Rapid Microbial Profiling via Multimodal Single Cell Biosensors

Introduction:: Recent studies have revealed the pivotal functions and complex interactions of the microbial community with the host environment. The ability to rapidly analyze the microbial community holds the potential to elucidate these complex interactions and harness the findings for diverse medical and biological applications. Microbial profiling is typically carried out through high-throughput sequencing of microbial DNA or RNA. One of the most common approaches is 16S rRNA gene sequencing, which uses the genetic sequence of the 16S rRNA gene to identify bacterial taxonomy. Metagenomic sequencing is another method that analyzes the entire DNA content of a microbial community and provides information on both bacterial and non-bacterial organisms. Metatranscriptomic sequencing, on the other hand, analyzes the RNA content of a microbial community, providing information on the gene expression in the community. However, conventional sequencing methods can suffer from long processing time, high cost, sequencing bias, inability to discriminate dead cells, complex bioinformatic workflow, and lack of spatial resolution. These limitations prevent their routine use in medical diagnostics and longitudinal studies. Effective methods that can rapidly characterize the taxonomy of the microbial community and their activities with high resolution will provide new opportunities to analyze, understand, and control the microbial community for biomedical applications. In this study, we present a multimodal single-cell biosensor platform for rapid profiling of microbial taxa and their gene expressions.

Materials and Methods:: The biosensor platform consists of a bioinformatic approach for designing molecular probes for phylogenetic analysis of bacterial 16S rRNA and gene expression analysis of mRNA. A custom R package was created to design double-stranded nucleic acid probes at multiple taxonomic levels. We employed an exhaustive algorithm to systematically determine the consensus sequence and performed statistical analyses on the binding energy to ensure that the selected probes have high sensitivity and specificity. Furthermore, we predicted the secondary structure of the 16S rRNA to identify the most appropriate probe binding positions. We also checked the hairpin and self-dimer areas of the donor sequence to avoid strong hairpin or self-dimer structures.



Fecal samples were collected from five healthy rats to demonstrate the biosensor platform. The samples were diluted in PBS at a 1:10 ratio and vortexed for 10 minutes. Low-speed centrifugation (200 x g for 2 minutes) was performed twice to separate bacteria and large debris. The supernatants containing bacteria were collected by high-speed centrifugation (14,000 rpm for 5 minutes). The fluorescently labeled double-stranded DNA probes were applied to detect the bacteria using a modified fluorescence in situ hybridization (FISH) protocol. Fluorescence microscopy and an automated image analysis workflow were used to quantify the microbial community and genes of interest at the single cell level (Figure 1). A portion of the samples was also sent to a commercial sequencing service (CD Genomics) for quantitative 16S rRNA sequencing.


Results, Conclusions, and Discussions:: Result: To demonstrate taxonomic analysis, we targeted Proteobacteria, Bacteriodota, Actinomycetota, and Firmicutes, which are predominant phyla in murine feces and play critical physiological roles. The microbial community profiles of the rat fecal samples were characterized with the biosensor platform. The heatmap shows the abundance of the bacteria at the phylum level (Figure 2A). The results are in good agreement with data from 16S rRNA sequencing (Figure 2A). For instance, the biosensor platform successfully captured the increase in abundance of Proteobacteria in sample 2 and Bacteriodota in sample 3.

We demonstrated the analysis of gene expression in bacteria at the single-cell level using the biosensor platform. Gut bacteria, such as L. plantarum, are crucial regulators of γ-aminobutyric acid (GABA), a neurotransmitter that regulates gut motility, immune function, mood, and behavior. These bacteria produce GABA through the activity of enzymes including glutamate decarboxylase B (gadB), which converts L-glutamate into GABA by removing the carboxyl group from glutamate. We designed an mRNA probe to identify bacterial communities that express gadB. Figure 2B shows gadB upregulation in Lactobacillus in response to 30 mg/mL of glutamate. The results demonstrate the biosensor's ability to detect time-dependent gadB upregulation upon glutamate stimulation.


Discussion: Our data suggest that the biosensors successfully captured microbial profiles compared to 16S rRNA sequencing. The biosensor requires only 3 hours of processing time compared to several days in conventional 16S rRNA sequencing workflows. Furthermore, we demonstrated the detection of gadB mRNA in bacteria involved in the synthesis of the neurotransmitter GABA. Our results demonstrate that the biosensor can resolve glutamate-induced gadB gene upregulation, suggesting a potential application of the biosensor platform for longitudinal studies1. These results support the use of the biosensor platform for rapid identification and quantification of bacterial taxonomy associated with various GI tract, neurological, and immunological conditions. Overall, our biosensor platform provides a valuable tool for understanding microbial communities and their interactions with their environments.

Conclusion: Our multimodal biosensors enable comprehensive analyses of bacterial taxonomy and gene expression at the single-cell level. This method offers a fast and efficient approach that complements sequencing-based techniques while providing high sensitivity and specificity.


Acknowledgements (Optional): :

References (Optional): : [1] Mancini, Andrea, et al. "In vitro probiotic characterization of high GABA producing strain Lactobacilluas brevis DSM 32386 isolated from traditional “wild” Alpine cheese." Annals of Microbiology 69 (2019): 1435-1443.