Ubiquity of arsenobetaine in marine animals and degradation of arsenobetaine by sedimentary micro-organisms

Arsenic compounds were extracted with chloroform/methanol/water from tissues of marine animals (four carnivores, five herbivores, five plankton feeders). The extracts were purified by cation and anion exchange chromatography. Arsenobetaine [(CH₃)₃As⁺CH₂COO^?], dimethylarsinic acid [(CH₃)₂AsOOH], trimethylarsine oxide [(CH₃)₃AsO] and arsenite, arsenate, and methylarsonic acid [(CH₃)AsO(OH)₂] as a group with the same retention time were identified by high-pressure liquid chromatography. Arsenic was determined in the collected fractions by graphite furnace atomic absorption spectrometry. Arsenobetaine found in all the animals was almost always the most abundant arsenic compound in the extracts. These results show that arsenobetaine is present in marine animals independently of their feeding habits and trophic levels. Arsenobetaine-containing growth media (ZoBell 2216E; solution of inorganic salts) were mixed with coastal marine sediments as the source of microorganisms. Arsenobetaine was converted in both media to trimethylarsine oxide and trimethylarsine oxide was converted to arsenite, arsenate or methylarsonic acid but not to dimethylarsinic acid. The conversion rates in the inorganic medium were faster than in the ZoBell medium. Two dominant bacterial strains isolated from the inorganic medium and identified as members of the Vibro–Aeromonas group were incapable of degrading arsenobetaine.     

Conversion of arsenobetaine to dimethylarsinic acid by arsenobetaine-decomposing bacteria isolated from coastal sediment

Arsenobetain [(CH₃)₃As⁺CH₂COO^-]-containing growth media (1/5 ZoBell 2216E and solution of inorganic salts) were inoculated with two bacterial strains, which were isolated from a coastal sediment and identified as members of the Vibro-Aeromonas group, and incubated under aerobic and anaerobic conditions. Arsenobetaine was converted to a metabolite only under aerobic conditions. This arsenic metabolite was identified as dimethylarsinic acid [(CH₃)₂AsOOH] by hydride generation/cold trap/GC MS/SIM analysis and high-performance liquid-chromatographic behaviour. The conversion pattern shown by these arsenobetaine-decomposing bacteria (that is, arsenobetaine → dimethylarsinic acid) was fairly different from that shown by the addition of sediment itself as the source of arsenobetaine-decomposing micro-organisms (that is, arsenobetaine → trimethylarsine oxide → inorganic arsenic). This result suggests to us that various micro-organisms, including the arsenobetaine-decomposing bacteria isolated in this study, participate in the degradation of arsenobetaine in marine environments.     

Arsenobetaine-decomposing Ability of Marine Microorganisms Occurring in Particles Collected at Depths of 1100 and 3500 Meters

The arsenobetaine-decomposing ability of microorganisms occurring in sinking particles, which play a main role in the vertical transport of organic substances produced in the photic zone, was investigated. The microorganisms in particles collected in the deep sea, 1100 and 3500 m in depth, clearly showed decomposing ability. With the particles from 1100 m, the degradation products were the same as those produced by microorganisms occurring in sources in the photic zone, i.e. trimethylarsine oxide (TMAO), dimethylarsinic acid (DMA) and inorganic arsenic(V). At 3500 m, the degradation activity was diminished, smalls amount of DMA and TMAO being produced. These results suggest that arsenobetaine contained in the animals starts to degrade immediately after the death of the animals and their transformation to particles. The degradation of arsenobetaine to inorganic arsenic in our tentative arsenic cycle in marine ecosystems (inorganic arsenic to inorganic arsenic via the biosynthesis of arsenobetaine) may apply to the deep sea as well as to the photic zone. © 1997 by John Wiley & Sons, Ltd.     

The fate of organoarsenic compounds in marine ecosystems

Microbial degradation experiments were performed with each standard arsenical [arsenobetaine, trimethylarsine oxide, dimethylarsinic acid, methanearsonic acid, inorganic arsenic(V) and inorganic arsenic(III)]. As typical origins for marine micro-organisms, sediments, macro-algae, mollusc intestine and suspended substances were used. The results were from these experiments led us to the following conclusions: (1) there is an arsenic cycle which begins with the methylation of inorganic arsenic on the route to arsenobetaine and terminates with the complete degradation of arsenobetaine to inorganic arsenic; (2) all the organoarsenic compounds which are derived from inorganic arsenic in seawater, through the food chains, have the fate that they, at least in part, finally return to the original inorganic arsenic.     

