We have developed a detection system which uses the colorimetric properties of gold nanoparticles(AuNPs) as well as their interactions with ssDNA aptamers to simply and rapidly detect the concentration of microcystin-LR (MC-LR) with high sensitivity and specificity. Unlike other microcystin detection methods like enzyme-linked immunosorbent assay (ELISA) and high-performance liquid chromatography (HPLC), this system not only has a quick turnaround time, but also requires minimal specialized skill and equipment [1]. Accurate microcystin concentrations can be determined in two hours with little more than a spectrophotometer.
The MC-LR specific aptamer used for the system is a 60 nucleotide long single stranded DNA aptamer (BBa_k2960000). These were conjugated with 10 nm AuNPs suspended in a citrate buffer. Citrate is a negatively charged ion that prevents the gold nanoparticles from aggregating into larger particles.
Normally, gold nanoparticles with a diameter on the order of 10 nm are a deep red color; however, when exposed to an excess salt solution, the ions reduce the charge shielding that the citrate buffer provides and thus allows the AuNPs to aggregate into larger particles [3] whose light absorption properties are different due to surface plasmon resonance [2].
In our system, we use sodium chloride with our DNA aptamer and AuNPs as was demonstrated by Li et al. [6]. When AuNPs are mixed with sodium chloride and aptamer, the negatively charged ssDNA aptamer binds to the AuNPs and provides charge shielding and prevents aggregation, preserving their red color. If there is microcystin present, the DNA aptamer will specifically prefer to bind to the microcystin, leaving the AuNP’s exposed to salt and allowing them to aggregate and change color.
As shown by Li et al., the color change has a linear relationship with the concentration of microcystin-LR toxin (MC-LR) [6]. Our aptamer preferentially binds to MC-LR, so when a solution of MC-LR is introduced into the aptamer and AuNP solution, MC-LR outcompetes the AuNPs for binding to the aptamer. This removes the protection from the AuNPs, allowing them to aggregate. Unlike the rightward shift in peak absorbance described by Li et al., we observed instead a decrease in the absorbance for higher concentrations of microcystin, with little to no change in peak absorbance [6]. This may be because higher concentrations of microcystin increase AuNP aggregation and make them absorb less high energy wavelengths, as the solution turns violet. We plotted a standard curve, which also shows a linear relationship (Figure 1). We determined that the ideal wavelength at which to measure microcystin concentration is 520 nm (peak absorbance) since the differences at this wavelength showed the most linear relationship with microcystin concentration.
This system has a quick turnaround time, and microcystin concentration can be determined in about 2 hours. The nanoparticles incubate with the salt and aptamer for 15 minutes and then it all incubates with a sample for at least two hours, all at room temperature. Longer incubation times were tested and these tests showed that the steady state absorbance is reached by 2 hours of incubation, as shown in Figure 2.
To test the specificity of the AuNP-Aptamer system, we analyzed tap water from Ithaca, New York, which is hard water that contains many ions and minerals, but no detectable levels of microcystin. Figure 3 shows that there is no difference in the system between very pure nuclease free water and tap water. This suggests that the specificity of our system for detecting MC-LR concentration is high, as other ions and minerals appear to have no effect on the affinity of the aptamer for the AuNPs. It also suggests that small amounts of other ions and minerals have little to no effect on the aggregation of the AuNPs.
In order to maximize the sensitivity of the reHAB detection system, salt and aptamer concentrations were optimized for the range of MC-LR concentrations that may be present in harmful algal blooms, being between 0 nM and 1 µM [4,5]. The optimal aptamer and salt conditions were observed to be 2 µM and .5 M, respectively, as shown in Figures 4 and 5.
As observable in Figure 4, as salt concentration decreases, the difference in the magnitude of the peak absorbance at 520 nm from adding MC-LR increases. This means that it appears as if our detection system is the most sensitive at low salt concentrations. It is worth mentioning that 1M, 2M, 3M, and 4M salt concentrations were tested together at 1 µM MC-LR, whereas .5M and 5M salt concentrations were tested separately from these other four at an MC-LR concentration of 0.1 µM. Following the trend observed above, one would expect that 0.5M NaCl produces the best sensitivity. It is clearly more sensitive than 5M NaCl. In comparing the sensitivity of our system with 0.5M NaCl with the concentrations of 1, 2, 3, and 4M NaCl, we can use some of the data from our standard curve to justify our pick of 0.5M NaCl as the most sensitive. According to our graph, the absorbance difference at 520 nm between 0 nM MC-LR and 0.1 µM for 0.5 M NaCl is 0.0246 OD. Because we have demonstrated that higher MC-LR concentration causes a more pronounced depression in our absorbance peak at 520 nm, it is reasonable to extrapolate that at MC-LR concentration of 1 µM will cause a much larger depression than the previously observed 0.1 µM. Therefore we decided on using 0.5 M NaCl.
