Trimethylaminuria (fish-odour syndrome): an inborn error of oxidative metabolism

The Accumulation and Molecular Effects of Trimethylamine N-Oxide on Metabolic Tissues: It's Not All Bad

Since elevated serum levels of trimethylamine N-oxide (TMAO) were first associated with increased risk of cardiovascular disease (CVD), TMAO research among chronic diseases has grown exponentially. We now know that serum TMAO accumulation begins with dietary choline metabolism across the microbiome-liver-kidney axis, which is typically dysregulated during pathogenesis. While CVD research links TMAO to atherosclerotic mechanisms in vascular tissue, its molecular effects on metabolic tissues are unclear. Here we report the current standing of TMAO research in metabolic disease contexts across relevant tissues including the liver, kidney, brain, adipose, and muscle. Since poor blood glucose management is a hallmark of metabolic diseases, we also explore the variable TMAO effects on insulin resistance and insulin production. Among metabolic tissues, hepatic TMAO research is the most common, whereas its effects on other tissues including the insulin producing pancreatic β-cells are largely unexplored. Studies on diseases including obesity, diabetes, liver diseases, chronic kidney disease, and cognitive diseases reveal that TMAO effects are unique under pathologic conditions compared to healthy controls. We conclude that molecular TMAO effects are highly context-dependent and call for further research to clarify the deleterious and beneficial molecular effects observed in metabolic disease research.

Metabolic phenotyping and systems biology approaches to understanding metabolic syndrome and fatty liver disease

Metabolic syndrome, a cluster of risk factors for type 2 diabetes mellitus and cardiovascular disease, is becoming an increasing global health concern. Insulin resistance is often associated with metabolic syndrome and also typical hepatic manifestations such as nonalcoholic fatty liver disease. Profiling of metabolic products (metabolic phenotyping or metabotyping) has provided new insights into metabolic syndrome and nonalcoholic fatty liver disease. Data from nuclear magnetic resonance spectroscopy and mass spectrometry combined with statistical modeling and top-down systems biology have allowed us to analyze and interpret metabolic signatures in terms of metabolic pathways and protein interaction networks and to identify the genomic and metagenomic determinants of metabolism. For example, metabolic phenotyping has shown that relationships between host cells and the microbiome affect development of the metabolic syndrome and fatty liver disease. We review recent developments in metabolic phenotyping and systems biology technologies and how these methodologies have provided insights into the mechanisms of metabolic syndrome and nonalcoholic fatty liver disease. We discuss emerging areas of research in this field and outline our vision for how metabolic phenotyping could be used to study metabolic syndrome and fatty liver disease.

Joint functions of protein residues and NADP(H) in oxygen activation by flavin-containing monooxygenase

The reactivity of flavoenzymes with dioxygen is at the heart of a number of biochemical reactions with far reaching implications for cell physiology and pathology. Flavin-containing monooxygenases are an attractive model system to study flavin-mediated oxygenation. In these enzymes, the NADP(H) cofactor is essential for stabilizing the flavin intermediate, which activates dioxygen and makes it ready to react with the substrate undergoing oxygenation. Our studies combine site-directed mutagenesis with the usage of NADP(+) analogues to dissect the specific roles of the cofactors and surrounding protein matrix. The highlight of this "double-engineering" approach is that subtle alterations in the hydrogen bonding and stereochemical environment can drastically alter the efficiency and outcome of the reaction with oxygen. This is illustrated by the seemingly marginal replacement of an Asn to Ser in the oxygen-reacting site, which inactivates the enzyme by effectively converting it into an oxidase. These data rationalize the effect of mutations that cause enzyme deficiency in patients affected by the fish odor syndrome. A crucial role of NADP(+) in the oxygenation reaction is to shield the reacting flavin N5 atom by H-bond interactions. A Tyr residue functions as backdoor that stabilizes this crucial binding conformation of the nicotinamide cofactor. A general concept emerging from this analysis is that the two alternative pathways of flavoprotein-oxygen reactivity (oxidation versus monooxygenation) are predicted to have very similar activation barriers. The necessity of fine tuning the hydrogen-bonding, electrostatics, and accessibility of the flavin will represent a challenge for the design and development of oxidases and monoxygenases for biotechnological applications.