Conversion of arsenobetaine by intestinal bacteria of a mollusc Liolophura japonica chitons

The intestinal micro-organisms of Liolophura japonica chitons converted arsenobetaine [(CH₃)₃As⁺CH₂COO^?] to trimethylarsine oxide [(CH₃)₃AsO] and dimethylarsinic acid [(CH₃)₂AsOOH] in the arsenobetaine-containing 1/5 ZoBell 2216E medium under aerobic conditions, no conversion being observed in an inorganic salt medium. This conversion pattern of arsenobetaine → trimethylarsine oxide ← dimethylarsinic acid was comparable with that shown by the microorganisms associated with marine macroalgae. On the other hand, no conversion was observed in either medium under anaerobic conditions.     

Can Humans Metabolize Arsenic Compounds to Arsenobetaine?

Arsenic compounds were determined in 21 urine samples collected from a male volunteer. The volunteer was exposed to arsenic through either consumption of codfish or inhalation of small amounts of (CH₃)₃As present in the laboratory air. The arsenic compounds in the urine were separated and quantified with an HPLC–ICP–MS system equipped with a hydraulic high-pressure nebulizer. This method has a determination limit of 0.5 μg As dm⁻³ urine. To eliminate the influence of the density of the urine, creatinine was determined and all concentrations of arsenic compounds were expressed in μg As g⁻¹ creatinine. The concentrations of arsenite, arsenate and methylarsonic acid in the urine were not influenced by the consumption of seafood. Exposure to trimethylarsine doubled the concentration of arsenate and increased the concentration of methylarsonic acid drastically (0.5 to 5 μg As g⁻¹ creatinine). The concentration of dimethylarsinic acid was elevated after the first consumption of fish (2.8 to 4.3 μg As g⁻¹ creatinine), after the second consumption of fish (4.9 to 26.5 μg As g⁻¹ creatinine) and after exposure to trimethyl- arsine (2.9 to 9.6 μg As g⁻¹ creatinine). As expected, the concentration of arsenobetaine in the urine increased 30- to 50-fold after the first consumption of codfish. Surprisingly, the concentration of arsenobetaine also increased after exposure to trimethylarsine, from a background of approximately 1 μg As g⁻¹ creatinine up to 33.1 μg As g⁻¹ creatinine. Arsenobetaine was detected in all the urine samples investigated. The arsenobetaine in the urine not ascribable to consumed seafood could come from food items of terrestrial origin that—unknown to us—contain arsenobetaine. The possibility that the human body is capable of metabolizing trimethyl- arsine to arsenobetaine must be considered. © 1997 by John Wiley & Sons, Ltd.     

Degradation of a tetramethylarsonium salt by microorganisms occurring in sediments and suspended substances under both aerobic and anaerobic conditions

Microbial degradation of a tetramethylarsonium salt during incubation at 25°C was investigated under both aerobic and anaerobic conditions. Two media (1/5 ZoBell 2216E and inorganic salt medium), added with the sediments or suspended substances as the sources of the microorganisms, were used. Degradation of the tetramethylarsonium salt occurred only in the ZoBell medium: under anaerobic conditions, trimethylarsine oxide and dimethylarsinic acid were derived with the sediments, and dimethylarsinic acid with the suspended substances, the salt degrading more rapidly with the former than with the latter. Small amounts of two metabolites, trimethylarsine oxide and inorganic arsenic(V), was also derived in the aerobically incubated ZoBell medium added with the suspended substances. This result means that the tetramethylarsonium salt is degraded to inorganic arsenic, which is the starting material for arsenic circulation in marine ecosystems, via trimethylarsine oxide and dimethylarsinic acid.     

