![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20081107120912im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Final Report: Metal Biosensors: Development and Environmental Testing
EPA Grant Number: R830907Title: Metal Biosensors: Development and Environmental Testing
Investigators: Anderson, Anne J. , McLean, Joan , Miller, Charles D.
Institution: Utah State University
EPA Project Officer: Savage, Nora
Project Period: May 1, 2003 through April 30, 2006 (Extended to April 30, 2007)
Project Amount: $336,000
RFA: Environmental Futures Research in Nanoscale Science Engineering and Technology (2002)
Research Category: Nanotechnology
Description:
Objective:Our research explored the development and use of bacterial biosensors for the rapid and sensitive detection of Cu and Cd in water. Our premise was that we could build biosensors based on the fusion of Cu- or Cd-specific promoters with a marker gene to enable rapid evaluation of changes in RNA expression upon metal exposure. We selected a luxAB cassette as the reporter, anticipating that the metals would stimulate expression because they would cause oxidative stress and consequently increase in light emission from cells. We used Pseudomonas putida strain KT2440 as the biosensor because this isolate was obtained from contaminated soils and has a completely sequenced genome.
Summary/Accomplishments (Outputs/Outcomes):Proteomic and Transcriptional Findings
P. putida cells responded differentially to Cd and Cu exposures at the proteomic and transcriptome levels. The cells displayed different stress responses that correlated with a more intense oxidative stress imposed by Cd than by Cu.
The proteomic studies involved two-dimensional (2-D) gel electrophoresis of extracts of cells exposed in minimal medium for 6 hours and demonstrated that some peptides increased in accumulation in response to both metals, whereas others were metal-specific (Figure 1).
Figure 1. Representation of Proteome Changes in Response to Cu or Cd Treatments (10 mg/L) in Minimal Medium for 6 Hours
We performed studies to determine transcript levels of the genes encoding the responsive proteins to determine whether the increases were due to transcriptional changes. For most of the 26 genes examined, transcript abundance was enhanced by 1 and/or 4 hours of exposure to 10 mg/L of metal in minimal medium, and the specificity of response was the same as that observed by the proteomics study. This finding suggests that a sensitive and specific detection system could be achieved using a gene chip array with selected responsive genes to detect altered gene expression in exposed P. putida cells.
Analysis by Geochem modeling showed that, in minimal medium, most of the Cd or Cu would exist as citrate, phosphate, and sulfate complexes (Table 1).
Table 1. Effect of the Composition of Minimal Medium on Complexation of Cu and Cd Ions
Cadmium concentration |
0.1 mg/L |
1 mg/L |
10 mg/L |
|||
Name |
Conc. M |
% Total |
Conc. M |
% Total |
Conc. M |
% Total |
Cd(2+) |
2.27E-08 |
2.6 |
2.47E-06 |
27.80 |
3.15E-06 |
3.50 |
CdSO4 (aq) |
2.23E-08 |
2.5 |
3.07E-07 |
3.50 |
3.92E-07 |
0.00 |
Cd[Citrate]2 (-4) |
2.70E-09 |
0 |
3.24E-07 |
3.60 |
4.12E-07 |
0.00 |
Cd[Citrate] (-1) |
8.39E-07 |
94.2 |
5.76E-06 |
64.70 |
7.34E-06 |
8.20 |
Cd3(PO4)2 |
|
|
|
|
2.59E-05 |
87.30 |
|
|
|
|
|
|
|
Copper concentration |
0.1 mg/L |
1 mg/L |
10 mg/L |
|||
Name |
Conc. M |
% Total |
Conc. M |
% Total |
Conc. M |
% Total |
Cu(2+) |
1.08E-10 |
0.00 |
1.52E-08 |
0.00 |
1.21E-07 |
0.00 |
Cu[Citrate] (-1) |
1.56E-06 |
99.10 |
1.37E-05 |
87.00 |
1.01E-04 |
64.30 |
Cu2[Citrate]2 (-2) |
1.39E-10 |
0.00 |
2.01E-08 |
0.00 |
1.10E-06 |
1.40 |
Cu[Citrate]2 (-4) |
1.29E-08 |
0.00 |
1.96E-06 |
12.50 |
1.35E-05 |
8.60 |
Cu3(PO4)2 |
|
|
|
|
1.34E-05 |
25.60 |
Thus, the cells showing the changes in transcription and protein accumulation occurred under conditions where the metals were not present as free ions but as complexes. This situation related directly to the problem for the U.S. Environmental Protection Agency (EPA) of how to assess the risk levels of Cd and Cu in aqueous solutions from the environment. The fact that complexes were bioavailable to the P. putida cell conflicts with the Biotic Ligand Model now used in risk assessment of Cu; the Biotic Ligand Model is built on the premise that only free ions would be perceived and active within cells.
Luminescent Biosensor Studies
To explore bioavailability further, we utilized biosensors constructed with fusions of promoters with the luxAB cassette to endow light production when the promoter was activated. One set of fusions were constructed with promoters of genes involved in protection against activated oxygen stress, because oxidative stress is one of the consequences of Cu and Cd exposures (see Figure 1).
Dose-dependent decreases in Lux output, compared with nonexposed control cells, was observed with the exception of Cu-treatment at 0.01 mg/L of the FeSOD mutant (Figure 2). For Cu, decreased Lux at 10 mg/L corresponded to cell death, but with Cd, the cells only underwent a temporary stasis in cell growth at this exposure dose.
