Metabolism of [14C]methamphetamine in man, the guinea pig and the rat

1. The metabolites of (+/-)-2-methylamino-1-phenyl[1-(14)C]propane ([(14)C]methamphetamine) in urine were examined in man, rat and guinea pig. 2. In two male human subjects receiving the drug orally (20mg per person) about 90% of the (14)C was excreted in the urine in 4 days. The urine of the first day was examined for metabolites, and the main metabolites were the unchanged drug (22% of the dose) and 4-hydroxymethamphetamine (15%). Minor metabolites were hippuric acid, norephedrine, 4-hydroxyamphetamine, 4-hydroxynorephedrine and an acid-labile precursor of benzyl methyl ketone. 3. In the rat some 82% of the dose of (14)C (45mg/kg) was excreted in the urine and 2-3% in the faeces in 3-4 days. In 2 days the main metabolites in the urine were 4-hydroxymethamphetamine (31% of dose), 4-hydroxynorephedrine (16%) and unchanged drug (11%). Minor metabolites were amphetamine, 4-hydroxyamphetamine and benzoic acid. 4. The guinea pig was injected intraperitoneally with the drug at two doses, 10 and 45mg/kg. In both cases nearly 90% of the (14)C was excreted, mainly in the urine after the lower dose, but in the urine (69%) and faeces (18%) after the higher dose. The main metabolites in the guinea pig were benzoic acid and its conjugates. Minor metabolites were unchanged drug, amphetamine, norephedrine, an acid-labile precursor of benzyl methyl ketone and an unknown weakly acidic metabolite. The output of norephedrine was dose-dependent, being about 19% on the higher dose and about 1% on the lower dose. 5. Marked species differences in the metabolism of methamphetamine were observed. The main reaction in the rat was aromatic hydroxylation, in the guinea pig demethylation and deamination, whereas in man much of the drug, possibly one-half, was excreted unchanged.     

Studies on flavonoid metabolism. Biliary and urinary excretion of metabolites of (+)-[U-14C]catechin

1. The biliary and urinary excretion of (+)-[U-(14)C]catechin was studied in normal male rats after a single injection of the flavonoid. 2. In rats large amounts of radioactivity (33.6-44.3% of the dose in 24h) were excreted in the bile as two glucuronide conjugates [one of which was a (+)-catechin conjugate] and three other unconjugated metabolites. 3. Excretion of radioactivity in the urine when the bile duct was not cannulated amounted to 44.5% of the dose. 4. In both the urine and bile the new metabolites showed maximum excretion in the (1/2)-1(1/2)h after intravenous injection of [(14)C]catechin. 5. The metabolites m-hydroxyphenylpropionic acid, p-hydroxyphenylpropionic acid, delta-(3-hydroxyphenyl)-gamma-valerolactone and delta-(3,4-dihydroxyphenyl)-gamma-valerolactione originate from the action of the intestinal micro-organisms on the biliary-excreted metabolites of (+)-catechin. These phenolic acid and lactone metabolites are then reabsorped and excreted in the urine. 6. It is proposed that, depending on the route of administration of (+)-catechin, there exists an alternative pathway, involving biliary excretion, for the metabolism of (+)-catechin.     

Biochemistry of metallocenes. Identification of the major metabolite of acetylruthenocene.

1. After oral administration of acetylruthenocene to rats, metabolites were detected in bile and urine. 2. The major metabolite, which is present in both bile and urine, is a glucuronide with the structure: C5H5-Ru-C5H4-CO-CH2-O-C6H9O6 3. The metabolite was identified by mass spectrometry of the permethylated glucuronide and mass spectrometry and n.m.r. of the aglycone. 4. The nature of the metabolite is discussed, and a comparison is made with the metabolism of the benzene analogue, acetophenone.     

Steroid metabolism in the rabbit. Biliary and urinary excretion of metabolites of [4-14C]progesterone

1. [4-(14)C]Progesterone was administered intravenously to anaesthetized male and female New Zealand White rabbits as a single injection or as a 45-60min. infusion. 2. After a single dose about 60% of the radioactivity was recovered in 6hr., and twice as much radioactivity was present in bile as in urine. After infusion total recovery of radioactivity was only about 40% in 6hr., but the relative proportions of metabolites in bile and urine were about the same as after a single dose. 3. Bile and urine samples were hydrolysed successively by beta-glucuronidase, cold acid and hot acid. 4. In bile the major proportion of metabolites appeared in the glucuronide fraction; in urine beta-glucuronidase hydrolysis yielded the greatest amounts of ether-extractable radioactivity, but the greatest proportion of radioactivity could not be extracted by ether from an alkaline solution of the hydrolysed urine. 5. There was no apparent difference in the quantity or distribution of metabolites excreted by male and female animals.     