Effects of the dietary supplements, activated charcoal and copper chlorophyllin, on urinary excretion of trimethylamine in Japanese trimethylaminuria patients

Trimethylaminuria (TMAU) is a metabolic disorder characterized by the inability to oxidize and convert dietary-derived trimethylamine (TMA) to trimethylamine N-oxide (TMAO). This disorder has been relatively well-documented in European and North American populations, but no reports have appeared regarding patients in Japan. We identified seven Japanese individuals that showed a low metabolic capacity to convert TMA to its odorless metabolite, TMAO. The metabolic capacity, as defined by the concentration of TMAO excreted in the urine divided by TMA concentration plus TMAO concentration, in these seven individuals ranged from 70 to 90%. In contrast, there were no healthy controls examined with less than 95% of the metabolic capacity to convert TMA to TMAO. The intake of dietary charcoal (total 1.5 g charcoal per day for 10 days) reduced the urinary free TMA concentration and increased the concentration of TMAO to normal values during charcoal administration. Copper chlorophyllin (total 180 mg per day for 3 weeks) was also effective at reducing free urinary TMA concentration and increasing TMAO to those of concentrations present in normal individuals. In the TMAU subjects examined, the effects of copper chlorophyllin appeared to last longer (i.e., several weeks) than those observed for activated charcoal. The results suggest that the daily intake of charcoal and/or copper chlorophyllin may be of significant use in improving the quality of life of individuals suffering from TMAU.

Quantification and cellular localization of expression in human skin of genes encoding flavin-containing monooxygenases and cytochromes P450

The expression, in adult human skin, of genes encoding flavin-containing monooxygenases (FMOs) 1, 3, 4, and 5 and cytochromes P450 (CYPs) 2A6, 2B6, and 3A4 was determined by RNase protection. Each FMO and CYP exhibits inter-individual variation in expression in this organ. Of the individuals analysed, all contained CYP2B6 mRNA in their skin, 90% contained FMO5 mRNA and about half contained mRNAs encoding FMOs 1, 3, and 4, and CYPs 2A6 and 3A4. The amount of each of the FMO and CYP mRNAs in skin is much lower than in the organ in which it is most highly expressed, namely the kidney (for FMO1) and the liver (for the others). In contrast to the latter organs, in the skin FMO mRNAs are present in amounts similar to, or greater than, CYP mRNAs. Only the mRNA encoding CYP2B6 decreased in abundance in skin with increasing age of the individual. All of the mRNAs were substantially less abundant in cultures of keratinocytes than in samples of skin from which the cells were derived. In contrast, an immortalized human keratinocyte cell line, HaCaT, expressed FMO3, FMO5, and CYP2B6 mRNAs in amounts that fall within the range detected in the whole skin samples analysed. FMO1, CYP2A6, and CYP3A4 mRNAs were not detected in HaCaT cells, whereas FMO4 expression was markedly increased in this cell line compared to whole skin. In situ hybridization showed that the expression of each of the FMOs and CYPs analysed was localized to the epidermis, sebaceous glands and hair follicles.

A novel mutation in the flavin-containing monooxygenase 3 gene, FM03, that causes fish-odour syndrome: activity of the mutant enzyme assessed by proton NMR spectroscopy

We have previously shown that primary trimethylaminuria, or fish-odour syndrome, is caused by an inherited defect in the flavin-containing monooxygenase 3 (FMO3) catalysed N-oxidation of the dietary-derived malodorous amine, trimethylamine (TMA). We now report a novel causative mutation for the disorder identified in a young girl diagnosed by proton nuclear magnetic resonance (NMR) spectroscopy of her urine. Sequence analysis of genomic DNA amplified from the patient revealed that she was homozygous for a T to C missense mutation in exon 3 of the FMO3 gene. The mutation changes an ATG triplet, encoding methionine, at codon 82 to an ACG triplet, encoding threonine. A polymerase chain reaction/restriction enzyme-based assay was devised to genotype individuals for the FMO3Thr82 allele. Wild-type and mutant FMO3, heterologously expressed in a baculovirus-insect cell system, were assayed by ultraviolet spectrophotometry and NMR spectroscopy for their ability to catalyse the N-oxidation of TMA. The latter technique has the advantage of enabling the simultaneous, direct and semi-continuous measurement of both of the products, TMA N-oxide and NADP, and of one of the reactants, NADPH. Results obtained from both techniques demonstrate that the Met82Thr mutation abolishes the catalytic activity of the enzyme and thus represents the genetic basis of the disorder in this individual. The combination of NMR spectroscopy with gene sequence and expression technology provides a powerful means of determining genotype-phenotype relationships in trimethylaminuria.