Arsenic Compounds Accumulated in Sedimentary Microorganisms Cultivated in Media Containing Several Arsenicals

Sediments, as sources of microorganisms, were added to two kinds of media, 1/5 ZoBell 2216E and a solution of inorganic salts, which contained inorganic arsenic(III), inorganic arsenic(V), methanearsonic acid, dimethyl- arsinic acid, trimethylarsine oxide, tetramethylarsonium salt or arsenocholine. After 17 days of incubation at 20 °C, the arsenicals that had accumulated in the microorganisms were analysed by high-performance liquid chromatography (HPLC). While the more toxic arsenicals [inorganic arsenic(III), inorganic arsenic(V), methanearsonic acid, dimethylarsinic acid] were not converted in the microorganisms, trimethylarsine oxide and tetramethylarsonium salt were considerably degraded to inorganic arsenic(V), and arsenocholine to arsenobetaine. Arsenobetaine that had accumulated in the microorganisms was extracted and confirmed by thin-layer chromatography (TLC) and fast atom bombardment (FAB) mass spectrometry.     

Arsenobetaine and arsenocholine: Two marine arsenic compounds without embryotoxity

The embryotoxicity of carboxymethyl(trimethyl)arsonium bromide [arsenobetaine,(CH₃)₃As⁺CH₂COO⁻] and of 2-hydroxyethyl(trimethyl)arsonium bromide [arsenobetaine, (CH₃)₃As⁺CH₂CH₂OHBr] was explored. Sprague-Dawley rat embryos with intact yolk sacs were removed on day 11 of gestation and grown in a culture medium for 24 h in the presence and absence of rat liver (S-9) homogenate. Solutions of arsenobetaine or arsenocholine in dimethyl sulfoxide [DMSO, (CH₃)₂SO] (0.03 cm³) were added to the media to achieve concentrations of 20 m̈g arsenic compound per cm³ of medium. After 24 h the circulation and heart beat were monitored (indicator of embryolethality); in addition the crown-to-rump lengths were measured and the neural structures (somites) and limb buds observed (indicator of embryotoxicity). No evidence for embryotoxicity or embryolethality was found in the absence or the presence of S-9. These results indicate that arsenobetaine, the most common arsenic compound found in seafood at concentrations from several micrograms to several hundred micrograms arsenic per gram, lacks subacute and acute prenatal toxicity.     

Preparation and certification of arsenobetaine reference material NMIJ CRM 7901-a

An arsenobetaine [(CH₃)₃As⁺CH₂COO⁻] solution reference material, NMIJ CRM 7901-a, intended for use in the speciation of arsenic compounds, was developed and certified by the National Metrology Institute of Japan (NMIJ), part of the National Institute of Advanced Industrial Science and Technology (AIST). The high-purity arsenobetaine powder was synthesized from trimethylarsine [(CH₃)₃As], and it was dissolved in water in order to prepare 20 mg kg⁻¹ of arsenobetaine standard solution. The solution was bottled in 500 bottles (each containing 10 ml). Certification of the CRM for arsenobetaine was conducted by NMIJ. The concentration of As was determined by four independent analytical techniques (ICP–MS, ICP–OES, GFAAS and LC–ICP–MS), and each result was converted to the arsenobetaine concentration by applying an appropriate factor. The arsenobetaine concentration in the CRM was thus certified.     

Formation of arsenobetaine from arsenocholine by micro-organisms occurring in sediments

As one of the experiments to pursue marine circulation of arsenic, we studied microbiological conversion of arsenocholine to arsenobetaine, because arsenocholine may be a precursor of arsenobetaine in these ecosystems. Two culture media, 1/5 ZoBell 2216E and an aqueous solution of inorganic salts, were used in this in vitro study. To each medium (25 cm³) were added synthetic arsenocholine (0.2%) and about 1 g of the sediment, and they were aerobically incubated at 25°C in the dark. These conversion experiments were performed in May and July 1990. In both seasons, two or three metabolites were derived in each mixture. These metabolites were purified using cation-exchange chromatography. Their structures were confirmed as arsenobetaine, trimethylarsine oxide and dimethylarsinic acid by high-performance liquid chromatography, thin-layer chromatography, FAB mass spectrometry and a combination of gas-chromatographic separation with hydride generation followed by a cold-trap technique and selected-ion monitoring mass spectrometric analysis. From this and other evidence it is concluded that, in the arsenic cycle in these marine ecosystems, as recently postulated by us, the pathway arsenocholine → arsenobetaine → trimethylarsine oxide → dimethylarsinic acid → methanearsonic acid → inorganic arsenic can be carried out by micro-organisms alone.     

