For the successful treatment of infections, real-time analysis and enhanced multiplex capacity, sensitivity and cost-effectiveness of the developed detection method are critical. In this work, surface-enhanced Raman scattering (SERS) was employed with the final aim of identification and discrimination of pathogenic bacteria, based on their detected SERS fingerprint at the single-cell level. Several genera of bacteria that are found in most of the isolated infections in bacteraemia were successfully identified in less than 5 minutes without the use of antibodies or other specific receptors. The key element of the SERS direct detection platform is the SERS substrate, which combines easy production at low costs with a high enhancement enabling single-cell detection. The innovative approach of detection required the in situ synthesis of silver nanoparticles (NPs), ensuring an intimate contact with the bacterial membrane. This protocol provided a good reproducibility of the single-cell SERS spectra and was successfully applied both on Gram-negative and Gram-positive microorganisms (E. coli, M. morganii, E. lactis, L. casei). Thus, a label-free SERS-based biosensor for pathogen detection was developed with low costs, minimal sample preparation, high-accuracy and a very short analysis time of less than 5 min, which is crucial for infection diagnosis.

Raman scattering and its particular effect, surface-enhanced Raman scattering (SERS), are whole-organism fingerprinting spectroscopic techniques that gain more and more popularity in bacterial detection. In this work, two relevant Gram-positive bacteria species, Lactobacillus casei (L. casei) and Listeria monocytogenes (L. monocytogenes) were characterized based on their Raman and SERS spectral fingerprints. The SERS spectra were used to identify the biochemical structures of the bacterial cell wall. Two synthesis methods of the SERS-active nanomaterials were used and the recorded spectra were analyzed. L. casei and L. monocytogenes were successfully discriminated by applying Principal Component Analysis (PCA) to their specific spectral data.

Early diagnosis based on screening is recognized as one of the most efficient ways of mitigating cancer-associated morbidity and mortality. Therefore, reliable but cost-effective methodologies are needed. By using a portable Raman spectrometer, a small and easily transportable instrument, the needs of modern diagnosis in terms of rapidity, ease of use and flexibility are met. In this study, we analyzed the diagnostic accuracy yielded by the surface-enhanced Raman scattering (SERS)-based profiling of serum, performed with a portable Raman device operating in a real-life hospital environment, in the case of 53 patients with gastrointestinal tumors and 25 control subjects. The SERS spectra of serum displayed intense bands attributed to carotenoids and purine metabolites such as uric acid, xanthine and hypoxanthine, with different intensities between the cancer and control groups. Based on principal component analysis-quadratic discriminant analysis (PCA-QDA), the cancer and control groups were classified with an accuracy of 76.92%. By combining SERS spectra with general inflammatory markers such as C-reactive protein levels, neutrophil counts, platelet counts and hemoglobin levels, the discrimination accuracy was increased to 83.33%. This study highlights the potential of SERS-based liquid biopsy for the point-of-care diagnosis of gastrointestinal tumors using a portable Raman device operating in a clinical setting.

Plasmonic noble metal nanostructured films have a huge potential for the development of efficient, tunable, miniaturized optical sensors. Herein, we report on the fabrication and characterization of gold-coated nanopost arrays, their use as refractometric sensors, and their optimization through photonics simulations. Monolithic square nanopost arrays having different period and nanopost size are fabricated by nanoimprint lithography on polymer foils, and sputter-coated by gold films. The reflectivity of these gold nanopost-shell arrays present dips in the visible range, which are efficient for refractometric sensing. By finite-difference time-domain (FDTD) simulations we reproduce the experimental spectra, describe the electric fields distribution around the nanopost-shells, and then explain their good sensitivity, around 450 nm/RIU. Furthermore, we determine by simulations the influence of several geometrical parameters, such as array period, nanopost width, gold film thickness, and nanopost side coverage on both reflectivity spectra and sensing capabilities. Fully coated nanoposts provide an extremely deep reflectivity minimum, approaching zero, which makes the relative reflectivity change extremely high, more than two orders of magnitude higher than for partially coated nanoposts. These results contribute to the understanding of the plasmonic properties of metal coated nanopost arrays, and to the development of efficient platforms for sensing and other surface plasmon based applications.

No abstract