Possible key intermediates in arsenic biochemistry: Synthesis and identification by liquid chromatography electrospray ionization mass spectrometry and high resolution mass spectrometry

Arsenic is a type 1 carcinogen and its toxicity is critically dependent on chemical speciation. However, after decades of research, the biogenesis of at least fifty naturally occurring arsenic species is still not well understood.Here, based on experimental work, it is proposed a set of pathways for the formation of multiple arsenic species that might help to clarify the present situation.These are focused on the thiol protein arsenic bond and on its interaction with reactive metabolites. In fact, arsenic bound to glutathione interacting with sulfur adenosyl methionine (SAM), MethylCB₁₂ and AdoCB₁₂, forms a number of complexes that might be key intermediates in arsenic biochemistry. These include dimethylarsino glutathione (DMAG) m/z 412 [M + H]⁺, synthesized non-enzymatically from glutathione and cacodylate. Trimethylarsonio glutathione (TMAG) m/z 426 [M]⁺ synthesized from DMA, GSH and SAM, apparently by a classical Challenger methylcarbonium attack. Tetramethyl arsonium ion m/z 135 [M]⁺ is formed in a third step, apparently by carbanion methylation. The presence of trimethylarsine oxide (TMAO) m/z 137 [M + H]⁺ is attributed to the hydrolysis of TMAG or TMA, or to carbanion methylation of dimethylarsinoyl glutathione (m/z 428 [M]⁺) formed from cacodylate and GSH. Cantoni type attacks of DMAG on SAM were unsuccessful, eventually due to competition of the trivalent S⁺ atom of SAM for the As^III atom attack. The presence of dimethylarsonio diglutathione (DMADG m/z 717 [M]⁺), is suggested to result from a GS^- attack on dimethylarsenoyl glutathione (m/z 428 [M + H]⁺). The presence of dimethylarsenoyladenosine (m/z 372 [M + H]⁺), trimethylarsenosugar adenine (m/z 370 [M]⁺), and dimethylthioarsenosugar adenine (m/z 388 [M + H]⁺), is explained by the synthesis of the pecursor dimethylarsonio-adenosine glutathione DMAAG (m/z 661 [M]⁺), a likely source of oxo-and trimethylated arsenosugars, as well as of thio-arsenosugars by the cleavage of its S-C bond. The results gathered suggest that cell vacuoles might play a major role in arsenic metabolism, and that the dominance of oxo-As sugars, in algae extracts, may be supported by a mechanism of synthesis independent of DMAAG (m/z 661).They also offer an explanation for the reason why arsenobetaine, and tetramethylarsonium are loosely bound to biotic tissues. Four arsenic species new to science, to the best of our knowledge, and a number of known arsenic compounds were synthesized in this work, identified by HPLC–ESI-MSⁿ and FTICR–ESI-MS, and suggestions regarding their mechanisms of synthesis were advanced. These results provide a framework for arsenic biochemistry which may explain the origin of a significant part of arsenic known metabolites.