Perturbations in choline metabolism cause neural tube defects in mouse embryos in vitro.

A role for choline during early stages of mammalian embryogenesis has not been established, although recent studies show that inhibitors of choline uptake and metabolism, 2-dimethylaminoethanol (DMAE), and 1-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine (ET-18-OCH3), produce neural tube defects in mouse embryos grown in vitro. To determine potential mechanisms responsible for these abnormalities, choline metabolism in the presence or absence of these inhibitors was evaluated in cultured, neurulating mouse embryos by using chromatographic techniques. Results showed that 90%-95% of 14C-choline was incorporated into phosphocholine and phosphatidylcholine (PtdCho), which was metabolized to sphingomyelin. Choline was oxidized to betaine, and betaine homocysteine methyltransferase was expressed. Acetylcholine was synthesized in yolk sacs, but 70 kDa choline acetyltransferase was undetectable by immunoblot. DMAE reduced embryonic choline uptake and inhibited phosphocholine, PtdCho, phosphatidylethanolamine (PtdEtn), and sphingomyelin synthesis. ET-18-OCH3 also inhibited PtdCho synthesis. In embryos and yolk sacs incubated with 3H-ethanolamine, 95% of recovered label was PtdEtn, but PtdEtn was not converted to PtdCho, which suggested that phosphatidylethanolamine methyltransferase (PeMT) activity was absent. In ET-18-OCH3 treated yolk sacs, PtdEtn was increased, but PtdCho was still not generated through PeMT. Results suggest that endogenous PtdCho synthesis is important during neurulation and that perturbed choline metabolism contributes to neural tube defects produced by DMAE and ET-18-OCH3.

Homocysteine-betaine interactions in a murine model of 5,10-methylenetetrahydrofolate reductase deficiency.

Hyperhomocysteinemia, a proposed risk factor for cardiovascular disease, is also observed in other common disorders. The most frequent genetic cause of hyperhomocysteinemia is a mutated methylenetetrahydrofolate reductase (MTHFR), predominantly when folate status is impaired. MTHFR synthesizes a major methyl donor for homocysteine remethylation to methionine. We administered the alternate choline-derived methyl donor, betaine, to wild-type mice and to littermates with mild or severe hyperhomocysteinemia due to hetero- or homozygosity for a disruption of the Mthfr gene. On control diets, plasma homocysteine and liver choline metabolite levels were strongly dependent on the Mthfr genotype. Betaine supplementation decreased homocysteine in all three genotypes, restored liver betaine and phosphocholine pools, and prevented severe steatosis in Mthfr-deficient mice. Increasing betaine intake did not further decrease homocysteine. In humans with cardiovascular disease, we found a significant negative correlation between plasma betaine and homocysteine concentrations. Our results emphasize the strong interrelationship between homocysteine, folate, and choline metabolism. Hyperhomocysteinemic Mthfr-compromised mice appear to be much more sensitive to changes of choline/betaine intake than do wild-type animals. Hyperhomocysteinemia, in the range of that associated with folate deficiency or with homozygosity for the 677T MTHFR variant, may be associated with disturbed choline metabolism.

Maternal dietary choline availability alters the balance of netrin-1 and DCC neuronal migration proteins in fetal mouse brain hippocampus.

Alterations in maternal dietary choline availability during days 12-17 of pregnancy led to an increase in the level of immunoreactive netrin-1 and a decrease in the level of DCC protein in the developing fetal mouse brain hippocampus compared with controls. Changes in the expression of cell migration cues during development could account for some of the lifelong consequences of maternal dietary choline availability for cognitive and memory processes.

Effects of betaine in a murine model of mild cystathionine-beta-synthase deficiency.

