Background

In many countries, water is a necessity. It's used for drinking, showering, cooking, washing dishes, and doing laundry. Yet as the recent crisis in Flint, Michigan shows, its purity and safety is not guaranteed. In Flint, city officials trying to save money on water temporarily supplied the city with water from the Flint River. The water contained bacteria, so residents were advised to boil the water. However, the chlorine used to treat the bacteria reacted with other compounds in the water to produce total trihalomethanes (TTHM), which could be carcinogenic. And, as news headlines declared, the water contained dangerously high levels of lead because the water corroded the pipelines and allowed lead to seep in. City officials did not proactively treat the water to prevent corrosion. And although residents complained about the water quality, city officials maintained for several months that the water was safe to drink (NPR).

Lead's toxicity is a result of its similarity to minerals our bodies need such as zinc, iron, and calcium. It can enter the body through inhalation, ingestion, or absorption through the skin. Through the pathways of these minerals, lead is distributed throughout the body. Lead is especially dangerous for young children and fetuses, in whom it interferes with development, resulting in symptoms ranging from speech and language problems to decreased bone and muscle growth. Other common sources of lead poisoning in children are lead-based paint, used in old houses, and contaminated soil (KidsHealth). In adults, lead poisoning can lead to high blood pressure and kidney problems (EPA). In children and adults, even a trace amount of lead can cause problems, but the maximum contaminant level (MCL) in drinking water set by the EPA is 15 parts per billion (ppb). In 2015, a year after Flint changed its water source to the Flint River, the a resident's water was found to contain over 13,000 ppb of lead. In the same time period, an increased percentage of children in Flint had elevated blood lead levels (NPR).

The Flint water crisis has sparked national concern about lead contamination in water. Cities are scrambling to test their water lest they become the "next Flint." In the past few months, the city of Pittsburgh has found lead levels that are dangerously close to the EPA's MCL or higher. This awareness is beneficial for the residents, but the power still lies within city officials. Residents need a device to test for lead in their own homes. Current lead sensors for home use exist, but they cost about twenty to thirty US dollars (USD)--inaccessible for over 40 percent of Flint's population and over 14 percent of the US population, which lives under the poverty line (NPR; US Census Bureau). Thus, we are developing Hot Metal Switch, a sensor that could cost mere cents, for residents to test their own water whenever they wish.

Hot Metal Switch

Hot Metal Switch is based on a paper-based sensor developed by the Collins group. In 2014, Pardee and colleagues developed a paper-based detection system that was cheap, easy to use, and efficient, producing results in less than an hour. Their system relied on RNA-based sensors to sense small molecules. These sensors were freeze-dried onto paper with cell extract, which contains the translational and transcriptional machinery a cell uses to express proteins. We are adapting their system to sense metals by using a DNAzyme-based sensor as part of the detection circuit. In the presence of lead, the circuit will express LacZ, producing a visible color change. .

The design of our circuit makes it easily adaptable to other metals. To demostrate this, we used a thallium DNAzyme to sense thallium. Thallium is the element between lead and mercury on the periodic table. It is a byproduct of glassmaking, electronics factories, drug production, and the extraction of metals from ore (EPA). Like its periodic table neighbors, thallium is a heavy metal that is highly toxic to humans--in fact, it is more potent than lead. The EPA’s MCL for for thallium is only 2 ppb. Thallium is also known as the “poisoner’s poison” because it takes several days to kill, is odorless, and is tasteless. Thallium owes much of its toxicity to its size, which is similar to that of potassium, so thallium can easily enter cells via potassium pathways (RSC). In low doses, thallium causes hair loss and problems with the kidney, liver, and intestines, which can lead to vomiting and diarrhea (ATSDR, EPA).

References

National Public Radio (NPR). Lead-Laced Water In Flint: A Step-By-Step Look At The Makings Of A Crisis. 2016, April 20.
KidsHealth from Neymour. Lead Poisoning. 2016.
Environmental Protection Agency (EPA). Table of Regulated Drinking Water Contaminants. 2016, July 15.
United States Census Bureau. Income and Poverty in the United States: 2014. 2015, September.
Royal Society of Chemistry (RSC). Chemistry in Its Element - Thallium.
Agency for Toxic Substances & Disease Registry (ATSDR). Toxicological Profile for Thallium. 2015, January 21.
National Environmental Methods Index (NEMI).
Back to Top

In Vitro Protein Synthesis

We worked with our genetic circuit in vitro for two reasons. First, the paper-based sensor developed by the Collins group necessitates that the circuit works in cell-free extract. Second, and more important, protein synthesis in vitro bypasses the fickle cell. With a cell-free extract, we can add as many plasmids as we need to complete our detection circuit. Some of our constructs are also toxic to cells. Cell-free systems provide much more flexibility for the development of our sensor. We synthesized proteins in vitro with PURExpress cell-free extract from New England BioLabs. PURExpress is a T7-mediated E. coli system.

