Development of an arsenic biosensor for testing arsenic contanimanation in drinking water
In Bangladesh, more than 25 million of people are drinking arsenic contaminated groundwater, what causes dreadful impacts. The limit for As-contamination in drinking water in Bangladesh is 50 μg/1, but As-concentrations up to 2.4 mg/I can occur. Until now, no sensitive enough chemical field test exists. The work presented here was aimed at developing a bacterial biosensor to measure arsenic contamination in groundwater. For proper applicability, the test should fullfil several demands. I) It needs to be sensitive enough and reliable. ii) The costs should be low. iii) The test should be so simple, that it can be applied by people, who don't have a laboratory education.
The bacterial detecting system was based on the naturally widely occuring arsenic resistance genes. This ars operon consists of five genes, arsR, arsD, arsA, arsB and arsC. The system is very sensitive and selective to arsenite and antimonite. The plasmids used in this study contained the arsR gene, one of the regulatory genes of the ars operon, coupled to a reporter genefor building the detection system. Three different Escherichia coli strains (DH5α, JM109, AWlO) were transformed with these arsR-reporter plasmids. To find the simplest and best test protocol, beta-galactosidase, luciferase and GFP were tested as reporter proteins.
The sensitivity to arsenite, arsenate and antimonite were measured in liquid solutions with betagalactosidase as reporter protein (plasmid pBGD23) and could be detected in a range between 0.01and10 μM for antimonite and between 0.9 and 9 μM for arsenite and arsenate, respectively. The used substrate was ONPG. With another substrate, X-Gal, no differences between samples with arsenite and samples without arsenite could be detected, due to the background expression of beta-galactosidase. ArsR was cloned into another plasmid (pJAMA8arsR) with the luxA and luxB genes. The luciferase expression was inducible with arsenite in a range between 0.09 and 9 μM. Another reporter protein, GFP (from plasmid pProbe-gfp(L V A)arsR), was also tried out. The arsenite inducible fluorescence was clearly less sensitive. Arsenite was detectable in a range between 9 and 90 μM.
To store the bacteria, freeze-drying techniques were applied. The number of viable cells was stable over 90 days. The bacteria could be reconstituted with water only. After an induction time of one hour with arsenite, a clear response could already be detected. This storage method can also be applied under the conditions in Bangladesh, with high temperatures and high air moisture.
The reporter systems could be applied with calibration standards for small laboratories in the field in Bangladesh. A more simple colorimetric test is not ready at the moment, but with further experiments like cloning other reporter proteins (i.e. catechol 2,3-dioxygenase) or reducing betagalactosidase background expression, a colorimetric test assay could be reached.