Cystathionine-beta-synthase (CBS) is required for transsulfuration of homocysteine, an amino acid implicated in vascular disease. We studied homocysteine metabolism in mice with mild hyperhomocysteinemia due to a heterozygous disruption of the Cbs gene. Mice were fed diets supplemented with betaine or dimethylsulfonioacetate (DMSA); betaine and DMSA provide methyl groups for an alternate pathway of homocysteine metabolism, remethylation by betaine:homocysteine methyltransferase (BHMT). On control diets, heterozygous mice had 50% higher plasma homocysteine than did wild-type mice. Betaine and DMSA had similar effects in both genotype groups: liver betaine increased dramatically, while plasma homocysteine decreased by 40% to 50%. With increasing betaine supplementation, homocysteine decreased by 75%. Plasma homocysteine and BHMT activity both showed a strong negative correlation with liver betaine. Homocysteinemia in mice is sensitive to a disruption of Cbs and to methyl donor intake. Because betaine leads to a greater flux through BHMT and lowers homocysteine, betaine supplementation may be beneficial in mild hyperhomocysteinemia.

Diethanolamine induces hepatic choline deficiency in mice.

The purpose of the present experiments was to test the hypothesis that diethanolamine (DEA), an alkanolamine shown to be hepatocarcinogenic in mice, induces hepatic choline deficiency and to determine whether altered choline homeostasis was causally related to the carcinogenic outcome. To examine this hypothesis, the biochemical and histopathological changes in male B6C3F1 mice made choline deficient by dietary deprivation were first determined. Phosphocholine (PCho), the intracellular storage form of choline was severely depleted, decreasing to about 20% of control values with 2 weeks of dietary choline deficiency. Other metabolites, including choline, glycerophosphocholine (GPC), and phosphatidylcholine (PC) also decreased. Hepatic concentrations of S-adenosylmethionine (SAM) decreased, whereas levels of S-adenosylhomocysteine (SAH) increased. Despite these biochemical changes, fatty liver, which is often associated with choline deficiency, was not observed in the mice. The dose response, reversibility, and strain-dependence of the effects of DEA on choline metabolites were studied. B6C3F1 mice were dosed dermally with DEA (0, 10, 20, 40, 80, and 160 mg/kg) for 4 weeks (5 days/week). Control animals received either no treatment or dermal application of 95% ethanol (1.8 ml/kg). PCho was most sensitive to DEA treatment, decreasing at dosages of 20 mg/kg and higher and reaching a maximum 50% depletion at 160 mg/kg/day. GPC, choline, and PC also decreased in a dose-dependent manner. At 80 and 160 mg/kg/day, SAM levels decreased while SAH levels increased in liver. A no-observed effect level (NOEL) for DEA-induced changes in choline homeostasis was 10 mg/kg/day. Choline metabolites, SAM and SAH returned to control levels in mice dosed at 160 mg/kg for 4 weeks and allowed a 2-week recovery period prior to necropsy. In a manner similar to dietary choline deficiency, no fatty change was observed in the liver of DEA-treated mice. In C57BL/6 mice, DEA treatment (160 mg/kg) also decreased PCho concentrations, without affecting hepatic SAM levels, suggesting that strain-specific differences in intracellular methyl group regulation may influence carcinogenic outcome with DEA treatment. Finally, in addition to the direct effects of DEA on choline homeostasis, dermal application of 95% ethanol for 4 weeks decreased hepatic betaine levels, suggesting that the use of ethanol as a vehicle for dermal application of DEA may exacerbate or confound the biochemical actions of DEA alone. Collectively, the results demonstrate that DEA treatment causes a spectrum of biochemical changes consistent with choline deficiency in mice and demonstrate a clear dose concordance between DEA-induced choline deficiency and hepatocarcinogenic outcome.

Deletion of the Pemt gene increases progenitor cell mitosis, DNA and protein methylation and decreases calretinin expression in embryonic day 17 mouse hippocampus.

Choline is a required nutrient and is derived from the diet as well as from de novo synthesis catalyzed by phosphatidylethanolamine N-methyltransferase (PEMT). We previously reported that choline availability during pregnancy alters mitosis and neuronal protein expression during fetal brain development in wild-type mice and rats, and that Pemt-/- mice become choline deficient. In this study, we examined brain development in these knockout mice. Pregnant Pemt-/- and wild-type mice were fed AIN-76A diet until gestation day 17 (E17) when the fetal brains were harvested. Phosphorylation of histone H3 (a measure of mitosis) and calretinin (a GABAergic neuronal marker) were assessed in hippocampal regions. We observed increased numbers of phosphorylated histone H3 positive cells in the Pemt-/- mice (up 54% compared to wild-type mice; p<0.01). We also found decreased calretinin labeling in Pemt-/- (down to 43% compared to wild-type mice; p<0.01). Thus, there was increased stem cell proliferation in the neuroepithelium and decreased GABAergic neuronal differentiation of these animals on E17. These results are opposite to what would have been expected in choline-deficient mice. The concentrations of S-adenosylmethionine (up 21%; p<0.05) and methylation of DNA (up 46%) and proteins (up 12%; p<0.01) in hippocampus were significantly increased in Pemt-/- mice, suggesting that increased S-adenosylmethionine availability may mediate the observed developmental changes. This is the first report of altered brain development in Pemt-/- mice.