Genetic Circuit

As described in the introductory section, the genetic circuit contains four components. When the target metal is present, a metal-specific DNAzyme is activated, resulting in cleavage of a second DNA strand: To bypass inefficient sequestration of the substrate strand by the catalytic strand, we developed hairpin DNAzymes as illustrated above. Read more about the DNAzymes below.

The cleaved DNA strand activates a toehold switch. which mediates expression of T3 RNA polymerase ("Ribo" is a ribosome): Toehold switches are generally RNA sequences that hide the ribosome binding site in a hairpin loop, thus preventing translation. Read more about the toehold switch below.

The T3 RNA polymerase transcribes two genes. The expression of lacZ results in a color change, which alerts the user that the target metal is present. In our system, LacZ converts the yellow substrate chlorophenol red-β-D-galactopyranoside into the purple β-galactosidase. Read more about the reporter below.
The expression of T3 RNA polymerase serves to amplify the original signal from the DNAzyme activity. The concentration of metal ion can be quite low, so an amplification system will ensure a visible signal from minimal DNAzyme cleavage. Read more about the amplification system below.

DNAzyme

DNAzymes, or deoxyribozymes, are short DNA strands which perform a specific chemical reaction, generally catalytic. They are analogous to protein enzymes or ribozymes. DNAzymes contain a catalytic domain and binding domains, and commonly require metal cofactors to be activated. This activation could either be a matter of directly facilitating the reaction, or being a part of the proper structural integrity of the DNAzyme. Most available DNAzymes catalyze the cleavage of strands containing RNA bases at the active site. No DNAzymes are known to exist in nature, and are therefore a fully synthetic construction (Silverman). We worked with two DNAzymes. The thallium-specific DNAzyme was based off the work of Huang, Vazin, and Liu. The lead-specific DNAzyme was based off the work of Lan, Furuya, and Lu.

The structure of the DNAzymes was created by combining aspects of the Collins toehold circuits, specifically D and G, and existing metalloDNAzymes. MetalloDNAzymes split their substrate at a specific sequence, releasing the products into solution (if properly designed), given the presence of the applicable cofactor(s). The Collins trigger sequences turn the Collins reporter sequence on by interacting with the toehold sequence ahead of the reporter (in this case, the LacZ sequence). The point at which the substrate splits is controlled; in the case of the thallium metalloDNAzyme, the substrate is split at the sequence rA*G, where rA is a RNA base, * is representative of a phosphothioate bond, and G is a normal DNA base. For the lead DNAzyme, it simply occurs at the sequence rAG.

Taking the initial structure of the DNAzyme and substrate, the substrate is first altered such that the split will create a piece which has the trigger sequence of the toehold circuit. Additionally, the trigger sequence is placed on the 5’ end of the altered substrate, so that the piece of the trigger sequence which initially bonds to the toehold sequence is bonded to the DNAzyme. This measure is to prevent the trigger sequence from activating the toehold sequence without the presence of the metallic cofactors. The DNAzyme is then altered such that both arms (the areas to the left and right of the catalytic loop) will bond to the appropriate places of the substrate.

Toehold Switch

Toehold switches are RNA sequences that regulate protein expression by hiding the ribosome binding site (RBS) in a hairpin loop. They require a trigger sequence to unfold, expose the RBS, and begin translation. Here is a video illustrating the mechanism of a toehold switch:

The Collins circuit utilizes a toehold switch to activate the lacZ (β-galactosidase) gene, which then is used to cleave a specialized substrate which turns purple (absorbance = 570 nm). The original Collins circuit utilized a RNA trigger and RNA toehold. Our modified circuit utilizes a DNA trigger sequence with an RNA toehold.

Reporter

Hot Metal Switch signals the presence of metal ion with a color change mediated by LacZ. The substrate chlorophenol red-β-D-galactopyranoside is added to the cell-free extract so that LacZ can produce β-galactosidase as soon as it's made. Because LacZ is an enzyme, the degree of color change is limited by the amount of substrate available, not the amount of enzyme produced, which is advantageous for our in vitro system. Furthermore, because the Collins toehold switches already contained lacZ, it was easier to test the rest of the system using LacZ.

We decided to use a colorimetric signal to make the sensor amenable to home use. Colors, unlike fluorescence, can be detected without additional equipment. Although an abundant amount of GFP becomes visible to the naked eye, GFP's strength lies in its fluorescence, not its color. Additionally, a cell-free system may not be able to produce enough GFP to be visible within a reasonable amount of time. Our first choice was to use the chromoprotein amilCP from the iGEM distribution kit, developed by the 2012 Uppsala Sweden team. However, sequences of the assembled plasmid suggested we had CFP, not amilCP. No color was observed in vitro or in the cells.

Amplifier

The amplification system was inspired by the work of Wang, Barahona, and Buck. In our system, the toehold switch produces T3 RNA polymerase in the presence of the target metal. T3 then translates both LacZ and itself to amplify the signal. The self-propagating amplifier decreases the detection limit of the system.