Riceguard: Protecting the World’s Staple from Invisible Threats

Current Solutions

---

Right now, detecting arsenic in rice requires specialized laboratory equipment. The gold standard is ICP-MS (Inductively Coupled Plasma Mass Spectrometry), a powerful tool capable of detecting trace amounts of heavy metals. While accurate, these machines are expensive, slow, and only available in research labs or government facilities. Farmers, families, and local markets rarely have access to them.

This means detection is limited, uneven, and often delayed. By the time contaminated rice is discovered, it may already be in homes, stores, or baby food products. Current solutions focus on centralized testing and regulations, leaving everyday people without the ability to check the safety of the food on their own table.

Our Project

---

Riceguard aims to close that gap by bringing detection out of the lab and into people’s hands. Instead of relying on bulky equipment, we use a cell-free system built with NEBExpress plasmids. This system avoids the complexity of living cells, works in a transcription-only environment, and allows us to quickly prototype and refine our biosensor.

At the center of Riceguard is a two-plasmid design. The first plasmid produces both ArsR and ArsC. ArsR is a repressor that binds to our second plasmid's promoter (Pars_OC2). ArsC is an arsenic reductase, an enzyme that transforms inorganic Arsenic (V) into Arsenic (III). In the presence of these contaminants, The ArsR repressor unbinds and allows T7 Polymerase to transcribe F30 Broccoli, an RNA aptamer, which binds to our fluorophore (DHFBI-1T).

To make this technology practical, we aim to freeze-dry (lyophilize) the system onto paper strips. Users only need to add a few drops of rice water extract to activate it. The excitation light our kit uses is a LED light, which will excite our RNA-dye complex. Our kit also comes with a small strip of plastic filters users can put on their camera lenses to eliminate unwanted emission. For even more precision, we’re developing a mobile app that captures the fluorescence and quantifies arsenic levels digitally. 

Riceguard transforms detection from an expensive, centralized process into something simple, affordable, and portable. Our goal is to give families and farmers a way to check food safety themselves anytime, anywhere.

Theory

Strategic Pivot to Cell-Free Synthesis

---

This necessitated an immediate pivot. We shifted to cell-free synthesis, utilizing the NEBExpress kit with T7 polymerase and no living cells. This method is safe, portable, and free of regulatory constraints.

Yet, our biosensor was not initially compatible with T7 systems, as our ArsR promoter was bacterial rather than phage-driven. The solution emerged from Chen et al. (2022), who described novel arsenic-inducible OC2 and OC3 promoters. We incorporated the optimized

(Chen, S.-Y., Zhang, Y., Li, R., Wang, B., & Ye, B.-C. (2022). De novo design of the ArsR regulated Pars promoter enables a highly sensitive whole-cell biosensor for arsenic contamination. Analytical Chemistry, 94(17), 7210–7218. © 2022 The Authors. Published by American Chemical Society). ParsOC2 promoter into our arsenic biosensor design, enhancing sensitivity for the NEBExpress system.

We fused these to T7 constructs creating hybrid systems T7-ParsOC2 with optimized spacers, ArsR/ArsC downstream, and Broccoli aptamer on a separate plasmid. Three sense variants and two reporters were developed.

(Senda, N., Enomoto, T., Kihara, K., Yamashiro, N., Takagi, N., Kiga, D., & Nishida, H. (2022). Development of an expression-tunable multiple protein synthesis system in cell-free reactions using T7-promoter-variant series. Synthetic Biology, 7(1), 1–8. © 2022 ) We decided on two separate plasmids in our biosensor. A sense plasmid to produce ArsR repressor and ArsC reductase. We chose the variant above as the best of the variants we tested. The Reporter Plasmid variant produces an mRNA aptamer called F-30 Brocolli, which binds with a flourophoroe called DHFBI-1T. Since it only requires transcription, fluorescence happens very quickly, ideal for a lyophilized biosensor.

The summer laboratory phase presented significant challenges: plasmids were cloned in-house, minipreps were conducted overnight, and the first faint green fluorescence was observed at 1 a.m., marking a critical milestone. Validation data revealed that OC2 outperformed OC3 in repression efficiency, leading to the exclusive adoption of OC2. The sensitivity optimization phase required six cycles and consumed five kits, with tests conducted at 25°C and 37°C using staggered and simultaneous additions. Persistent leakiness was observed, with 44,000 RFU at 0 ppb and no signal at 50 ppb, prompting adjustments to ratios, temperatures, and dye considerations. New plasmids were reordered, and fresh bacterial cultures were prepared to determine if the issue stemmed from the plasmids. Despite these obstacles, the wet-lab team gained substantial experience in synthetic biology techniques, fostering resilience, character, and camaraderie.

Success was achieved on October 2. The optimized master mix comprised lysate, T7 polymerase, both plasmids added simultaneously, 20 µM DFHBI-1T (final concentration 3.12 µM), 12.5 µL per well, and 1 µL of rice extract or arsenic solution. Measurements were conducted using an ELISA plate reader at 37°C without shaking or sealing, capturing raw kinetics. The fluorescence exhibited a rapid initial increase—a transcription burst—followed by a crash at zero ppb as ArsR bound the promoter, causing the dye to unbind. At 5 to 100 ppb, the rise persisted longer; at 50 ppb, fluorescence was approximately 1.5-fold above zero, and at 100 ppb, it reached about 2-fold, though it stalled after 20 minutes due to an arsenic toxicity ceiling. Above 100 ppb, the signal diminished. The dynamic range of 5–100 ppb aligns perfectly with rice contamination levels reported by HBBF (0.1–0.4 µg/g), detectable at a 100-fold dilution.

Sensitivity testing for (0, 5, 10, 25, 50, 100, 200, 500) in parts per billion As(III), rice extract testing on 2 different rice samples from Thailand (jasmine & wild rice varieties), interferent testing on Cu (100ppb), Pb(50ppb) and Fe(200ppb). *We omitted Cd and Hg due to the extremely high toxicities involved in experimenting and disposal of these contaminants.

For rice extract compatibility, we ground 1 g of rice in 10 mL of water, centrifuged it for 3 minutes at 5,000 rpm, and used 1 µL of supernatant. Samples E1–E3 and E4–E6 produced glows equivalent to 5–10 ppb arsenic, confirming real contamination without artificial spiking—a practical approach, as farmers would naturally grind their own dinner. The system proved effective. Specificity testing with interferents (Fe, Cu, Pb)) showed flat responses, contributing less than 5% of the signal from 50 ppb arsenic, indicating clean specificity.