Monomethylarsonous acid (MMA(III)) has been detected for the first time in the urine of some humans exposed to inorganic arsenic in their drinking water. Our experiments have dealt with subjects in Romania who have been exposed to 2.8, 29, 84, or 161 microg of As/L in their drinking water. In the latter two groups, MMA(III) was 11 and 7% of the urinary arsenic while the monomethylarsonic acid (MMA(V)) was 14 and 13%, respectively. Of our 58 subjects, 17% had MMA(III) in their urine. MMA(III) was not found in urine of any members of the group with the lowest level of As exposure. If the lowest-level As exposure group is excluded, 23% of our subjects had MMA(III) in their urine. Our results indicate that (a) future studies concerning urinary arsenic profiles of arsenic-exposed humans must determine MMA(III) concentrations, (b) previous studies of urinary profiles dealing with humans exposed to arsenic need to be re-examined and re-evaluated, and (c) since MMA(III) is more toxic than inorganic arsenite, a re-examination is needed of the two hypotheses which hold that methylation is a detoxication process for inorganic arsenite and that inorganic arsenite is the major cause of the toxicity and carcinogenicity of inorganic arsenic.

The drinking of water containing large amounts of inorganic arsenic is a worldwide major public health problem because of arsenic carcinogenicity. Yet an understanding of the specific mechanism(s) of inorganic arsenic toxicity has been elusive. We have now partially purified the rate-limiting enzyme of inorganic arsenic metabolism, human liver MMA(V) reductase, using ion exchange, molecular exclusion, and hydroxyapatite chromatography. When SDS-beta-mercaptoethanol-PAGE was performed on the most purified fraction, seven protein bands were obtained. Each band was excised from the gel, sequenced by LC-MS/MS and identified according to the SWISS-PROT and TrEMBL Protein Sequence databases. Human liver MMA(V) reductase is 100% identical, over 92% of sequence that we analyzed, with the recently discovered human glutathione-S-transferase Omega class hGSTO 1-1. Recombinant human GSTO1-1 had MMA(V) reductase activity with K(m) and V(max) values comparable to those of human liver MMA(V) reductase. The partially purified human liver MMA(V) reductase had glutathione S-transferase (GST) activity. MMA(V) reductase activity was competitively inhibited by the GST substrate, 1-chloro 2,4-dinitrobenzene and also by the GST inhibitor, deoxycholate. Western blot analysis of the most purified human liver MMA(V) reductase showed one band when probed with hGSTO1-1 antiserum. We propose that MMA(V) reductase and hGSTO 1-1 are identical proteins.

No abstract