The effect of arginine-428 mutation on modulation of activity of human liver flavin monooxygenase 3 (FMO3) by imipramine and chlorpromazine

This study was carried out to investigate the molecular basis for modulation of recombinant FMO3-catalyzed activity by the tricyclic antidepressants, imipramine and chlorpromazine. A mutant of human liver FMO3 (T428R) was formed by site-directed mutagenesis and characterized along with the native enzyme in order to elucidate a possible structure-function relationship. Functional properties of native and T428R human FMO3s were studied with methimazole as substrate. Both enzymes catalyzed the S-oxidation of methimazole with the same Km value. Imipramine modulated the activities of the native and T428R human FMO3s differently; the activity of the native FMO3 was increased at all concentrations, whereas the activity of the mutant enzyme was inhibited at concentrations above 300 microM. Chlorpromazine activated the native enzyme at all concentrations of methimazole but activated the mutant enzyme only at high substrate concentrations. The direction (activation or inhibition) and extend of modulation of FMO3 activity is not only dependent on the concentration of the modulator, it is also dependent on the substrate concentration. This study confirms our previous findings with FMO1 that position 428 is important in the interaction of the FMO with modulators.

Sequence variations in the flavin-containing mono-oxygenase 3 gene (FMO3) in fish odour syndrome

Trimethylaminuria is inherited recessively as a defect in hepatic N-oxidation of dietary derived trimethylamine (TMA) which causes excess excretion of TMA so that affected individuals have a body odour resembling rotten fish. Flavin-containing mono-oxygenase 3 (FMO3) catalyses TMA oxidation and mutations in the FMO3 gene have recently been shown to underlie trimethylaminuria/fish odour syndrome. We searched for FMO3 mutations in a previously unreported individual with this disorder using polymerase chain reaction of genomic DNA, heteroduplex analysis and direct sequencing of heteroduplex band shifts. We identified a heterozygous missense Pro153-->Leu153 mutation in exon 4. Leu153 has been reported previously as a homozygous mutation in two unrelated siblings with trimethylaminuria and has been shown to result in total loss of FMO3 enzyme activity. In our patient, two further missense mutations were identified on the other FMO3 allele, Val143-->Glu143 and Glu158-->Lys158. Lys158 is known to be a common polymorphism, but has functional significance in reducing enzyme activity by 10%. Glu143 has not been documented previously, but was shown to be a rare polymorphism and may be of further relevance in reducing FMO3 activity. Mutagenesis studies and enzyme assays will be necessary to confirm or refute the potential pathogenic significance of Glu143 in this patient, but the mutation Pro153-->Leu153 appears to be a recurrent cause of this distressing metabolic disorder.

Missense mutation in flavin-containing mono-oxygenase 3 gene, FMO3, underlies fish-odour syndrome

Individuals with primary trimethylaminuria exhibit a body odour reminiscent of rotting fish, due to excessive excretion of trimethylamine (TMA; refs 1-3). The disorder, colloquially known as fish-odour syndrome, is inherited recessively as a defect in hepatic N-oxidation of dietary-derived TMA and cannot be considered benign, as sufferers may display a variety of psychosocial reactions, ranging from social isolation of clinical depression and attempted suicide. TMA oxidation is catalyzed by flavin-containing mono-oxygenase (FMO; refs 7,8), and tissue localization and functional studies have established FMO3 as the form most likely to be defective in fish-odour syndrome. Direct sequencing of the coding exons of FMO3 amplified from a patient with fish-odour syndrome identified two missense mutations. Although one of these represented a common polymorphism, the other, a C-->T transition in exon 4, was found only in an affected pedigree, in which it segregated with the disorder. The latter mutation predicts a proline-->leucine substitution at residue 153 and abolishes FMO3 catalytic activity. Our results indicate that defects in FMO3 underlie fish-odour syndrome and that the Pro 153-->Leu 153 mutation described here is a cause of this distressing condition.