Investigations on metabolites of vitamin D in rat bile. Separation and partial identification of a major metabolite

1. Young rats with cannulated bile ducts were given 0.34mg. of [1alpha-(3)H]cholecalciferol or 0.54mg. of [(14)C]ergocalciferol by intravenous infusion. Of the radioactivity in the dose of [1alpha-(3)H]cholecalciferol 31% was recovered in bile within 24hr. 2. The metabolites in bile were separated by gradient-elution column chromatography on silicic acid into five components, all more polar than cholecalciferol or 25-hydroxycholecalciferol. [(14)C]Ergocalciferol gave a similar pattern of metabolites in bile. 3. The three most polar metabolites were shown to be ionic. The major component has been identified as a glucuronide conjugate, which was not identical with synthetic cholecalciferyl glucuronide.     

Bile duct ligation promotes covalent drug-protein adduct formation in plasma but not in liver of rats given zomepirac

Acyl glucuronides are reactive electrophilic metabolites of carboxylate drugs, capable of undergoing hydrolysis, rearrangement and covalent binding reactions with proteins in vivo. Such covalent drug-protein adducts may be prerequisites for certain idiosyncratic immune and toxic responses in susceptible individuals. The present study examined the effect of experimental cholestasis on the extent and pattern of formation of protein adducts in plasma and liver of rats given the non-steroidal antiinflammatory drug (NSAID) zomepirac (ZP). Groups of intact, bile-exteriorized and bile duct-ligated rats given a 50 mg/kg i.v. dose of ZP were studied for 24 hr. In intact rats, only 1.4% of the dose was recovered as the sum of ZP, ZP acyl glucuronide (ZAG) and its rearrangement isomers (iso-ZAG) in urine in 24 hr. In bile-exteriorized animals, 0.5% of the dose was recovered in urine in 24 hr, with 31.6% of the dose being recovered in bile (2.7% as ZP, 20.0% as ZAG and 8.9% as iso-ZAG). In the bile duct-ligated group, recovery of dose in 24 hr urine totalled 17.5% (1.7% as ZP, 6.7% as ZAG and 9.1% as iso-ZAG). ZAG and iso-ZAG were measurable in plasma only in the bile duct-ligated group, and covalent binding of ZP to plasma proteins was much higher (5-6 fold) than in intact or bile-exteriorized rats. Total adduct concentrations in liver were not significantly different among the three groups. Immunoblotting using a polyclonal ZP antiserum confirmed that serum albumin was a major target protein in plasma. The major ZP-modified bands in the livers of intact and bile-exteriorized rats were at about 110, 140 and 200 kDa. However, the bands at 110 and 140 kDa were much lower in the livers of bile duct-ligated rats. The results show that about 30% of ZP doses are normally excreted as ZAG and its isomers in bile, with only minor excretion in urine. Bile duct ligation shunts the glucuronide into blood (and urine), strongly promoting adduct formation with plasma proteins, and alters the pattern but not the total quantity of drug-modified proteins formed in the liver.     

Excretion of cholate glucuronide

[3-3H]Cholic acid glucuronide [7 alpha,12 alpha-dihydroxy-3 alpha-O-(beta-D-glucopyranosyluronate)-5 beta- cholan-24-oate] was synthesized and administered to rats prepared with either an external biliary fistula or a ligated bile duct. When bile fistula animals were given either microgram or milligram amounts of the glucuronide, biliary secretion of label was rapid and efficient: greater than 90% of the administered label was secreted within 60 min and total recovery of label in bile was 98.6 +/- 1.2%. Studies in which [14C]taurocholate was included in the dose indicated that this bile acid was secreted into bile significantly more rapidly than was the glucuronide. In animals with ligated bile ducts, urinary excretion was the major route of elimination: after 20 hr, 83.4 +/- 9.3% of the administered dose had been excreted in urine. Urinary excretion of cholate glucuronide was significantly more rapid than that of taurocholate. Gas-liquid chromatographic analysis of the methyl ester acetate derivatives of labeled compounds isolated from bile and urine by chromatography established that the bulk (greater than 70%) of the administered material was secreted in bile or excreted in urine as the intact cholate glucuronide. From these results, we conclude that the glucuronidation of cholic acid produces a derivative which is rapidly and effectively cleared from the circulation and excreted.     

