the absence of ethanolamine, but only slightly in its presence. p-Hydroxymercuribenzoate inhibited markedly even in the presence of ethanolamine. Dithiothreitol and 2-mercaptoethanol (less effectively) restored activity to the enzyme dialysed against buffer containing ethanolamine. 6. Although K(+) ions stabilized the enzyme during dialysis or storage, they were not necessary for activity. 7. Gel filtration showed the enzyme to be of high molecular weight, ultracentrifugal studies giving s(20,w) of 16.4 and an estimated mol.wt. 560400. The isoelectric point for the apoenzyme was approx. pH5.0. inhibited enzyme activity at concentrations above 1m (95% inhibition at 3m) and sodium dodecyl sulphate/polyacrylamide-gel electrophoresis indicated protein subunits of mol.wt. 61400. 8. Immunological studies showed that the E.coli enzyme was closely related to those of other enterobacteria, but only distantly to that of Clostridium sp. A double precipitin band suggested that the apoenzyme may be made up of two protein components.

1. Ten bacteria utilizing [2-14C]ethanol-2-amine as the sole or major source of nitrogen for growth on glycerol + salts medium incorporated radioactivity into a variety of bacterial substances. A high proportion was commonly found in lipid fractions, particularly in the case of Erwinia carotovora. 2. Detailed studies of [14C]ethanolamine incorporation into lipids by five bacteria, including E. carotovora, showed that all detectable lipids were labelled. Even where phosphatidylethanolamine was the major lipid labelled, radioactivity was predominantly in the fatty acid rather than the base moiety. The labelled fatty acids were identified in each case. 3. The addition of acetate to growth media decreased the incorporation of radioactivity from ethanolamine into both fatty acid and phosphatidyl-base fragments of lipids from all the bacteria except Mycobacterium smegmatis. Experiments with [3H]ethanolamine and [14C]acetate confirmed that unlabelled acetate decreased the incorporation of both radioactive isotopes into lipids, except in the case of M. smegmatis. 4. Enzyme studies suggested one of two metabolic routes between ethanolamine and acetyl-CoA for each of four bacteria. A role for ethanolamine O-phosphate was not obligatory for the incorporation of [14C]ethanolamine into phospholipids, but correlated with CoA-independent aldehyde dehydrogenase activity.

Ethanolamine phospholipid head groups in Paramecium were synthesized directly from ethanolamine. As in other cell types, radioactivity from ethanolamine failed to incorporate significantly into head groups of ethanolamine phosphonolipids, indicating that the phosphonolipids are not derived from their phospholipid analogues. Unlike other systems previously examined, radioactivity from serine is incorporated into both ethanolamine phospholipid and phosphonolipid head groups of glycerolipids and sphingolipids in this ciliate. These observations suggest that synthesis of ethanolamine phosphonolipids involves synthesis de novo of free phosphonoserine, which is then incorporated into lipids, and then lipid-bound phosphonoserine intermediates (glycerolipids or sphingolipids) undergo decarboxylation, forming lipidbound phosphonoethanolamine compounds.

The concentration of unesterified choline in the plasma in the jugular vein of the rat (0.85 nmol/ml) was found to be three times that of the arterial supply to the brain (0.25 nmol/ml), indicating a higher efflux than uptake of unesterified choline by the brain. No such difference was found for the rabbit and no arterio-venous difference for phosphatidylcholine or lysophosphatidylcholine was observed in either species. No arterio-venous difference was found for choline in blood cells. The infusion of [Me-3H]choline into the circulation of the rat or rabbit indicated an uptake of radioactive choline by the brain and an efflux of non-radioactive choline. In the rabbit such an infusion produced a steady rise in the labelling of phosphatidylcholine and lysophosphatidylcholine in the plasma. When [14C2]ethanolamine was injected intraperitoneally into the rat there was a labelling of phosphatidylcholine, lysophosphatidylcholine and sphingomyelin in the plasma and cells of blood from the jugular vein and the arterial supply, as well as in the brain tissue. However, no labelling of unesterified choline in these tissues could be detected. Unesterified choline was shown to be liberated into the plasma when whole blood from the rat or man, but not the rabbit, was incubated for short periods at 30 degrees C.

Exposure of isolated rat hepatocytes to glucagon or chlorophenylthio cyclic AMP led to an inhibition of the incorporation of [1,2-14C]ethanolamine into phosphatidylethanolamine. Pulse-chase experiments and measurement of the activities of the enzymes involved in the CDP-ethanolamine pathway provided evidence that the inhibitory effect of glucagon on the synthesis de novo of phosphatidylethanolamine was not caused by a diminished conversion of ethanolamine phosphate into CDP-ethanolamine. The observations suggested that the glucagon-induced inhibition of the biosynthesis of phosphatidylethanolamine is probably due to a decreased supply of diacylglycerols, resulting in a decreased formation of phosphatidylethanolamine from CDP-ethanolamine and diacylglycerols.