Structural organization of the human flavin-containing monooxygenase 3 gene (FMO3), the favored candidate for fish-odor syndrome, determined directly from genomic DNA

The inherited metabolic disorder trimethylaminuria (fish-odor syndrome) is associated with defective hepatic N-oxidation of dietary-derived trimethylamine catalyzed by flavin-containing monooxygenase (FMO). As FMO3 encodes the major form of FMO expressed in adult human liver, it represents the best candidate gene for the disorder. The structural organization of FMO3 was determined by sequencing the products of exon-to-exon and vectorette PCR, the latter through the use of vectorette libraries constructed directly from genomic DNA. The gene contains one noncoding and eight coding exons. Knowledge of the exon/intron organization of the human FMO3 gene enabled each of the coding exons of the gene, together with their associated flanking intron sequences, to be amplified from genomic DNA and will thus facilitate the identification of mutations in FMO3 in families affected with fish-odor syndrome.

Differential developmental and tissue-specific regulation of expression of the genes encoding three members of the flavin-containing monooxygenase family of man, FMO1, FMO3 and FM04

We have previously described the isolation and sequencing of cDNA clones encoding flavin-containing monooxygenases (FMOs) 1 and 4 of man [Dolphin, C., Shephard, E. A., Povey, S., Palmer, C. N. A., Ziegler, D. M., Ayesh, R., Smith, R. L. & Phillips, I. R. (1991) J. Biol. Chem. 266, 12379-12385; Dolphin, C., Shephard E. A., Povey, S., Smith, R. L. & Phillips, I. R. (1992) Biochem. J. 287, 261-267]. We present here the isolation of a cDNA for FM03 of man. The sequence of this CDNA and the amino acid sequence deduced from it differ substantially from those previously reported for this member of the FMO family of man. In addition, we have investigated, by quantitative RNase protection assays, the expression in several foetal and adult human tissues of genes encoding FMO1, FMO3 and FMO4, Our results demonstrate that, in the adult, FMO1 is expressed in kidney but not in liver, whereas in the foetus it is expressed in both organs. The lack of expression of FMO1 in adult human liver is in marked contrast to the situation in other mammals, such as pig and rabbit, in which FMO1 constitutes a major form of the enzyme in the liver of the adult animal. The mRNA encoding FMO3 is abundant in adult liver and is also present, in low abundance, in some foetal tissues. Thus, FMO1 and FMO3 are both subject to developmental and tissue-specific regulation, with a developmental switch in the expression of the genes taking place in the liver. FMO4 mRNA is present in low abundance in several foetal and adult tissues and thus the corresponding gene appears to be expressed constitutively.

The molecular biology of the flavin-containing monooxygenases of man

cDNA clones encoding five distinct members of the FMO family of man (FMOs 1, 2, 3, 4 and 5) were isolated by a combination of library screening and reverse transcription-polymerase chain reaction techniques. The deduced amino acid sequences of the human FMOs have 82-87% identity with their known orthologues in other mammal but only 51-57% similarity to each other. The hydropathy profiles of the proteins are very similar. From the calculated rate of evolution of FMOs (a 1% change in sequence per 6 million years) it would appear that individual members of the FMO gene family arose by duplication of a common ancestral gene some 250-300 million years ago. Each of the FMO genes was mapped by the polymerase chain reaction to the long arm of human chromosome 1. The localization of the FMO1 gene was further refined to 1q23-q25 by in situ hybridization of human metaphase chromosomes. RNase protection assays demonstrated that in man each FMO gene displays a distinct developmental and tissue-specific pattern of expression. In the adult, FMO1 is expressed in kidney but not in liver, whereas in the foetus its mRNA is abundant in both organs. FMO3 expression is essentially restricted to the liver in the adult and the mRNA is either absent, or present in low amounts, in foetal tissues. FMO4 is expressed more constitutively. Human FMO1 and FMO3 cDNAs were functionally expressed in prokaryotic and eukaryotic cells. FMO1 and FMO3, expressed in either system, displayed product stereoselectivity in their catalysis of the N-oxidation of the pro-chiral tertiary amines, N-ethyl-N-methylaniline (EMA) and pargyline. Both enzymes were stereoselective with respect to the production of the (-)-S-enantiomer of EMA N-oxide. But in the case of pargyline, the enzymes displayed opposite stereoselectivity, FMO1 producing solely the (+)-enantiomer and FMO3 predominantly the (-)-enantiomer of the N-oxide.