The methylation of arsenic in marine sediments

Laboratory studies have shown that microorganisms present in both natural marine sediments and sediments contaminated with mine-tailings are capable of methylating arsenic under aerobic and anaerobic conditions. Incubation of sediments with culture media produced volatile arsines [including AsH₃, (CH₃)AsH₂, and (CH₃)₃As] as well as the methylarsenic(V) compounds (CH₃)_nAs(O) (OH)_3?n (n = 1, 2, 3). The concentration of the arsines increased and then decreased in a growth and decay pattern reminiscent of the methylation and demethylation of mercury. Thus, arsenic speciation varied with time, being controlled by the biochemical activity of the dominant microbe(s) at the time of sampling, and changing in response to the ecological succession within the microbial community. The analysis of the interstitial waters of sediments collected from several British Columbia (Canada) coastal sites gave results that were consistent with the culture experiments, in that the methylarsenicals were ubiquitous, but present only in small amounts. It is estimated that methylarsenic(V) species account for less than 1% of the arsenic present in porewaters. The actual proportion was dependent on a number of factors but, contrary to prevailing viewpoints, there was no relationship to the organic content of the sediments, nor did methylation occur only in the presence of high arsenic concentrations. Instead, all of the evidence was consistent with in situ microbial methylation and demethylation processes that are similar to the arsenic transformations that occur in soil ecosystems. The results are discussed in terms of the cycling of arsenic in the marine environment and within the marine food web.     

Identification of arsenobetaine and a tetramethylarsonium salt in the clam Meretrix lusoria

Chemical forms of arsenic were examined in six tissues (gill, mid-gut gland, siphon, foot, mantle and adductor muscle) of the clam Meretrix lusoria. The gill was found to contain higher levels of arsenic than the other tissues. Regardless of the nature of the tissues, the presence of arsenobetaine was established by HPLC0ICP; it was a minor arsenic compound in gill but a major one in the other tissues. The major arsenic compound in gill, which was more cationic than arsenobetaine, was obtained in a relatively pure state by ion-exchange chromatography, gel filtration and HPLC. It was positive to the Dragendorff reagent and iodine vapor but negative to ninhydrin reagent. Its ¹HNMR spectrum exhibited only one signal at δ 1.7 (singlet) and its FAB mass spectrum gave a base peak at m/e 135 [(CH₃)₄As⁺] and two significant peaks at m/e 120 [(CH₃)₃As] and 106 [(CH₃)₂AsH]. These results suggested that the major arsenic compound in gill exists as a tetramethylarsonium salt (CH₃)₄As⁺ · X^?. The tetramethylarsonium salt was also found as a minor component in the tissues other than the gill.     

Decomposition of organoarsenic compounds for total arsenic determination in marine organisms by the hydride generation technique

The conditions necessary for the complete decomposition of six organic arsenic compounds, namely methylarsonic acid (MMAA), dimethylarsinic acid (DMAA), trimethylarsine oxide, tetramethylarsonium iodide, arsenocholine bromide (AsC) and arsenobetaine (AB), were investigated. The degree of decomposition of the arsenic compounds was monitored using a hydride generation (HYD) technique, because the response from this system depends strongly on the chemical species of arsenic, with inorganic arsenic (the expected product from these decomposition experiments) giving a much more intense HYD signal than the organic arsenic compounds. The arsenic compounds were decomposed by heating them with three types of acid mixture, namely HNO₃? HClO₄, HNO₃? HClO₄? HF, or HNO₃? HClO₄? H₂SO₄. Both MMAA and DMAA were decomposed completely using any of the mixed acids at a decomposition temperature of 200 °C or higher. The HNO₃? HClO₄? H₂SO₄ mixture was the most effective for decomposing AsC and AB, which are the most difficult compounds among all types of organic arsenic compound to decompose and render inorganic. The complete decomposition of AB was only achieved, however, when the temperature was 320 °C or higher, and the sample was evaporated to dryness. When the residue from this treatment was examined by high‐performance liquid chromatography combined with inductively coupled plasma atomic emission spectrometry, all of the arsenic was found to be present as arsenic(V). The optimized conditions (HNO₃? HClO₄? H₂SO₄ at 320 °C) for decomposing AB were then used to determine the total amount of arsenic in marine organisms known to contain AB. Copyright © 2005 John Wiley & Sons, Ltd.     