Biliary metabolites of all-trans-retinoic acid in the rat

Biliary metabolites from physiological doses of all-trans-[10-3H]retinoic acid were examined in normal and vitamin A-deficient rats. The bile from normal and vitamin A-deficient rats contained approximately 60% of the administered dose following a 24-h collection period. However, vitamin A-deficient rats show a 6-h delay in the excretion of radioactivity compared to normal rats. Retinoyl-beta-glucuronide excretion was particularly sensitive to the vitamin A status of the rats. In normal rats, retinoyl-beta-glucuronide reached a maximum concentration of 235 pmol/ml of bile 2 h following the dose and then rapidly declined. Vitamin A-deficient rats show a relatively constant concentration of this metabolite (100-150 pmol/ml of bile) over a 10-h collection period. Retinoic acid excretion was low in both normal and deficient rats. The concentration of retinotaurine, a recently identified biliary metabolite, was approximately equal to retinoyl-beta-glucuronide in normal rats and appeared in the bile 2 h later than the glucuronide.     

Analysis of amphetamine-type stimulants and their metabolites in plasma, urine and bile by liquid chromatography with a strong cation-exchange column-tandem mass spectrometry

The aim of this work was to develop and validate a method for analysing amphetamine-type stimulants (ATSs) and their metabolites in plasma, urine and bile by liquid chromatography with a strong cation-exchange column-tandem mass spectrometry, and to apply it to the pharmacokinetic study of ATSs. 3,4-Methylenedioxymethamphetamine, methamphetamine, ketamine and their main metabolites, 4-hydroxy-3-methoxymethamphetamine, 3,4-methylenedioxyamphetamine, p-hydroxymethamphetamine, amphetamine and norketamine, were simultaneously quantified by the new method (50-5000 ng/ml). The coefficients of variation and the percent deviations for the eight compounds were in the range of 0.2 to 5.3% and -9.4 to +12.8%, respectively. The recoveries were over 90% in all biological samples tested. This method was effective for the separation and the identification of ATSs and their main metabolites having amine moieties in plasma, urine and bile, and was applicable to pharmacokinetic analysis of methamphetamine, ketamine and their main metabolites in biological samples. This analytical method should be useful for the pharmacokinetic analysis of ATSs.     

Application of inductively coupled plasma mass spectrometry and high-performance liquid chromatography--with parallel electrospray mass spectrometry to the investigation of the disposition and metabolic fate of 2-, 3- and 4-iodobenzoic acids in the rat

ICP-MS, HPLC-ICP-MS and HPLC-ICP-MS/ESI-MS have been applied to determine the disposition and metabolic fate of 2-, 3- and 4-iodobenzoic acids following intraperitoneal administration at 50 mg kg(-1) to male bile duct cannulated rats. Quantitative excretion balance studies based on the determination of the total iodine content of urine and bile showed that all three iodobenzoic acids were rapidly excreted. Recoveries ranging from 95 to 105% of the administered doses were achieved within 24 h of administration. Metabolite profiles for urine and bile showed extensive metabolism with unchanged iodobenzoic acids forming a minor part of the total. A combination of alkaline hydrolysis and MS enabled the identification of the major metabolites of all three iodobenzoic acids as glycine and ester glucuronide conjugates with very little if any of the parent compounds excreted unchanged.     

Metabolism of coenzyme Q10: biliary and urinary excretion study in guinea pigs.