Trimethylaminuria in a girl with Prader-Willi syndrome and del(15)(q11q13)

We report on an individual with trimethylaminuria, Prader-Willi syndrome, and del(15) (q11q13). To our knowledge, such an association has never been reported. Skin sores secondary to choline-rich foods and amenable to dietary control have not been described in trimethylaminuria, although they are seen in some patients with Prader-Willi syndrome. Pathogenesis, clinical diagnosis, and management of reported cases with trimethylaminuria are reviewed. Serious social and behavioral problems may result from strong body odor. Amelioration of the "fish odor" by dietary choline restriction makes trimethylaminuria detection important. Association of trimethylaminuria with Prader-Willi syndrome and del(15) (q11q13) in this patient is of particular interest. It may represent a contiguous gene syndrome, or deletion of the normal allele leading to expression of a single recessive trimethylaminuria gene, or an unrelated association, such as in Noonan syndrome. However, recent development of mapping of flavin-containing monooxygenase 2 (FMO2), the likely enzyme that is defective in fish odor syndrome, to chromosome 1q probably excludes pathogenetic association of fish odor syndrome with the Prader-Willi syndrome.

Cloning, primary sequence and chromosomal localization of human FMO2, a new member of the flavin-containing mono-oxygenase family

We have previously reported the cloning of cDNAs for a flavin-containing mono-oxygenase (FMO) of man, designated FMO1 [Dolphin, Shephard, Povey, Palmer, Ziegler, Ayesh, Smith & Phillips (1991) J. Biol. Chem. 266, 12379-12385], that is the orthologue of pig and rabbit hepatic FMOs. We now describe the isolation and characterization of cDNA clones for a second human FMO, which we have designated FMO2. The polypeptide encoded by the cDNAs is 558 amino acid residues long, has a calculated M(r) of 63337, and contains putative FAD- and NADP-binding sites that align exactly with those described in other mammalian FMOs. Human FMO2 has 51-53% primary sequence identity with human FMO1, rabbit pulmonary FMO and rabbit liver FMO form 2, and thus represents a fourth, distinct, member of the mammalian FMO family. The corresponding mRNA is present in low abundance in adult human liver. Southern blot hybridization with single-exon probes demonstrated that human FMO2 and FMO1 are the products of single genes. The gene encoding FMO2 (designated FMO2) was mapped, by the polymerase chain reaction, to human chromosome 1, the same chromosome on which FMO1 is located.

Experimental study on the metabolism of dimethylethylamine in man

Dimethylethylamine (DMEA) is an aliphatic tertiary amine, which is used as a catalyst in the mould core manufacturing. During 8 h, four healthy volunteers were exposed to four different DMEA air concentrations (10, 20, 40 and 50 mg/m3; 20 mg/m3, two subjects only). DMEA was biotransformed into dimethylethylamine N-oxide (DMEAO). On average, DMEAO, accounted for 90% of the combined amount of DMEA and DMEAO excreted into the urine. The half-lives of DMEA and DMEAO in plasma were 1.3 and 3.0 h, respectively. The urinary excretion of DMEA and DMEAO followed a two-phase pattern. The half-lives in the first phase were 1.5 h for DMEA and 2.5 h for DMEAO. In the second phase, which started about 9 h after the end of exposure, half-lives of 7 h for DMEA and 8 h for DMEAO were recorded. The combined concentration of DMEA and DMEAO, in both plasma and urine, showed an excellent correlation with the air concentration of DMEA. Thus, both urinary excretion and plasma concentration can be used for biological monitoring of exposure to DMEA. An 8-h exposure to 10 mg DMEA/m3 corresponds to a postexposure plasma concentration and 2-h postexposure urinary excretion of 4.9 mumol/l and 75 mmol/mol creatinine, respectively.

Dimethylethylamine in mould core manufacturing: exposure, metabolism, and biological monitoring

The exposure and metabolism of dimethylethylamine (DMEA) was studied in 12 mould core makers in four different foundries using the Ashland cold box technique. The mean time weighted average (TWA) full work shift DMEA exposure concentration was 3.7 mg/m3. Inhaled DMEA was excreted into urine as the original amine and as its metabolite dimethylethylamine-N-oxide (DMEAO). This metabolite made up a median of 87 (range 18-93) % of the sum of DMEA and DMEAO concentrations excreted into the urine. Occupational exposure did not significantly increase the urinary excretion of dimethylamine or methylethylamine. The data indicate half lives after the end of exposure for DMEA in urine of 1.5 hours and DMEAO of three hours. The postshift summed concentration of DMEA and DMEAO in plasma and urine is a good indicator of the TWA concentration in air during the workday, and might thus be used for biological monitoring. An air concentration of 10 mg/m3 corresponds to a urinary excretion of the summed amount of DMEA and DMEAO of 135 mmol/mol creatinine.