Choline- and betaine-defined diets for use in clinical research and for the management of trimethylaminuria.

This article describes the development of a series of choline- and betaine-controlled diets that were served to research subjects as part of an ongoing study of diet requirements in humans. These diets were developed based on the analysis of choline and betaine in individual foods. The calculated diets were compared with analyses of all foods combined into a single sample for each day. The laboratory analyses of choline and betaine in the whole-diet aliquots matched the estimated amounts in the diets that were calculated from the analyses of individual foods. These diets were adjusted for several levels of choline and betaine and were well accepted by research subjects who consumed them for a time period of up to 2 months. This article describes applications of this diet for use in clinical research on methyl-group requirements in humans and for use in clinical practice for counseling the client who requires a choline-controlled diet.

Regulation of choline deficiency apoptosis by epidermal growth factor in CWSV-1 rat hepatocytes.

Previous studies show that acute choline deficiency (CD) triggers apoptosis in cultured rat hepatocytes (CWSV-1 cells). We demonstrate that prolonged EGF stimulation (10 ng/mL x 48 hrs) restores cell proliferation, as assessed by BrdU labeling, and protects cells from CD-induced apoptosis, as assessed by TUNEL labeling and cleavage of poly(ADP-ribose) polymerase. However, EGF rescue was not accompanied by restoration of depleted intracellular concentrations of choline, glycerphosphocholine, phosphocholine, or phosphatidylcholine. In contrast, we show that EGF stimulation blocks apoptosis by restoring mitochondrial membrane potential (Delta Psi(m)), as determined using the potential-sensitive dye chloromethyl-X-rosamine, and by preventing the release and nuclear localization of cytochrome c. We investigated whether EGF rescue involves EGF receptor phosphorylation and activation of the down-stream cell survival factor Akt. Compared to cells in control medium (CT, 70 micromol choline x 48 hrs), cells in CD medium (5 micromol choline) were less sensitive to EGF-induced (0-300 ng/mL x 5 min) receptor tyrosine phosphorylation. Compared to cells in CT medium, cells in CD medium treated with EGF (10 ng/mL x 5 min) exhibited higher levels of phosphatidylinositol 3-kinase (PI3K)-dependent phosphorylation of AktSer473. Inactivation of PI3K was sufficient to block EGF-stimulated activation of Akt, restoration of mitochondrial Delta Psi(m), and prevention of cytochrome c release. These studies indicate that stimulation with EGF activates a cell survival response against CD-apoptosis by restoring mitochondrial membrane potential and preventing cytochrome c release and nuclear translocation which are mediated by activation of Akt in hepatocytes.

Ad libitum choline intake in healthy individuals meets or exceeds the proposed adequate intake level.

Choline is an essential nutrient for humans that is used to synthesize membrane phospholipids and the neurotransmitter acetylcholine. Betaine, a metabolite of choline, functions as a methyl-group donor in the conversion of homocysteine to methionine, and is important for renal function. Accurate analysis of choline intake was previously not possible because the choline content of most foods was not known. Using new and recently published data on the concentrations of choline in common foods, we measured the choline content of diets consumed ad libitum by healthy adult volunteers housed in a clinical research center and compared these with estimates of choline intake derived from 3-d food records kept by subjects immediately before study enrollment. Mean choline intake in this subject population met or slightly exceeded the current Adequate Intake (AI) of 7 mg/(kg . d) set by the Institute of Medicine. Men and women consumed similar amounts of choline per day (8.4 and. 6.7 mg/kg, respectively; P = 0.11). Choline intakes estimated from the 3-d food records were significantly lower than this (when expressed as mg/kg, or as total mg, but not when normalized to energy intake), suggesting underreporting of food intake. Intake of betaine, which may spare choline utilization as a methyl-group donor, was 5.3 mg/(kg . d) in men and 4.7 mg/(kg . d) in women. Intake of folate, vitamin B-12, and methionine + cysteine, were similar and sufficient in all subjects. The current recommended AI for choline seems to be a good approximation of the actual intake of this nutrient.