Changes in isoprenaline-sensitive phospholipid methyltransferase were studied in hepatocytes isolated from juvenile, mature and adrenalectomized rats. Isoprenaline produced greater stimulation of cyclic AMP accumulation in juvenile and mature adrenalectomized rats than in mature animals. Similarly, isoprenaline stimulated phospholipid methyltransferase in juvenile and mature adrenalectomized rats but had no effect in mature animals. Isoprenaline-mediated activation of phospholipid methyltransferase in adrenalectomized rats was time- and dose-dependent. In hepatocytes isolated from adrenalectomized rats incubated with [Me-3H]methionine or [3H]-ethanolamine the addition of isoprenaline increased the amount of radioactivity incorporated into phosphatidylcholine. The activation by isoprenaline of phospholipid methyltransferase was abolished by the beta-blocker propranolol and by insulin. These results indicate that rat liver the occupation of functional beta-receptors causes a stimulation of phospholipid methylation. It is suggested that, as reported previously, cyclic AMP activates phospholipid methyltransferase.

Each salivary gland contains about 135 pmol of phosphatidylinositol. In glands prelabelled by incubation for 1 h with [32P]Pi or [3H]inositol there was a subsequent breakdown of 80% of the labelled phosphatidylinositol over a 2 h incubation period with 10 micrometer-5-hydroxytryptamine. However, there was no detectable decrease either in total phosphatidylinositol based on phosphorus analysis by chemical estimation or in the radioactivity of [32P]phosphatidylinositol in salivary glands of flies raised from the larval stage on diets containing[32P]Pi and whose phospholipids were uniformly labelled. These results suggest that the pool of phosphatidylinositol involved with Ca2+ gating is a small fraction of the total phosphatidylinositol content. Furthermore it is this small compartment that is preferentially radioactively labelled during short-term incubations with radioactively labelled precursors. In salivary glands incubated for 2 h with 10 micrometer-5-hydroxytryptamine there was a marked decrease in the flux of 45Ca2+ across the gland. After removal of the hormone, incubation of salivary glands for 1 h in the presence of 2mM-inositol, but not choline or ethanolamine, resulted in a recovery of hormone-responsive 45Ca2+ flux. Quantitative studies revealed that less than 9 pmol of phosphatidylinositol must be formed to fully restoret he 5-hydroxytryptamine-responsive 45Ca2+ flux.

1. Incorporation of [Me-14C]choline and [2-14C]ethanolamine into lipids was studied in germinating soya bean (Glycine max L.) seeds. The precursors are only incorporated into phosphatidylcholine and into phosphatidylethanolamine respectively. 2. Base-labelling via a phospholipase-D type of reaction was eliminated as a significant factor. 3. Cyclo heximide inhibited labelling of phosphatidylcholine from [Me-14C]choline but did not affect labelling of the aqueous choline pool. It had no effect on [2-14C]ethanolamine uptake or incorporation into phosphatidylethanolamine. 4. Hemicholinium-15 at 10mM concentrations decreased uptake and lipid labelling from the both bases. 5. There was no evidence for base competition. 6. The endogenous pool of choline was much larger than that of ethanolamine, which resulted in higher specific radioactivities for phosphatidyl-ethanolamine than for phosphatidylcholine. 7. The results can be interpreted as indicating that the kinase and phosphoryltransferase enzymes of the CDP-base pathways are separate for each phospholipid.

Previous results have shown that 1,25-dihydroxycholecalciferol [1,25(OH)2D3] enhances the synthesis of phosphatidylserine (PS) and suppresses the synthesis of phosphatidylethanolamine (PE) in osteoblast-like rat osteogenic sarcoma UMR 106 cells [Matsumoto, Kawanobe, Morita & Ogata (1985) J. Biol. Chem. 260, 13704-13709]. In the present study, the effect of parathyroid hormone (PTH) on phospholipid metabolism is examined by using these cells. Treatment of UMR 106 cells with human PTH-(1-34)-peptide suppresses the synthesis of phosphatidylethanolamine in a dose- and time-dependent manner without affecting the synthesis of PS. The maximal effect on PE synthesis is obtained with 2.4 nM-human PTH-(1-34)-peptide when the cells are treated for 48 h or longer. In addition, when human PTH-(1-34)-peptide is added together with the maximal dose of 1,25(OH)2D3, there is a further decline in PE synthesis, whereas the stimulation of PS synthesis by 1,25(OH)2D3 is not altered. Because methylation of PE is suggested to affect hormone receptor-adenylate cyclase coupling, the observed change in PE metabolism by PTH and 1,25(OH)2D3 may be, at least in part, involved in the development of desensitization phenomenon to PTH in these cells.