Flavin-containing monooxygenases: enzymes adapted for multisubstrate specificity

Unlike all other oxidases, microsomal flavin-containing monooxygenases (FMO) discriminate between essential and foreign compounds by excluding the former rather than selectively binding the latter. As Daniel Ziegler describes here, xenobiotics that readily cross cell membranes can enter the catalytic cavity, whereas charged groups on essential metabolites that prevent their passive diffusion out of the cell also block their access to FMO. FMO appears to be ideally adapted to catalyse the detoxification of structurally diverse soft nucleophiles (e.g. alkaloids with basic side-chains and organic sulfur xenobiotics) so abundant in food derived from plants.

Pharmacokinetics of triethylamine and triethylamine-N-oxide in man

The pharmacokinetics of the industrially important compound triethylamine (TEA) and its metabolite triethylamine-N-oxide (TEAO) were studied in four volunteers after oral and intravenous administration. TEA was efficiently absorbed from the gastrointestinal (GI) tract, rapidly distributed, and in part metabolized into TEAO. There was no significant first pass metabolism. TEAO was also well absorbed from the GI tract. Within the GI tract, TEAO was reduced into TEA (19%) and dealkylated into diethylamine (DEA; 10%). The apparent volumes of distribution during the elimination phase were 192 liters for TEA and 103 liters for TEAO. Gastric intubation showed that there was a close association between levels of TEA in plasma and gastric juice, the latter levels being 30 times higher. The TEA and TEAO in plasma had half-lives of about 3 and 4 hr, respectively. Exhalation of TEA was minimal. More than 90% of the dose was recovered in the urine as TEA and TEAO. The urinary clearances of TEA and TEAO indicated that in addition to glomerular filtration, tubular secretion takes place. For TEAO at high levels, the secretion appears to be saturable. The present data, in combination with those of earlier studies, indicate that the sum of TEA and TEAO in urine may be used for biological monitoring of exposure to TEA.

Formation of aliphatic amine precursors of N-nitrosodimethylamine after oral administration of choline and choline analogues in the rat

Trimethylamine and dimethylamine are important precursors of N-nitrosodimethylamine, which is a potent carcinogen in a wide variety of animal species. Choline, a component of the normal human diet, is metabolized by bacteria within the intestine to form trimethylamine and dimethylamine. However, animals on a choline-free diet continue to excrete some trimethylamine and dimethylamine, suggesting that other dietary precursors of these methylamines might exist. To determine whether C-N bond cleavage by the intestinal bacteria is specific to the choline molecule, we measured monomethylamine, dimethylamine, trimethylamine and trimethylamine oxide excretion in rat urine after the administration of compounds that shared structural features with choline. Water, choline, dimethylaminoethanol, diethylaminoethanol, phosphocholine, betaine, carnitine, beta-methylcholine or dimethylaminoethyl chloride were administered by orogastric intubation, and the urine was collected for 24 hr. Administration of choline (15 mmol/kg body weight) resulted in increased urinary excretion of dimethylamine, trimethylamine and trimethylamine oxide (increases of approximately twofold, 500-fold and 50-fold, respectively). Of the administered choline, 12% was converted to trimethylamine or trimethylamine oxide and excreted in the urine within 24 hr. Phosphocholine administration resulted in similar increases in dimethylamine, trimethylamine and trimethylamine oxide excretion by rats. Modification of the ethyl-backbone or quaternary amine end of the choline molecule resulted in marked suppression of methylamine formation. Though administration of some analogues of choline (methylcholine, betaine and carnitine) resulted in the formation of small amounts of trimethylamine or trimethylamine oxide, and the administration of others (dimethylaminoethanol and dimethylaminoethyl chloride) resulted in the formation of some dimethylamine, the amounts formed were minimal compared with the amounts of trimethylamine and trimethylamine oxide formed after choline administration. Thus, of the many components of foods, only choline and its esters are likely to be significant substrates for trimethylamine and dimethylamine formation. How then can we explain the persistence of trimethylamine and dimethylamine excretion observed in choline-deficient rats? We suggest that endogenous (non-bacterial) synthesis of trimethylamine and dimethylamine occurs within some tissue of the rat.