An autonomous microbial sensor enables long-term detection of TNT explosive in natural soil

Abstract

Microbes can be engineered to sense target chemicals for environmental and geospatial detection. However, when engineered microbes operate in real-world environments, it remains unclear how competition with natural microbes affect their performance over long time periods. Here, we engineer sensors and memory-storing genetic circuits inside the soil bacterium Bacillus subtilis to sense the TNT explosive and maintain a long-term response, using predictive models to design riboswitch sensors, tune transcription rates, and improve the genetic circuit's dynamic range. We characterize the autonomous microbial sensor's ability to detect TNT in a natural soil system, measuring single-cell and population-level behavior over a 28-day period. The autonomous microbial sensor activates its response by 14-fold when exposed to low TNT concentrations and maintains stable activation for over 21 days, exhibiting exponential decay dynamics at the population-level with a half-life of about 5 days. Overall, we show that autonomous microbial sensors can carry out long-term detection of an important chemical in natural soil with competitive growth dynamics serving as additional biocontainment.

Fig. 1. A synthetic genetic circuit for detecting TNT in soil.

A Engineered Bacillus subtilis cells sense TNT inside the soil in competition with natural microbes, using optical or olfactory output signals to transmit response information for stand-off detection. B The engineered synthetic genetic circuit contains cell sensing, memory, and response modules. A constitutive promoter transcribes a TNT-sensing riboswitch, which activates the translation of a site-specific integrase. An antisense promoter reduces sensor leakiness via transcriptional interference. When expressed, the integrase binds to the attB/attP sites and flips the orientation of a promoter region. When flipped, the promoter expresses the output response module. Modules are insulated using transcriptional terminators.

Fig. 2. Design and characterization of a TNT-sensing riboswitch.

A The Riboswitch Calculator model predicted mRNA structures of the RS14 riboswitch before and after induction with 35 µM of TNT. The “OFF” state (state 1) shows the mRNA structure and translation initiation rate calculated when TNT is unbound (r_TL,OFF). The “ON” state (state 4) shows the change in mRNA structure and translation rate when both TNT is fully bound to its aptamer and the ribosome is bound to the mRNA (r_TL,35mM). Nucleotides are color-coded based on their interactions, (light blue) the TNT aptamer sequence, (green) the Shine-Dalgarno sequence (SD), (dark blue) the last 9 nucleotides of the 16S ribosomal RNA, and (pink) the start codon for mRFP1. The orange bar is the ribosomal footprint for initiation. For riboswitch characterizations, we added a strong B. subtilis promoter upstream to measure (B) the mRFP1 fluorescence levels and (C) the growth rates (doubling time in minutes) of the TNT-RS14 riboswitch strain (purple dots) and mRFP1-only control (yellow squares) in response to varied concentrations of TNT, up to 35 µM. Data points and error bars are the mean and standard deviation of N = 3 biological replicates. RFU is a relative fluorescence unit.

Fig. 3. Design and testing of autonomous microbial TNT sensors in liquid culture.

A TNT activates the synthetic genetic circuit by binding to the RS14 riboswitch, which activates the expression of the Int2 recombinase. The Int2 recombinase binds to the att recognition sites and flips the state-switchable promoter’s orientation, which activates the expression of a mRFP1 reporter (output module). Designed sense promoters control the transcription rate of the mRNA encoding the RS14 riboswitch and Int2 recombinase. Designed anti-sense promoters control the reduction in mRNA levels, due to transcriptional interference. Numbers are predicted transcription and translation rates. B Average mRFP1 fluorescence levels of engineered B. subtilis cells during exponential growth in liquid culture carrying the most performant synthetic genetic circuit in response to varied TNT concentrations (0, 15, 25, and 35 µM). Bars and error bars are the mean and standard deviation of 3 biological replicates. Dots are individual data points. Numbers are activation ratios with statistical significance (two-tailed T test p-values are 0.0263 and 0.0112). C The single-cell mRFP1 fluorescence distributions are shown for these engineered B. subtilis cells responding to each TNT concentration versus a wild-type control (no mRFP1). D Average mRFP1 fluorescence levels of engineered B. subtilis cells carrying different synthetic genetic circuits growing in the same conditions, comparing responses at 0 µM TNT (blue bars) versus responses at 35 µM TNT (red bars). Bars and error bars are the mean and standard deviation of 3 biological replicates. Dots are individual data points. Numbers are activation ratios with statistical significance (two-tailed T-test p-values are 0.0337, 0.0329, 1.2E-05, 0.000196, 2.8E-06, 0.0195). All stars denote the statistical significance of comparisons based on p-value thresholds (*0.05, **0.01, ***0.001). Quantitative models of (E) transcriptional interference and (F) riboswitch regulation show how changes in sense promoter and antisense promoter transcription rates alter Int2 mRNA levels and translation activation. Transcription rate predictions are shown using the Promoter Calculator v1.0 scale. RFU is a relative fluorescence unit.

Fig. 4. Long-term testing of an autonomous microbial TNT sensor in wild soil.

A The experimental workflow and measurement timelines for characterizing the function and persistence of the TNT-sensing autonomous microbial sensor in wild soil. Measurements on soil samples included quantitative PCR (qPCR), colony-forming units (CFUs), flow cytometry (FLOW), and next-generation sequencing (NGS). B Measured red fluorescence levels of the TNT-sensing autonomous microbial in wild soil at 0 mg TNT / kg soil (blue circles), 4.5 mg TNT / kg soil (green squares), 45 mg TNT / kg soil (red diamonds), and 110 mg TNT / kg soil (tan stars) over a 28-day period. C Measured red fluorescence levels of WT B. subtilis cells (gray triangles), mRFP1 only B. subtilis cells (orange diamonds), and the no added cells control (purple octagons) in wild soil without TNT over a 28-day period. D Measured mRFP1 activation ratios of the TNT-sensing autonomous microbial sensor in wild soil with 4.5 mg TNT / kg soil (green bars), 45 mg TNT / kg soil (red bars), and 110 mg TNT / kg soil (tan bars) over a 28-day period. Bars and error bars are the mean and standard deviation of 3 biological replicates. Dots are individual data points. The statistical significance of the starred comparisons is quantified by two-tailed T test p-values (0.000003, 0.0001, 0.0021, 0.000049, 0.0000047, 0.00021, 0.000091, 0.0000006, 0.000495, 0.01198, 0.00060, 0.037026, 0.009956, 0.00448). E The percentage of autonomous microbial sensors in wild soil with varied TNT amounts over a 28-day period. F The percentage of auto-fluorescent control cells in wild soil without TNT over a 28-day period. G The percentage of other natural microbial cells in wild soil with varied conditions across a 28-day period. H Measured cell viability counts on selective agar plates (LB/Cm5) from wild soil samples containing the autonomous microbial sensor with varied TNT amounts. I Measured cell viability counts on selective (LB/Cm5) or non-selective (LB) agar plates from wild soil samples containing controls. J Measured copy number of autonomous microbial sensor genomes in wild soil with varied TNT amounts as quantified by qPCR. Data points and error bars are the mean and standard deviation of (B–I) 3 or (J) 2 biological replicates (independent soil containers), each with (B–I) 3 or (J) 2 technical replicates (samples per container). RFU is a relative fluorescence unit.