Highly sensitive SERS sensors for glucose detection based on enzyme@MOFs and ratiometric Raman

Diabetes is a common chronic metabolic disease. The frequent fluctuation of glucose is the main cause of most diabetes complications, which in turn causes harm to the health of patients. Surface-enhanced Raman scattering (SERS) spectroscopy has attracted much attention in the rapid detection of glucose due to its unique molecular fingerprinting ability, ultra-high sensitivity and fast response. However, due to the low affinity between glucose and SERS substrate, poor signal, susceptibility to complex environmental interference, and poor stability of SERS detection, it is still a challenge for SERS to accurately and sensitively determine glucose in complex environments. In this work, we encapsulated 4-mercaptobutyronitrile (4-MBN) as an internal standard (IS) in Au@Ag NRs inside and then Au@4-MBN@Ag NRs, Leucomalachite Green (LMG), glucose oxidase (GOx) and horseradish peroxidase (HPR) were encapsulated in ZIF-8 to prepare a tandem enzyme catalytic ratiometric SERS sensor Au@4-MBN@Ag@LMG@ZIF-8(GOx, HPR) for the detection of glucose in saliva. Because ZIF-8 enhanced the catalytic activity of the enzyme, the ability of glucose enrichment, and weakens the aggregation of Ag NRs. The internal standard signal molecule improves the accuracy and sensitivity of detection. The ratiometric Raman signal I₄₁₂/I₂₂₃₃ of glucose has a good linear relationship with the concentration in the range of 0.1-100 μM, and the limit of detection (LOD) could be down to 0.03 μM. At the same time, it has excellent selectivity, repeatability and accuracy. The recovery rate of glucose in saliva is 96.50%-105.56 %, which proves the feasibility of the method. The Au@4-MBN@Ag@LMG@ZIF-8(GOx, HPR) sensor prepared in this study showed excellent SERS performance, which was able to detect glucose quickly, sensitively and accurately. This work provides a new strategy for the design of enzyme-catalyzed SERS sensors.