Tissue accumulation and distribution of arsenic compounds in three marine fish species: relationship to trophic position

In this study the accumulation and distribution of arsenic compounds in marine fish species in relation to their trophic position was investigated. Arsenic compounds were measured in eight tissues of mullet Mugil cephalus (detritivore), luderick Girella tricuspidata (herbivore) and tailor Pomatomus saltatrix (carnivore) by high performance liquid chromatography–inductively coupled plasma‐mass spectrometry. The majority of arsenic in tailor tissues, the pelagic carnivore, was present as arsenobetaine (86–94%). Mullet and luderick also contained high amounts of arsenobetaine in all tissues (62–98% and 59–100% respectively) except the intestines (20% and 24% respectively). Appreciable amounts of dimethylarsinic acid (1–39%), arsenate (2–38%), arsenite (1–9%) and trimethylarsine oxide (2–8%) were identified in mullet and luderick tissues. Small amounts of arsenocholine (1–3%), methylarsonic acid (1–3%) and tetramethylarsonium ion (1–2%) were found in some tissues of all three species. A phosphate arsenoriboside was identified in mullet intestine (4%) and from all tissues of luderick (1–6%) except muscle. Pelagic carnivore fish species are exposed mainly to arsenobetaine through their diet and accumulate the majority of arsenic in tissues as this compound. Detritivore and herbivore fish species also accumulate arsenobetaine from their diet, with quantities of other inorganic and organic arsenic compounds. These compounds may result from ingestion of food and sediment, degradation products (e.g. arsenobetaine to trimethylarsine oxide; arsenoribosides to dimethylarsinic acid), conversion (e.g. arsenate to dimethylarsinic acid and trimethylarsine oxide by bacterial action in digestive tissues) and/or in situ enzymatic activity in liver tissue. Copyright © 2001 John Wiley & Sons, Ltd.     

A comparative study on acute toxicity of methylarsonic acid,dimethylarsinic acid and trimethylarsine oxide in mice

The acute toxicity of methylarsonic acid, CH₃AsO(OH)₂ (MAA), dimethylarsininc acid, (CH₃)₂AsO(OH) (DMAA), and trimethylarsine oxide, (CH₃)₃AsO (TMAO), were examined in mice with oral administration. The LD₅₀ values of MAA, DMAA and TMAO were 1.8, 1.2 and 10.6 g kg^?1 respectively. The toxicity of MAA and DMAA was very much lower than that for inorganic arsenic compounds. It was shown that TMAO has a similar acute toxicity to arsenobetaine. On the other hand, when the mice were administered 14.4 g kg^?1 of TMAO once only orally, a garlic-like odor (trimethylarsine, (CH₃)₃As) was definitely detectable in the exhalation of the animals by the human olfactory sense within about a few minutes.     

Organoarsenic compounds in plants and soil on top of an ore vein

Plants and soil collected above an ore vein in Gasen (Austria) were investigated for total arsenic concentrations by inductively coupled plasma mass spectrometry (ICP‐MS). Total arsenic concentrations in all samples were higher than those usually found at non‐contaminated sites. The arsenic concentration in the soil ranged from ∼700 to ∼4000 mg kg⁻¹ dry mass. Arsenic concentrations in plant samples ranged from ∼0.5 to 6 mg kg⁻¹ dry mass and varied with plant species and plant part. Examination of plant and soil extracts by high‐performance liquid chromatography–ICP‐MS revealed that only small amounts of arsenic (<1%) could be extracted from the soil and the main part of the extractable arsenic from soil was inorganic arsenic, dominated by arsenate. Trimethylarsine oxide and arsenobetaine were also detected as minor compounds in soil. The extracts of the plants (Trifolium pratense, Dactylis glomerata, and Plantago lanceolata) contained arsenate, arsenite, methylarsonic acid, dimethylarsinic acid, trimethylarsine oxide, the tetramethylarsonium ion, arsenobetaine, and arsenocholine (2.5–12% extraction efficiency). The arsenic compounds and their concentrations differed with plant species. The extracts of D. glomerata and P. lanceolata contained mainly inorganic arsenic compounds typical of most other plants. T. pratense, on the other hand, contained mainly organic arsenicals and the major compound was methylarsonic acid. Copyright © 2002 John Wiley & Sons, Ltd.     