Choline availability during embryonic development alters progenitor cell mitosis in developing mouse hippocampus.

Previously, we reported that dietary choline influences development of the hippocampus in fetal rat brain. It is important to know whether similar effects of choline occur in developing fetal mouse brain because interesting new experimental approaches are now available using several transgenic mouse models. Timed-pregnant mice were fed choline-supplemented (CS), control (CT) or choline-deficient (CD) AIN-76 diet from embryonic day 12 to 17 (E12-17). Fetuses from CD dams had diminished concentrations of phosphocholine and phosphatidylcholine in their brains compared with CT or CS fetuses (P < 0.05). When we analyzed fetal hippocampus on day E17 for cells with mitotic phase-specific expression of phosphorylated histone H3, we detected fewer labeled cells at the ventricular surface of the ventricular zone in the CD group (14.8 +/- 1.9) compared with the CT (30.7 +/- 1.9) or CS (36.6 +/- 2.6) group (P < 0.05). At the same time, we detected more apoptotic cells in E17 hippocampus using morphology in the CD group (11.8 +/- 1.4) than in CT (5.6 +/- 0.6) or CS (4.2 +/- 0.7) group (P < 0.05). This was confirmed using terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-digoxigenin anti-digoxigenin fluorescein conjugate antibody nick end-labeling (TUNEL) and activated caspase-3 immunoreactivity. We conclude that the dietary availability of choline to the mouse dam influences progenitor cell proliferation and apoptosis in the fetal brain.

A method for the determination of betaine in tissues using high performance liquid chromatography.

Betaine is a major metabolite of choline in liver and kidney and may be an important product of choline metabolism in other tissues. The available methods for assay of betaine, however, are time consuming and relatively insensitive. We describe a modification of published methods that provides a sensitive and specific assay for betaine by derivatization and HPLC separation with UV quantitation. Betaine and other water-soluble choline metabolites are extracted from biological samples and separated by HPLC based on mobility of 14C-labeled internal standards. The betaine fraction is collected and derivatized with 4'-bromo-phenacyl triflate. The betaine-triflate derivative is quantitated by UV absorbance and the result is corrected for possible losses due to incomplete extraction recovery and incomplete derivatization by simultaneous measurement of radioactivity from the derivatized 14C-betaine internal standard. Betaine concentrations determined with this procedure are reported for several adult and fetal rat tissues.

Concentrations of choline-containing compounds and betaine in common foods.

Choline is important for normal membrane function, acetylcholine synthesis and methyl group metabolism; the choline requirement for humans is 550 mg/d for men (Adequate Intake). Betaine, a choline derivative, is important because of its role in the donation of methyl groups to homocysteine to form methionine. In tissues and foods, there are multiple choline compounds that contribute to total choline concentration (choline, glycerophosphocholine, phosphocholine, phosphatidylcholine and sphingomyelin). In this study, we collected representative food samples and analyzed the choline concentration of 145 common foods using liquid chromatography-mass spectrometry. Foods with the highest total choline concentration (mg/100 g) were: beef liver (418), chicken liver (290), eggs (251), wheat germ (152), bacon (125), dried soybeans (116) and pork (103). The foods with the highest betaine concentration (mg/100 g) were: wheat bran (1339), wheat germ (1241), spinach (645), pretzels (237), shrimp (218) and wheat bread (201). A number of epidemiologic studies have examined the relationship between dietary folic acid and cancer or heart disease. It may be helpful to also consider choline intake as a confounding factor because folate and choline methyl donation can be interchangeable.

Phosphatidylethanolamine N-methyltransferase (PEMT) knockout mice have hepatic steatosis and abnormal hepatic choline metabolite concentrations despite ingesting a recommended dietary intake of choline.

