Edinburgh summary page
Summary of the arsenic biosensor project
In our initial brainstorming sessions, we considered a variety of projects, including a biosensor for detection of water contamination, and a hybrid biological-electrical device such as a variable resistor. Ultimately we decided to combine these two ideas to develop a whole cell biosensor that responds to arsenic by producing a measurable pH change which can be easily detected with a pH electrode.
Arsenic contamination in drinking water is a serious problem in many parts of the world, and is particularly associated with Bangladesh and West Bengal, where many tube wells were inadvertently drilled through arsenic bearing sediments, resulting in drinking water contaminated with arsenate and arsenite anions (Meharg, 2005; Chowdhury, 2004; Tareq et al, 2003). Consumption of water with elevated arsenic levels over a prolonged period leads to arsenicosis, resulting in skin lesions and various cancers. Many millions of people worldwide are at risk. The current WHO recommended limit for drinking water is 10 ppb arsenic; in many countries a more relaxed limit of 50 ppb is still in operation.
A simple, cheap and sensitive field assay for arsenic levels would therefore be extremely useful. A whole cell microbial biosensor, with an arsenic-responsive pomoter linked to a suitable reporter gene, might be one way of achieving this (Belkin, 2003; Daunert et al, 2000). Arsenic biosensors have been previously reported (for example, Tauriainen et al, 1997; Stocker et al, 2003), but have mainly relied on luminescent or fluorescent reporter genes, which require expensive equipment and trained technicians, and are not really suitable for field use. Other biosensors have used the LacZ/Xgal reporter system, but this is difficult to quantify, and Xgal is expensive and requires refrigeration. By contrast, a sensor giving a pH response would allow a simple quantitative measurement using a cheap pH electrode or solid state device (ISFET), or even just a pH indicator solution giving a colour change.
We devised a system based on the plasmid-encoded arsenic resistance operon. This is controlled by two repressor proteins, ArsR (responding to low concentrations of arsenate or arsenite) and ArsD (responding to higher concentrations)(Wu and Rosen, 1993; Chen and Rosen, 1997). Each is negatively autoregulated. To induce an increase in pH, we chose to use urease, which breaks down urea, (NH2)2CO, to release ammonium ions. This is used in diagnostic microbiology to distinguish urease-positive bacteria such as Proteus, since the pH can rise above 9. To induce a decrease in pH, we chose to use lacZ. This encodes beta-galactosidase, which catalyses the essential first step in the fermentation of lactose to acetic and lactic acids (mixed acid fermentation) in E. coli and related organisms. This reaction is also used in diagnostic microbiology, since the pH can fall below 4.5.
In our design, the activity of the biosensor is initiated by exposure to lactose. Urease is expressed from a hybrid promoter repressed by both lambda cI repressor and LacI repressor. In the presence of lactose, but absence of arsenate, urease is induced and the pH rises. When low amounts of arsenate are present, an ArsR-repressed promoter is induced, leading to expression of lambda cI repressor, switching off urease production. Thus the pH remains neutral. If higher amounts of arsenate are present, lacZ expression is induced through an ArsD-responsive promoter, leading to a fall in pH. By using multiple promoters in this way, a high sensitivity and high dynamic range (the range of arsenate concentrations over which arsenate concentration can be estimated from the reponse) are achieved.
This system was modelled using an ODE-based model, with parameters estimated based on the literature. The model showed good induction of urease and repression of lacZ in the absence of arsenate, and repression of urease and induction of lacZ at high arsenate levels. Sensitivity analysis was also conducted in order to determine which parameters had the greatest effect on the urease and LacZ responses. For example, it was found that the parameter having the greatest effect on the steady-state level of LacZ expression in the presence of arsenate was the degradation rate of ArsD.
Testing the Concept
To demonstrate that a detectable pH change could be achieved in the laboratory, we constructed a biobrick bearing the E. coli chrososomal ars promoter and negatively autoregulated arsR gene (BBa_J33201). The chromosomal ars operon is similar to the plasmid-encoded one we had originally envisaged using, but is controlled solely by ArsR and has no equivalent of the second repressor, ArsD (Diiorio et al, 1995; Cai and DuBow, 1996). We also made a biobrick of the lacZ’ gene encoding the N-terminus of lacZ, which complements the lacZ-delta-M15 mutation found on the chromosome of laboratory strains of E. coli such as JM109 and XL1Blue (BBa_J33202). These were joined to generate BBa_J33203. Unfortunately, we were not able to obtain template DNA for the plasmid encoded arsR and arsD genes we had intended to use within the time frame of the competition. We therefore also cloned the ars promoter and arsR gene from Bacillus subtilis (Sato and Kobayashi, 1998), to test whether this might have a sufficiently different affinity for arsenate to be useful in this context. This was joined to lacZ’ to generate BBa_J33206. In experiments using JM109/pSB1A2-BBa_J33203, concentrations of arsenate as low as 5 ppb gave a significant decrease in pH at incubation times above 5 hours, persisting to over 20 hours, in a non-optimized medium based on LB with 2% w/v lactose. The response was easily detected with a pH electrode and could also be visually assessed using the pH indicator methyl red, which has a pKa around 4.8 to 5. The equivalent system using BBa_J33206 unfortunately did not show a response to arsenate, with even arsenate-free controls giving a rapid drop in pH, suggesting high background activity due to incomplete repression of this promoter in E. coli.