Metabolism of arsenic compounds by the blue mussel mytilus edulis after accumulation from seawater spiked with arsenic compounds

Blue mussels (Mytilus edulis) were exposed to 100 μg As dm^?3 in the form of arsenite, arsenate, methylarsonic acid, dimethylarsinic acid, arsenobetaine, arsenocholine, trimethylarsine oxide, tetramethylarsonium iodide or dimethyl-(2-hydroxyethyl)arsine oxide in seawater for 10 days. The seawater was renewed and spiked with the arsenic compounds daily. Analyses of water samples taken 24 h after spiking showed that arsenobetaine and arsenocholine had been converted to trimethylarsine oxide, whereas trimethylarsine oxide and tetramethylarsonium iodide were unchanged. Arsenobetaine was accumulated by mussels most efficienty, followed in efficiency by arsenocholine and tetramethylarsonium iodide. None of the other arsenic compounds was significantly accumulated by the mussels. Extraction of mussel tissues with methanol revealed that control mussels contained arsenobetaine, a dimethyl-(5-ribosyl)arsine oxide and an additional arsenic compound, possibly dimethylarsinic acid. Mussels exposed to arsenobetaine contained almost all their experimentally accumulated arsenic as arsenobetaine, and mussels exposed to tetramethylarsonium iodide contained it as the tetramethylarsonium compound. Mussels exposed to arsenocholine had arsenobetaine as the major arsenic compound and glycerylphosphorylarsenocholine as a minor arsenic compound in their tissues. The results show that arsenobetaine and arsenocholine are efficiently accumulated from seawater by blue mussels and that in both cases the accumulated arsenic is present in the tissues as arsenobetaine. Consequently arsenobetaine and/or arsenocholine present at very low concentrations in seawater may be responsible for the presence of arsenobetaine in M. edulis and probably also among other marine animals. The quantity of arsenobetaine accumulated by the mussels decreases with increasing concentrations of betaine. HPLC-ICP-MS was found to be very powerful for the investigation of the metabolism of arsenic compounds in biological systems.     

The chemical form and acute toxicity of arsenic compounds in marine organisms

A method for the separation and identification of inorganic and methylated arsenic compounds in marine organisms was constructed by using a hydride generation/cold trap/gas chromatography mass spectrometry (HG/CT/GC MS) measurement system. The chemical form of arsenic compounds in marine organisms was examined by the HG/CT/GC MS system after alkaline digestion. It was observed that trimethylarsenic compounds were distributed mainly in the water-soluble fraction of muscle of carnivorous gastropods, crustaceans and fish. Also, dimethylated arsenic compounds were distributed in the water-soluble fraction of Phaeophyceae. It is thought that most of the trimethylated arsenic is likely to be arsenobetaine since this compound released trimethylarsine by alkaline digestion and subsequent reduction with sodium borohydride. The major arsenic compound isolated from the water-soluble fraction in the muscle and liver of sharks was identified as arsenobetaine from IR, FAB Ms data, NMR spectra and TLC behaviour. The acute toxicity of arsenobetaine was studied in male mice. The LD₅₀ value was higher than 10 g kg⁻¹. This compound was found in urine in the non-metabolized form. No particular toxic symptoms were observed following administration. These results suggest that arsenobetaine has low toxicity and is not metabolized in mice. The LD₅₀ values of other minor arsenicals in marine organisms, trimethylarsine oxide, arsenocholine and tetramethylarsonium salt, were also examined in mice.     

Arsenic Compounds in Terrestrial Organisms I: Collybia maculata,Collybia butyracea and Amanita muscaria from Arsenic Smelter Sites in Austria

Three mushroom species from two old arsenic smelter sites in Austria were analyzed for arsenic compounds. The total arsenic concentrations were determined by ICP–MS. Collybia maculata contained 30.0 mg, Collybia butyracea 10.9 mg and Amanita muscaria 21.9 mg As kg⁻¹ dry mass. The arsenic compounds extracted with methanol/water (9:1) from the dried mushroom powders were separated by HPLC on anion-exchange and reversed-phase columns and detected by ICP-MS using a hydraulic high-pressure nebulizer. In Collybia maculata almost all arsenic is present as arsenobetaine. Collybia butyracea contained mainly arsenobetaine (8.8 mg As kg⁻¹ dry mass) and dimethylarsinic acid (1.9 mg As kg⁻¹). Amanita muscaria contained arsenobetaine (15.1 mg As kg⁻¹), traces of arsenite, dimethylarsinic acid and arsenate, and surprisingly arsenocholine (2.6 mg As kg⁻¹) and a tetramethylarsonium salt (0.8 mg As kg⁻¹). © 1997 by John Wiley & Sons, Ltd.