Choline is an essential nutrient for humans and is derived from the diet as well as from de novo synthesis involving methylation of phosphatidylethanolamine catalysed by the enzyme phosphatidylethanolamine N -methyltransferase (PEMT). This is the only known pathway that produces new choline molecules. We used mice with a disrupted Pemt-2 gene (which encodes PEMT; Pemt (-/-)) that have previously been shown to possess no hepatic PEMT enzyme. Male, female and pregnant Pemt (-/-) and wild-type mice ( n =5-6 per diet group) were fed diets of different choline content (deficient, control, and supplemented). Livers were collected and analysed for choline metabolites, steatosis, and apoptotic [terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labelling (TUNEL)] positive cells. We found that, in livers of Pemt (-/-) mice fed any of the diets, there was hepatic steatosis and significantly higher occurrence of TUNEL positive cells compared with wild-type controls. In male, female and pregnant mice, liver phosphatidylcholine concentrations were significantly decreased in Pemt (-/-) choline deficient and in Pemt (-/-) choline control groups but returned to normal in Pemt (-/-) choline supplemented groups. Phosphocholine concentrations in liver were significantly diminished in knockout mice even when choline was supplemented to above dietary requirements. These results show that PEMT normally supplies a significant portion of the daily choline requirement in the mouse and, when this pathway is knocked out, mice are unable to attain normal concentrations of all choline metabolites even with a supplemental source of dietary choline.

Quantitation of choline and its metabolites in tissues and foods by liquid chromatography/electrospray ionization-isotope dilution mass spectrometry.

Choline is important for normal membrane function, acetylcholine synthesis, lipid transport, and methyl metabolism. The U.S. National Academy of Sciences recently set requirements for choline in the human diet. In tissues and foods, there are multiple choline compounds that contribute to choline content. Betaine, a derivative of choline, is also important because of its role in donation of methyl groups to homocysteine to form methionine. Radioisotopic, high-pressure liquid chromatography, and gas chromatography/isotope dilution mass spectrometry (GC/IDMS) methods are available for measurement of choline. However, these existing methods are cumbersome and time-consuming, and none measures all of the compounds of interest. In this study, we describe a new method for quantitation of choline, betaine, acetylcholine, glycerophosphocholine, cytidine diphosphocholine, phosphocholine, phosphatidylcholine, and sphingomyelin in liver, plasma, various foods, and brain using liquid chromatography/electrospray ionization-isotope dilution mass spectrometry (LC/ESI-IDMS). Choline compounds were extracted by and partitioned into organic and aqueous phases using methanol and chloroform and analyzed directly by LC/ESI-IDMS without the need for isolation and derivatization of each compound separately as was required by the GC/IDMS method. The new LC/ESI-IDMS method was validated using the existing published GC/IDMS method.

Folic acid deficiency during late gestation decreases progenitor cell proliferation and increases apoptosis in fetal mouse brain.

In mice and rats, maternal dietary choline intake during late pregnancy modulates mitosis and apoptosis in progenitor cells of the fetal hippocampus and septum. Because choline and folate are interrelated metabolically, we investigated the effects of maternal dietary folate availability on progenitor cells in fetal mouse telencephalon. Timed-pregnant mice were fed a folate-supplemented (FS), control (FCT) or folate-deficient (FD) AIN-76 diet from d 11-17 of pregnancy. FD decreased the number of progenitor cells undergoing cell replication in the ventricular zones of the developing mouse brain septum (46.6% of FCT), caudate putamen (43.5%), and neocortex (54.4%) as assessed using phosphorylated histone H3 (a specific marker of mitotic phase) and confirmed by bromodeoxyuridine (BrdU) labeling of the S phase. In addition, 106.2% more apoptotic cells were found in FD than in FCT fetal septum. We observed 46.8% more calretinin-positive cells in the medial septal-diagonal band region of FD compared with pups from control dams. FS mice did not differ significantly from FCT mice in any of these measures. These results suggest that progenitor cells in fetal forebrain are sensitive to maternal dietary folate during late gestation.

Deficiency in methionine, tryptophan, isoleucine, or choline induces apoptosis in cultured cells.

