EXPRESS: Bioinspired and μPDMS Substrates for SERS-Based Detection of Microbial Adhesion and Colonization

Microbiological contamination of water and food remains a pressing concern in public health, the environment, and industry, largely due to the ability of bacteria to form biofilms that confer high resistance to antibiotics and disinfectants. One promising strategy to address this challenge involves the design of bioinspired surfaces combined with advanced analytical methods. This study investigated how the morphology and surface density of metal nanostructures influence bacterial adhesion and colonization by employing bioinspired patterns and micro-structured polydimethylsiloxane (PDMS) substrates. Model strains of Lactobacillus acidophilus , Lactobacillus casei , and Lactococcus lactis were characterized using surface-enhanced Raman spectroscopy (SERS) with complementary multivariate analysis. Results indicated that the Asian Lilium-inspired surface exhibited the strongest aseptic effect, significantly reducing bacterial adhesion and colonization compared with microstructured PDMS, which was more prone to microbial accumulation. Among the multivariate approaches, the support vector machine (SVM) model achieved superior predictive performance and sensitivity (R 2 Cal = 0.9999; R 2 CV = 0.958), with an estimated detection limit of approximately 3 × 10 1 CFU/mL; the partial least squares (PLS) model exhibited a practical detection of approximately 1.7 × 10 2 CFU/mL. Collectively, these findings demonstrate that integrating bioinspired substrates with SERS and advanced statistical models offers a robust, non-invasive approach to bacterial detection, with potential applications in microbiology, food safety, and environmental monitoring.