For our system, we also needed a urease part to increase pH in the absence of arsenate. The most obvious choice would have been the urease gene cluster present in some strains of E. coli; however, this consists of around 5 kb of DNA with 7 genes (ureDABCEFG, where ureABC are the genes encoding the urease subunits, and the other genes encode accessory factors required for proper insertion of the nickel cofactor) and contains six forbidden restriction sites which would have to be mutated out individually before the gene cluster could be converted to a biobrick. After searching the literature, we found that the Bacillus subtilis urease gene cluster consists of only three genes, ureABC, which can nevertheless be assembled into a functioning urease in E. coli without the requirement for the usual accessory proteins (Kim et al, 2005). Unfortunately, this gene cluster also contains two forbidden restriction sites, EcoRI and SpeI.
To check that this urease would be suitable for our purposes, the ureABC region was cloned in pGemT-easy (Promega) and pBluescript SK+ (Stratagene). In both constructs, activity was demonstrated in E. coli, with pH rising to 9 after incubation in LB with 0.2% w/v urea. Having decided that this urease would be suitable, we used site-directed mutagenesis to remove the two forbidden restriction sites. This was successfully achieved, but the mutant gene cluster gave no detectable urease activity. Sequencing revealed a possible frameshift mutation in ureC as well as two non-silent single nucleotide changes as compared to the published sequence. Thus we were unable to generate a urease biobrick during the time available.
The final part required for our system was the hybrid promoter repressed by both lambda cI and LacI. This was generated by fusing the PRM-PR region of bacssteriophage lambda, including cI binding sites OR1, OR2 and OR3, to the 3’ end of the lac promoter region including the LacI binding site. The N-terminal region of lacZ was also included, so that lacZ’ expression could be used to test regulation of the promoter. This biobrick was designated BBa_J33205. Unfortunately, we did not have time to build the constructs necessary to test the regulation of this part.
Even though we were not able to build our complete design in the time available, we have demonstrated that a simpler version, E. coli JM109/pSB1A2-BBa_J33203, gives a good pH response to arsenate concentrations as low as 5 ppb arsenic, with a dynamic range in the region of 0 to 20 ppb, in a non-optimized system. This can be detected with a pH electrode or a pH indicator (methyl red) which changes from yellow to red when the pH falls below about 5. Recalling that the WHO limit is 10 ppb, this device is suitable for further development, and could potentially be the basis for a cheap and useful sensor to help prevent the ongoing tragedy of chronic arsenic poisoning. Also, we have submitted the functioning arsenic-responsive promoter to the Registry (BBa_J33201), so we hope that others may be inspired to develop even better arsenic biosensors in the future.
Belkin, S. 2003. Microbial whole-cell sensing systems of environmental pollutants. Current Opinion in Microbiology 6, 206-212.
Cai, J., and DuBow, M.S. 1996. Expression of the Escherichia coli chromosomal ars operon. Canadian Journal of Microbiology 42, 662-671.
Chen, Y.-X., and Rosen, B.P. 1997. Metalloregulatory properties of the ArsD repressor. Journal of Biological Chemistry 272, 14257-14262.
Chowdhury, A.M.R. 2004. Arsenic crisis in Bangladesh. Scientific American (August 2004), 291 (2), 71-75.
Daunert, S., Barret, G., Feliciano, J.S., Shetty, R.S., Shrestha, S., and Smith-Spencer, W. 2000. Genetically engineered whole-cell sensing systems: coupling biological recognition with reporter genes. Chemical Reviews 100, 2705-2738.
Diorio, C., Cai, J., Marmor, J., Shinder, R., and DuBow, M.S. 1995. An Escherichia coli chromosomal ars operon homolog is functional in arsenic detoxification and is conserved in Gram negative bacteria. Journal of Bacteriology 177, 2050-2056.
Kim, J.-K., Mulrooney, S.B., and Hausinger, R.P. 2005. Biosynthesis of active Bacillus subtilis urease in the absence of known urease accessory proteins. Journal of Bacteriology 187, 7150-7154.
Meharg, A. 2005. Venomous Earth: how arsenic caused the world's worst mass poisoning. Macmillan Publishers Ltd.
Sato, T., and Kobayashi, Y. 1998. The ars operon in the skin element of Bacillus subtilis confers resistance to arsenate and arsenite. Journal of Bacteriology 180, 1655-1661.
Stocker, J., Balluch, D., Gsell, M., Harms, H., Feliciano, J., Daunert, S., Malik, K.A., and Van der Meer, J.R. 2003. Development of a set of simple bacterial biosensors for quantitative and rapid measurements of arsenite and arsenate in potable water. Environmental Science and Technology 37, 4734-4750.
Tareq, S.M., Safiullah, S., Anawar, H.M., Rahman, M.M., and Ishizuka, T. 2003. Arsenic pollution in groundwater: a self-organizing complex geochemical process in the deltaic sedimentary environment. Science of the Total Environment 313, 213-226.
Tauriainen, S., Karp, H., Chang, W., and Virta, M. 1997. Recombinant luminescent bacteria for measuring bioavailable arsenite and antimonite. Applied and Environmental Microbiology 63, 4456-4461.
Wu, J.H. and Rosen, B.P. 1993. The arsD gene encodes a second trans-acting regulatory protein of the plasmid-encoded arsenical resistance operon. Molecular Microbiology 8, 615-623.