Cells in culture die by apoptosis when deprived of the essential nutrient choline. We now report that cells (both proliferating PC12 cells and postmitotic neurons isolated from fetal rat brains) undergo apoptosis when deprived of other individual essential nutrients (methionine, tryptophan or isoleucine). In PC12 cells, deficiencies of each nutrient independently led to ceramide accumulation and to caspase activation, both recognized signals of several apoptotic pathways. A similar profile of caspases was activated in PC12 cells deprived of choline, methionine, tryptophan or isoleucine. More than one caspase was involved and these caspases appeared to transmit parallel signals for apoptosis induction because only broad-spectrum caspase inhibitors, but not inhibitors for specific individual caspases inhibited apoptosis in choline- or methionine-deprived cells. The induction of these caspase-dependent apoptosis pathways likely did not involve the same upstream signals. Choline deficiency perturbed choline metabolism but did not affect protein synthesis, whereas amino acid deficiencies inhibited protein synthesis but did not perturb choline metabolism. In addition, a subclone of PC12 cells that was resistant to choline deficiency-induced apoptosis was not resistant to tryptophan deficiency-induced apoptosis. These observations suggest that deficiency of each studied nutrient activates different pathways for signaling apoptosis that ultimately converge on a common execution pathway.

Mitochondrial and microsomal derived reactive oxygen species mediate apoptosis induced by transforming growth factor-beta1 in immortalized rat hepatocytes.

Transforming growth factor-beta1 (TGFbeta1) is a multifunctional cytokine that is over expressed during liver hepatocytes injury and regeneration. SV40-transformed CWSV-1 rat hepatocytes that are p53-defective undergo apoptosis in response to choline deficiency (CD) or TGFbeta1, which mediates CD-apoptosis. Reactive oxygen species (ROS) are essential mediators of apoptosis. We have shown that apoptosis induced by TGFbeta1 is accompanied by ROS generation and the ROS-trapping agent N-acetylcysteine (NAC) inhibits TGFbeta1-induced apoptosis. While persistent induction of ROS contributes to this form of apoptosis, the source of ROS generated downstream of TGFbeta1 is not clear. The mitochondria and the endoplasmic reticulum both harbor potent electron transfer chains that might be the source of ROS essential for completion of TGFbeta1-apoptosis. Here we show that CWSV-1 cells treated with cyclosporine A, which prevents opening of mitochondrial membrane pores required for ROS generation, inhibits TGFbeta1-induced apoptosis. A similar effect was obtained by treating these cells with rotenone, an inhibitor of complex 1 of the mitochondrial electron transfer chain. However, we demonstrate that TGFbeta1 induces cytochrome P450 1A1 and that metyrapone, a potent inhibitor of cytochrome P450 1A1, inhibits TGFbeta1-induced apoptosis. Therefore, our studies indicate that concurrent with promoting generation of ROS from mitochondria, TGFbeta1 also promotes generation of ROS from the cytochrome P450 electron transfer chain. Since inhibition of either of these two sources of ROS interferes with apoptosis, it is reasonable to conclude that the combined involvement of both pathways is essential for completion of TGFbeta1-induced apoptosis.

Choline availability during embryonic development alters the localization of calretinin in developing and aging mouse hippocampus.

Choline availability in the diet during pregnancy alters fetal brain biochemistry with resulting behavioral changes that persist throughout the lifetime of the offspring. In the present study, the effects of dietary choline on the onset of GABAergic neuronal differentiation in developing fetal brain, as demarcated by the expression of calcium binding protein calretinin, are described. In these studies, timed-pregnant mice were fed choline supplemented, control or choline deficient AIN-76 diet from day 12-17 of pregnancy and the brains of their fetuses were studied on day 17 of gestation. In the primordial dentate gyrus, we found that pups from choline deficient-dams had more calretinin protein (330% increase), and pups from choline supplemented-dams had less calretinin protein (70% decrease), than did pups from control-dams. Importantly, decreased calretinin protein was still detectable in hippocampus in aged, 24-month-old mice, born of choline supplemented-dams and maintained since birth on a control diet. Thus, alterations in the level of calretinin protein in fetal brain hippocampus could underlie the known, life long effects of maternal dietary choline availability on brain development and behavior.