Biliary and urinary metabolites were examined after intravenous administration of 14C-coenzyme Q10 (14C-CoQ) to guinea pigs. Cumulative recovery of administered radioactivity for up to 8 hours by bile drainage was 4.8%. The greater part of radioactivity was detected in conjugate form. After hydrolyzing with beta-glucuronidase, aglycone fragments were subjected to methylation and reductive acetylation. The main metabolite was demonstrated to be Q acid-1 1,4-hydroquinone diacetate methyl ester (M-1) on HPLC. Then, the main metabolite was assumed to be glucuronide of 2,3-dimethoxy-5-methyl-6-(3'-methyl-5'-carboxy-2'-pentenyl)-1, 4-benzohydroquinone [Q acid-I hydroquinone]. The cumulative urinary recovery of the administered radioactivity over 48 hours was 8.3%. The labeled samples were treated similarly to bile. The urinary metabolites of CoQ10 consisted of unconjugated and conjugated forms. Lyophilized urine was treated as a bile sample and analyzed. The two major metabolites were assigned to be M-1 and Q acid-II 1,4-hydroquinone diacetate methyl ester (M-2). Then, the two metabolites were assumed to be composed of Q acid-I and 2,3-dimethoxy-5-methyl-6-(3'-carboxypropyl)-1,4-benzoquinone (Q acid-II) in free and corresponding hydroquinone conjugate forms. To investigate the effect of exogenous labeled CoQ10 on unlabeled CoQ10 (endogenous) metabolites in urine, simultaneous quantitative determination was performed using deuterium labeled CoQ10 (CoQ10-d5). Urine collected over a 72-hour period after intravenous administration of CoQ10-d5 was processed similarly to that described above and two derivatized metabolites (M-1 and M-2) were quantified by gas chromatography-mass fragmentography with the multi-ion detection method. The analytical results showed that the addition of exogenous labeled CoQ10 did not influence the metabolism (or breakdown) of unlabeled (endogenous) CoQ10.     

In vivo and in vitro metabolism of gomphoside, a cardiotonic steroid with doubly-linked sugar

The metabolism of gomphoside, a cardiotonic steroid glycoside with doubly-linked 4,6-dideoxyhexosulose sugar was studied in vivo in rats, and in vitro using rat liver microsomes. The biliary excretion of metabolites, following intraperitoneal administrative of [3H]gomphoside, was rapid with 68% of radioactivity being collected over 8 h. The metabolites in the bile were principally a water-soluble glucuronide conjugate of gomphoside, and a small amount of chloroform-soluble metabolites. Conversion of [3H]gomphoside to metabolites by microsomes at 37 degrees C reached a maximum of 16% under optimum conditions, producing the same set of metabolites as those in the chloroform-soluble fraction of the bile. The major chloroform-soluble metabolite was the aglycone of gomphoside, viz. gomphogenin or 2 alpha,3 beta, 14-trihydroxy-5 alpha-card-20(22)-enolide. The other major component was recovered gomphoside. Other metabolites were calactin, calotropin, and 2 alpha-hydroxyuzarigenin 3-(4,6-dideoxy-beta-D-arabino-hexopyranoside). Another metabolite, which is a new cardenolide was shown to be 3-epi-gomphogenin or 2 alpha,3 alpha, 14-trihydroxy-5 alpha-card-20(22)-enolide. Gomphoside glucuronide was shown spectroscopically to have the glucuronide residue attached to position 3' of the hexosulose sugar. It was cleaved by beta-D-glucuronidase to gomphoside, and is thus gomphoside 3'-beta-D-glucuronide. The metabolic transformations of gomphoside are summarized in Fig. 5.     

A high pressure liquid chromatographic method for the separation and quantitation of water-soluble radiolabeled benzene metabolites

The glucuronide and sulfate conjugates of benzene metabolites as well as muconic acid and pre-phenyl- and phenylmercapturic acids were separated by ion-pairing HPLC. The HPLC method developed was suitable for automated analysis of a large number of tissue or excreta samples. p-Nitrophenyl [14C]glucuronide was used as an internal standard for quantitation of these water-soluble metabolites. Quantitation was verified by spiking liver tissue with various amounts of phenylsulfate or glucuronides of phenol, catechol, or hydroquinone and analyzing by HPLC. Values determined by HPLC analysis were within 10% of the actual amount with which the liver was spiked. The amount of metabolite present in urine following exposure to [3H]benzene was determined using p-nitrophenyl [14C]glucuronide as an internal standard. Phenylsulfate was the major water-soluble metabolite in the urine of F344 rats exposed to 50 ppm [3H]benzene for 6 h. Muconic acid and an unknown metabolite which decomposed in acidic media to phenylmercapturic acid were also present. Liver, however, contained a different metabolic profile. Phenylsulfate, muconic acid, and pre-phenylmercapturic acids as well as an unknown with a HPLC retention time of 7 min were the major metabolites in the liver. This indicates that urinary metabolite profiles may not be a true reflection of what is seen in individual tissues.     

Specific deuterium labelling and computerized gas chromatography -- mass spectrometry in studies on the metabolism in vivo of a steroid sulphate in the rat.

