Label-Free Classification of Bacterial Extracellular Vesicles by Combining Nanoplasmonic Sensors With Machine Learning

Abstract

Bacterial extracellular vesicles (EVs) are nanoscale lipidenclosed packages that are released by bacteria cells and shuttle various biomolecules between bacteria or host cells. They are implicated in playing several important roles, from infectious disease progression to maintaining proper gut health, however the tools available to characterise and classify them are limited and impractical for many applications. Surface-enhanced Raman Spectroscopy (SERS) provides a promising means of rapidly fingerprinting bacterial EVs in a label-free manner by taking advantage of plasmonic resonances that occur on nanopatterned surfaces, effectively amplifying the inelastic scattering of incident light. In this study, we demonstrate that by applying machine learning algorithms to bacterial EV SERS spectra, EVs from cultures of the same bacterial species (Escherichia coli) can be classified by strain, culture conditions, and purification method. While these EVs are highly purified and homogeneous compared to complex samples, the ability to classify them from a single species demonstrates the incredible power of SERS when combined with machine learning, and the importance of considering these parameters in future applications. We anticipate that these findings will play a crucial role in developing the laboratory and clinical utility of bacterial EVs, such as the label-free, noninvasive, and rapid diagnosis of infections without the need to culture samples from blood, urine, or other fluids.

Classification of Preeclamptic Placental Extracellular Vesicles Using Femtosecond Laser Fabricated Nanoplasmonic Sensors

“…In contrast, t-SNE and UMAP offer a more clear boundary between NT and preeclamptic EV data. This finding is in agreement with our previously reported use of nonlinear manifold learning for the purpose of the EV Raman data visualization . However, due to the inherent heterogeneity of EVs, particularly those produced from mixed cell populations, an improved unsupervised technique was needed to clearly demonstrate the difference between NT, EOPE, and LOPE samples as well as the heterogeneity within individual samples.…”

“…As previously established, plasmonic nanostructures can concentrate incoming light near their nanometric features and sharp edges. 24,26 Such hotspots lead to significantly greater but highly localized electric field amplitudes, effectively enhancing any nonlinear phenomena such as Raman scattering. To quantify this effect for various geometries and conditions, we calculated the area of regions which have an electric field enhancement of more than 5 (corresponding to more than 5 4 fold Raman enhancement) as shown Figure 1h and i for 532 and 785 nm excitation wavelengths, respectively.…”

“…This finding is in agreement with our previously reported use of nonlinear manifold learning for the purpose of the EV Raman data visualization. 24 However, due to the inherent heterogeneity of EVs, particularly those produced from mixed cell populations, an improved unsupervised technique was needed to clearly demonstrate the difference between NT, EOPE, and LOPE samples as well as the heterogeneity within individual samples. To accomplish this, spatial K-mean clustering was performed for all of the samples' normalized spectra as shown in Figure 5a.…”

“…mesenchymal stromal cell (MSC) EV classification. 24,25 Clearly, the utilization of SERS for EV characterization in preeclampsia or other pregnancy diseases could provide immense value given its label-free and generally nondestructive nature.…”

“…Compared to conventional EV characterization tools, label-free surface-enhanced Raman spectroscopy (SERS) is ultrasensitive, nondestructive, and information-rich, with the resulting spectra providing a multiomics fingerprint of the EV sample of interest. The use of SERS to classify EVs has been primarily seen in cancer applications, − but has recently begun to expand into other EV-related research fields such as bacterial or mesenchymal stromal cell (MSC) EV classification. , Clearly, the utilization of SERS for EV characterization in preeclampsia or other pregnancy diseases could provide immense value given its label-free and generally nondestructive nature.…”