Figure 2A. Cd response
Figure 2B. Cu response
Figure 2. Effect on Light Output (Lux) of Exposure to Cd (A) and Cu (B) of Logarithmic Phase Cells with Promoter Fusions With the bfr Gene (Encoding an Iron Binding Protein), catA (Encoding a Major Catalase Gene), and sodB (Encoding the Major Fe Superoxide Dismutase)
To derive biosensors with more specific and sensitive responses, we generated fusions with the promoters identified as being metal responsive in our RNA assays with isolate KT2440. We housed these fusions in a stable plasmid rather than creating a knockout mutant as in the cells used in Figure 2. Representative data for these biosensors are shown in Figure 3. Unexpectedly, we found that the constructs showed no little specificity in response to Cu or Cd; rather a rapid decline in Lux (instead of the anticipated increase) was observed. The promoter in construct Pp_0588 was for a Cu-specific gene, yet both Cu and Cd decreased Lux output to the same extent as a fusion with a nonmetal responsive promoter (data not shown). We postulate that the metal ion exposure reduced the supply of FMNH2 so that the luciferase could not function. This is an intriguing finding because similar constructs in Escherichia coli function well as biosensors, demonstrating increased promoter activity. We suspect that there may be fundamental differences between the metabolism of these two cell types.
Figure 3. Response of Promoter Fusion in P. putida KT2440 to Cu and Cd Free Ions
Although the biosensors lacked specificity, we have used them to examine the effect of competing metal background ions and complexation.
- We find that a background of K+ does not affect the metal response, but that Ca2+ protects cells more against Cd than Cu exposure. This finding suggests that modification of risk assessments for hardness factors, as presently in place for Cu evaluation, may have to be tailored to the metal.
- We also see that complexation with sulfate and citrate for Cu and Cd exposure or phosphate and chloride for Cd exposure reduced the toxicity, but that each of these complexes was bioactive. This finding for pseudomonad cells questions how Biotic Ligand modeling should be involved in risk assessment for heavy metals.
Cell Sorbtion Studies
The relevance of the Biotic Ligand Model to interactions of the pseudomonad cells with Cu and Cd was examined directly using chemical methods to assess how the metal was associated with the cell (at exchangeable surface sites, at nonexchangeable sites [periplasmic] and internal). As shown in Figure 4, partitioning between Cd and Cu was different with Cu mainly being held in the periplasm and Cd at exchangeable surface sites. These findings agree with the ability of Ca to protect KT2440 cells from Cd2+ but not Cu2+.
Figure 4. Partitioning of Cu and Cd Within P. putida Cells
Testing the experimental sorbtion data with the Langmuir isotherm showed that free metal exposure correlated well with the Biotic Ligand Model. However, when the same analyses were performed with the data from use of the metal complexes, none of the associations could be well explained by the Biotic Ligand Model. Increases in internal levels of Cd were especially notable for the citrate complexes. These observations demonstrate with a fourth technique that heavy metal complexes are bioavailable to the pseudomonad cells. The findings also emphasize how the Biotic Ligand Model should be used in the EPA assessment of Cu and Cd risk levels.
Conclusions:Our goal of generating specific and sensitive biosensors to assess Cu and Cd in waters was not successful. The promoter fusions in P. putida strain KT2440 responded to higher concentrations of Cu and Cd (mg/L rather than the μg/L levels associated with risk thresholds). Additionally, the metals unexpectedly had a nonspecific toxicity in the biosensor pseudomonad host cell. Further work could explore the use of a different marker gene or expression in another host bacterium. Alternatively, a gene chip array for specific detection could be developed based on the genes that show differential transcription upon Cu- and Cd-exposure of strain KT2440. We have demonstrated that these transcriptional changes are robust in a mixed inorganic background.
Bioavailability of both inorganic and organic complexes of Cu and Cd to the soil bacterium P. putida KT2440 was demonstrated at the proteomic and transcriptome levels and with use of the promoter-fusion biosensors. The findings call into question whether the Biotic Ligand Model should be used in adjusting the levels of Cu or Cd that are considered bioavailable in risk assessment. The studies also demonstrate that the use of hardness factors in assessment should be considered differentially for each metal.
The bioavailability studies for the bacteria raise the possibility that life forms at many different levels should be studied to understand better risk assessment of heavy metals. For instance, the increased internal levels of Cd seen after exposure to Cd-citrate in the pseudomonad cells could have an impact on life forms higher in the food chain.
In summer 2007, Mindy Pabst completed a BS/ME degree based on the biosensor and cell binding work, and she has presented posters on her findings.
The biosensors have been used in training programs for several undergraduate and high school students during the summer months each year. The sensors are being used in the laboratory portion of a new biosensor/biophotonics course held in the last two years at Utah State University (USU).
Three papers are in preparation: (1) a Lux biosensor paper, (2) the differential partitioning of free ions of Cd and Cu into pseudomonad cells, and (3) comparisons of fit of the Biotic Ligand Model for the interactions of pseudomonads with free and complexed Cd and Cu).
Journal Articles:No journal articles submitted with this report: View all 8 publications for this project
Supplemental Keywords:, Ecosystem Protection/Environmental Exposure & Risk, Toxics, POLLUTANTS/TOXICS, Water, Sustainable Industry/Business, Scientific Discipline, RFA, Engineering, Chemistry, & Physics, Chemicals, Environmental Engineering, Environmental Chemistry, New/Innovative technologies, 33/50, Monitoring/Modeling, Environmental Monitoring, heavy metals, nanotechnology, nanosensors, chromium & chromium compounds, cadmium & cadmium compounds, field detection, metal biosensors, cadmium, biosensors, bioengineering, bacterial sensors, nanoengineering, environmental measurement
Progress and Final Reports:
2003 Progress Report
2004 Progress Report
Original Abstract