Synthetic toehold biosensors for detecting Listeria monocytogenes in raw milk

Víctor M. Carballo-Uicab, Luz E. Casados-Vázquez, Drew Endy, J. E. Barboza-Corona

Introduction To our knowledge, direct detection of foodborne bacterial pathogens using toehold biosensors has not been demonstrated in complex food matrices such as raw milk, particularly for Listeria monocytogenes . Here, we report the design and evaluation of toehold biosensors for detecting synthetic triggers and L. monocytogenes in both pure cultures and raw milk. Methods We performed a phylogenetic analysis of the V2-V3 hypervariable regions of the 16S rRNA gene, showing that Listeria spp. rRNA are distinct from other bacterial groups; a 36-nt trigger in the V2 region matched Listeria spp., whereas non- Listeria bacteria exhibited multiple mismatches. We designed several toehold sequences specific to the V2 region of L. monocytogenes and constructed three (A, B, C), selecting toehold B because of its highest experimental stability. Results Without NASBA amplification, toehold B responded to synthetic RNA triggers, inputs ranging from 10 to 10 12 copies (50 aM to 5 μM); these were used to evaluate sensor activation and to select a working RNA concentration for subsequent assays with biological RNA. Maximum activation was observed at 10 12 and 10 11 RNA copies, corresponding to 5 μM and 500 nM, respectively, yielding 17- and 5-fold changes at 30 min. A TRIzol-based protocol was used to extract bacterial RNA from pure culture or milk, producing consistent results in cell-free assays. Listeria RNA extracted with either a column-based kit or TRIzol reached a maximum absorbance of 3.5 at 600 nm. RNA from pure bacterial cultures of Listeria exhibited a fold change of 1.8, compared to approximately 1.5 when the toehold was activated with RNA obtained from crude milk contaminated with Listeria , confirming that the toehold biosensor can detect L. monocytogenes -derived RNA in a complex milk matrix. RNA from uncontaminated crude milk showed an activation ratio of 0.73, which differed statistically from the previous values. Conclusion Although the activation ratios are low, this first proof-of-concept demonstration shows that toehold-switch biosensors are activated by L. monocytogenes RNA from a complex food matrix (raw milk), not only by short synthetic triggers, suggesting a promising path toward more prevalent and affordable food security diagnostics. This finding remains a meaningful advance because detection is achieved in a cell-free format without nucleic acid amplification.