The metabolism of 3beta-hydroxy-5alpha-pregnan-20-one sulphate was studied in bile fistula rats and in isolated perfused livers. Computerized gas chromatography--mass spectrometry, in combination with specific deuterium-labelling, was employed to follow the metabolic transformations. Male animals excreted metabolites into bile more rapidly than females, a finding which could be correlated with the preferential formation of glucuronide conjugates in the male liver. The major metabolic pathway in male rats involved the steps: hydrolysis, 2alpha-hydroxylation, oxidoreduction at C-3 and glucuronide conjugation, yielding 2alpha, 3alpha-dihydroxy-5alpha-pregnan-20-one glucuronide as the major metabolite. Only traces of the injected steroid sulphate were detected in bile from male animals. In contrast, the administered compound was the major steroid excreted in bile of female rats, where the main metabolite was identified as 3beta,15beta-dihydroxy-5alpha-pregnan-20-one sulphate. A minor metabolite, 3beta,16alpha-dihydroxy-5alpha-pregnan-20-one, was found as a monosulphate in female rats and as both a disulphate and a glucuronide conjugate in male rats. The deuterium content of the sulphated 15beta-and 16alpha-hydroxylated metabolites was consistent with metabolic pathways involving direct hydroxylation of the injected steroid sulphate. The results obtained from the liver perfusions were essentially the same as those from the experiments with bile fistula animals. This indicates that all the observed metabolic reactions took place in the liver.     

Identification of the major antioxidative metabolites in biological fluids of the rat with ingested (+)-catechin and (-)-epicatechin.

(+)-Catechin and (-)-epicatechin are known to be biologically effective antioxidants present in the human diet, particularly in wine and tea. We studied the metabolism of these compounds to elucidate the truly active structures in biological fluids by their oral administration to rats. Without any treatment with beta-glucuronidase and sulfatase, a pair of metabolites were detected at much higher concentrations in the plasma, bile, and urine than the originally ingested compounds. Each major metabolite found in the plasma at the highest concentration was excreted in both the bile and urine, and was purified from urine. Their chemical structures were established to be (+)-catechin 5-O-beta-glucuronide and (-)-epicatechin 5-O-beta-glucuronide by MS and NMR analyses. These glucuronide conjugates exhibited high antioxidative activities as superoxide anion radical scavengers like their parent compounds. It is concluded that (+)-catechin 5-O-beta-glucuronide and (-)-epicatechin 5-O-beta-glucuronide are the biologically active in vivo structures of the ingested polyphenolic antioxidants.     

Effects of Δ9-tetrahydrocannabinol on the disposition of d-amphetamine in the rat

Δ⁹-Tetrahydrocannabinol (THC), 10 or 50 mg/kg, administered intragastrically one hour before intraperitoneal injection of ¹⁴C-d-amphetamine, 4 mg/kg, did not modify the disappearance from the blood or the tissue distribution of amphetamine in fasted rats. Furthermore, THC did not influence the urinary excretion of unchanged amphetamine or its major metabolite, p-hydroxyamphetamine, in these animals. However, when the interval between drug treatments was increased to two hours, THC, 10 mg/kg, minimally reduced the rate of disappearance of ¹⁴C-amphetamine from the blood of fasted rats. This effect was much more pronounced in rats which had food available throughout the experiment. THC also inhibited the urinary excretion of total radioactivity as well as ¹⁴C-amphetamine metabolites in fed but not in fasted animals during the first 4 hours following ¹⁴C-amphetamine injection. In addition, fasted rats excreted significantly more total radioactivity and unchanged ¹⁴C-amphetamine than fed rats during the 0 – 4 hour urine collection interval. The pH of urine collected during this and all other periods was significantly more acid for faster than fed rats. It is concluded that THC can inhibit amphetamine metabolism in rats depending upon the time interval between the administration of the two drugs and the dietary state of the animals.     

Differences between the biliary excretion of tri[14C]methyl-1-(3-hydroxyphenyl)ammonium iodide in Wistar and Gunn rats

1. The biliary excretion of [¹⁴C]trimophonium iodide [tri[¹⁴C]methyl(3-hydroxyphenyl)ammonium iodide] was studied in normal Wistar animals and in jaundiced homozygous Gunn rats. 2. In normal Wistar rats small amounts of radioactivity (approx. 3% of the dose in 4h) were excreted in bile as two glucuronide conjugates, i.e. [¹⁴C]trimophonium glucuronide [tri[¹⁴C]methyl-(3-oxyphenyl)ammonium glucuronide] (85%) and 3-di[¹⁴C]methylaminophenyl glucuronide (10–15%). Only minor amounts of the unchanged drug were detected in bile. 3. In the homozygous jaundiced Gunn rat large amounts of radioactivity (26% of the dose in 4h) were eliminated in bile as [¹⁴C]trimophonium glucuronide alone. The quantitative excretion of this metabolite in Gunn rat bile was about ten times that in normal animals. 4. It is proposed that the biochemical lesion in the homozygous Gunn rat may indirectly affect the biliary transport of exogenous glucuronides across the canalicular membrane.     