It was determined that the aptamer concentration made no consistent difference in detecting microcystins. Figure 5 is a series of graphs, each of which represent measurements for a different concentration of DNA aptamer: 2 µM, 20 µM, and 200 µM. Each demonstrate the absorbance values observed across a range of MC-LR concentrations. Note that the y-axis contains relative units which describe the difference in absorbance values before and after adding a sample of MC-LR. Absorbance was measured across a 300-700 nm spectrum. There is a lack of consistent differences before and after adding microcystin due to increased aptamer concentration, so a lower concentration of 2 μM is used in our system.
The bioreactor is composed of immobilized E. coli which are enabled to express the mlr cassette into their periplasm. Immobilization was achieved by adapting a technique to encapsulate cells within millimeter-sized alginate beads [7]. The immobilized cells were then packed into a cylindrical bioreactor and lake water was passed through. As water passes through the bioreactor, microcystins diffuse into the beads and then into the E.coli periplasm where it is then degraded by the mlr enzymes.
To ensure that encapsulation was achieved, and that E. coli was capable of surviving within the alginate beads over an extended period of time, the beads were placed in a water bath with comparable water temperature for cyanobacterial growth (27.5°C) [8]. Every day a select number of beads were dissolved in sodium citrate, creating a solution of bacteria. This solution was then plated. Presence of colonies on the plates indicated survival of the E. Coli and successful encapsulation. To further test viability, wet cultures of the beads were successfully made by piercing the bead and placing the sample into LB.
To test the efficacy of the bioreactor, water treated with a known concentration of MC-LR was then passed through a bioreactor model packed with beads containing strains expressing either either mlrA (BBa_k2960012) or mlrA with translocase tag (BBa_k2960001), and the outflow then underwent aptamer testing. The results of the treated sample were compared to the initial results. This test was then replicated multiple times (n=3) to validate the bioreactor’s efficacy. To estimate the microcystinase activity, we compared the MC-LR concentration of the outflow to that of a control reaction, in which the same solution was passed through the empty chamber.
As compared to the control, the outflow from the bioreactor packed with beads containing mlrA has lower relative MC-LR concentration. Relative concentration was determined by normalizing the test (mlrA) absorbance values in the outflow by those of the control and then using the equation given by the standard curve to calculate corresponding relative concentration.
[1] Detection Methods for Cyanotoxins. (2019, August 12). Retrieved from https://www.epa.gov/ground-water-and-drinking-water/detection-methods-cyanotoxins.
[2] Gold Nanoparticle Properties. (n.d.). Retrieved October 19, 2019, from http://www.cytodiagnostics.com/store/pc/viewcontent.asp?idpage=2.
[3] Pamies, R., Cifre, J. G. H., Espín, V. F., Collado-González, M., Baños, F. G. D., & Torre, J. G. D. L. (2014). Aggregation behaviour of gold nanoparticles in saline aqueous media. Journal of Nanoparticle Research, 16(4). doi: 10.1007/s11051-014-2376-4
[4] Turner, A., Dhanji-Rapkova, M., O’Neill, A., Coates, L., Lewis, A., & Lewis, K. (2018). Analysis of Microcystins in Cyanobacterial Blooms from Freshwater Bodies in England. Toxins, 10(1), 39. doi: 10.3390/toxins10010039
[5] US Department of Commerce, Noaa, Great Lakes Environmental Research Laboratory, & Institute for Limnology and Ecosystems Research. (n.d.). Microcystin Guidelines. Retrieved October 19, 2019, from https://www.glerl.noaa.gov/res/HABs_and_Hypoxia/microcystinGuidelines.html.
[6] Li, X., Cheng, R., Shi, H., Tang, B., Xiao, H., & Zhao, G. (2016). A simple highly sensitive and selective aptamer-based colorimetric sensor for environmental toxins microcystin-LR in water samples. Journal of Hazardous Materials, 304, 474–480. doi: 10.1016/j.jhazmat.2015.11.016
[7] Dziga, D., Sworzen, M., Wladyka, B., & Wasylewski, M. (2013). Genetically Engineered Bacteria Immobilized in Alginate as an Option of Cyanotoxins Removal. International Journal of Environmental Science and Development, 360–364. https://doi.org/10.7763/IJESD.2013.V4.371
[8] Jiaqi You, Kevin Mallery, Jiarong Hong, Miki Hondzo, Temperature effects on growth and buoyancy of Microcystis aeruginosa, Journal of Plankton Research, Volume 40, Issue 1, January-February 2018, Pages 16–28, https://doi.org/10.1093/plankt/fbx059