1. Phenethylamines were extracted from brain and liver of rats with phenylketonuria-like characteristics produced in vivo by inhibition of phenylalanine hydroxylase (EC 1.14.3.1) with p-chlorophenylalanine, with or without phenylalanine administration. To protect amines against oxidation by monoamine oxidase, pargyline was also administered. 2. beta-Phenethylamine was the major compound found in brain and liver. beta-Phenethanolamine and octopamine were also present, in lesser amounts, and the concentrations of these three amines paralleled blood phenylalanine concentrations. By comparison, tissues from control animals had only very low concentrations of these amines. 3. Small amounts of normetadrenaline, m-tyramine and 3-methoxytyramine were also found. 4. The inhibitors used, p-chlorophenylalanine and pargyline, gave rise to p-chlorophenethylamine and benzylamine respectively, the first via decarboxylation, the second probably by breakdown during extraction. 5. Distribution of phenethylamines in different brain regions and in subcellular fractions of rat brain cells was also investigated. The content of phenethylamine was highest in the striatum. 6. These findings are discussed in the light of changes occurring in human patients with uncontrolled phenylketonuria.

1. Growth and manometric experiments showed that a Pseudomonas sp. P6 (N.C.I.B. 10431), formerly known as Achromobacter sp. P6, was capable of growth on both stereoisomers of 1-aminopropan-2-ol, and supported the hypothesis that assimilation involved metabolism to propionaldehyde, propionate and possibly 2-hydroxyglutarate. A number of alternative intermediary metabolites were ruled out. 2. Accumulation of propionaldehyde from 1-aminopropan-2-ol by intact cells occurred only during active growth, was transitory and was accompanied by morphological changes in the pseudomonad. 3. Enzymic and radioactive tracer evidence showed that 1-aminopropan-2-ol O-phosphate was the intermediate between amino alcohol and aldehyde. The operation of an inducibly formed ATP-amino alcohol phosphotransferase was established by measuring substrate disappearance, ADP formation and amino alcohol O-phosphate formation. This novel kinase had two activity peaks at about pH7 and 9. It acted on both l- and d-isomers of 1-aminopropan-2-ol, and also on l-threonine and ethanolamine, but had only low activity towards choline. The enzyme was partially purified by ion-exchange chromatography. 4. An amino alcohol O-phosphate phospho-lyase (deaminating) produced propionaldehyde from dl- and d-1-aminopropan-2-ol O-phosphate, and also formed acetaldehyde less rapidly from ethanolamine O-phosphate. It had optimum activity at about pH8 in Tris-HCl buffers. The enzyme was partially purified and evidence was obtained that a single enzyme was responsible for both activities. Apparent K(m) values for the substrates were determined. Activity was inhibited by dl-threonine O-phosphate, dl-serine O-phosphate, choline O-phosphate and P(i). Enzyme formation was induced by growth with either amino alcohol substrate. 5. Radioactive tracer experiments with dl-1-amino[3-(14)C]propan-2-ol confirmed the operation of the amino alcohol kinase and demonstrated coupling with the phospho-lyase enzyme in vitro to produce [(14)C]-propionaldehyde. 6. An aldehyde dehydrogenase, found in extracts of the pseudomonad after growth on 1-aminopropan-2-ol, was characterized and concluded to be responsible for propionaldehyde and acetaldehyde oxidation. The enzyme was inactive with methylglyoxal. 7. Propionate and acetate were concluded to be metabolized via propionyl-CoA and acetyl-CoA, and studies were made of a CoA ester synthase found in extracts. 8. Studies of a strain of Pseudomonas putida N.C.I.B. 10558 suggested that 1-aminopropan-2-ols were metabolized via their O-phosphates, propionaldehyde and propionate. Amino alcohol kinase activity was detected and extracts contained a phospho-lyase showing higher activity with the 1-aminopropan-2-ol O-phosphate than with ethanolamine O-phosphate.

1. Growth of Erwinia carotovora N.C.P.P.B. 1280 on media containing 1-aminopropan-2-ol compounds or ethanolamine as the sole N source resulted in the excretion of propionaldehyde or acetaldehyde respectively. The inclusion of (NH(4))(2)SO(4) in media prevented aldehyde formation. 2. Growth, microrespirometric and enzymic evidence implicated amino alcohol O-phosphates as aldehyde precursors. An inducibly formed ATP-amino alcohol phosphotransferase was partially purified and found to be markedly stimulated by ADP, unaffected by NH(4) (+) ions and more active with ethanolamine than with 1-aminopropan-2-ol compounds. Amino alcohol O-phosphates were deaminated by an inducible phospho-lyase to give the corresponding aldehydes. This enzyme, separated from the kinase during purification, was more active with ethanolamine O-phosphate than with 1-aminopropan-2-ol O-phosphates. Activity of the phospho-lyase was unaffected by a number of possible effectors, including NH(4) (+) ions, but its formation was repressed by the addition of (NH(4))(2)SO(4) to growth media. 3. E. carotovora was unable to grow with ethanolamine or 1-aminopropan-2-ol compounds as sources of C, the production of aldehydes during utilization as N sources being attributable to the inability of the microbe to synthesize aldehyde dehydrogenase. 4. Of seven additional strains of Erwinia examined similar results were obtained only with Erwinia ananas (N.C.P.P.B. 441) and Erwinia milletiae (N.C.P.P.B. 955).