Metabolism of a glutathione conjugate of 2-hydroxyoestradiol-17β in the adult male rat

Adult male rats with cannulated or ligated bile ducts were given S-(2-hydroxyoestradiol-1-yl)[(35)S]glutathione, S-(2-hydroxy[6,7-(3)H(2)]oestradiol-1-yl)glutathione or S-(2-hydroxyoestradiol-1-yl)[glycine-(3)H]glutathione by intraperitoneal injection. The recovery of radioactivity in the bile of bile duct-cannulated rats was 33-86% and in the urine of bile duct-ligated rats was 54-105%. Oestrogen thioether derivatives of glutathione, cysteinylglycine, cysteine and N-acetylcysteine were isolated from bile; only the N-acetylcysteine derivatives could be identified in the urine. The steroid moiety was characterized by microchemical tests before and after treatment with Raney nickel: 2-hydroxyoestradiol-17beta was released from the glutathione conjugate, and 2-hydroxyoestrone and 2-hydroxyoestrone 3-methyl ether from the other conjugates. From intact rats the recovery of administered radioactivity was about 15% in the urine and 5% in the faeces over a period of several days and the radioactivity appeared to be largely protein-bound. The results demonstrate that injected oestrogen-glutathione conjugate undergoes conversion into N-acetylcysteine derivatives in vivo. Oestrogen-glutathione conjugates formed in the intact rat may be excreted in an apparently non-steroidal, possibly protein-bound form, which would not be detected by current analytical techniques.     

Estrogen mercapturic acids in the adult male rat

N-Acetylcysteine derivatives of catechol estrogens have been isolated from the urine of adult male hooded rats with ligated bile ducts, following injection of [4-¹⁴C]2-hydroxyestradiol-17β and of [4-¹⁴C]estradiol-17β-By application of double isotope methods previously described, it was shown that 2-hydroxyestradiol-17β was converted into mercapturic acids in a yield of 6–8%, confirming two previous experiments with bile duct cannulated rats, and that estradiol-17β was converted into mercapturic acids to the extent of 3–6%. Since these figures are small, and since it has been shown that in two women estrogen mercapturic acids were not formed, it appears that this class of compound will not provide an answer to the problem of unidentified water-soluble metabolites of the estrogens.     

Species variations in the biliary and urinary excretion of arsenate, arsenite and their metabolites

In most mammalian species, inorganic arsenicals are extensively biotransformed and excreted both in unchanged form and as metabolites. In the bile of rats receiving arsenate (AsV) or arsenite (AsIII) we have identified monomethylarsonous acid (MMAsIII), purportedly the most toxic metabolite of inorganic arsenic. As rats are not commonly accepted for studying arsenic metabolism, we carried out a comparative investigation on the excretion of AsV, AsIII and their metabolites in five animal species in order to determine whether they also form MMAsIII from AsV and AsIII. Anaesthetised bile duct-cannulated rats, mice, hamsters, rabbits, and guinea pigs were injected with AsV or AsIII (50 micromol/kg, i.v.) and their bile and urine was collected for 2 h. Arsenic in bile and urine was speciated by HPLC-hydride generation-atomic fluorescence spectrometry and the excretion rates of AsV, AsIII, monomethylarsonic acid (MMAsV), MMAsIII and dimethylarsinic acid (DMAsV) were quantified. All species injected with AsV excreted arsenic preferentially into urine, whereas all animals receiving AsIII, except rabbits, delivered more arsenic into bile than urine. Bile contained almost exclusively trivalent arsenic (i.e. AsIII and/or MMAsIII), whereas AsV, AsIII and DMAsV appeared in urine. Except for guinea pigs, which do not methylate arsenic, the other species formed MMAsIII and excreted it into bile. Having excreted as much as 8% of the dose of AsIII or AsV in 2 h as MMAsIII, rats were by far the most efficient producers of this supertoxic metabolite. Thus, although the rat is not a good model for studying long-term arsenic disposition, this species appears especially valuable in studies on AsIII methyltransferase and in vivo formation of MMAsIII.