Surface-Enhanced Raman Spectroscopy-Based Label-Free Insulin Detection at Physiological Concentrations for Analysis of Islet Performance

Label-free optical detection of insulin would allow in vitro assessment of pancreatic cell functions in their natural state and expedite diabetes-related clinical research and treatment; however, no existing method has met these criteria at physiological concentrations. Using spatially uniform 3D gold-nanoparticle sensors, we have demonstrated surface enhanced Raman sensing of insulin in the secretions from human pancreatic islets under low and high glucose environments without the use of labels such as antibodies or aptamers. Label-free measurements of the islet secretions showed excellent correlation among the ambient glucose levels, secreted insulin concentrations, and measured Raman emission intensities. When excited at 785 nm, plasmonic hotspots of the densely arranged 3D gold-nanoparticle pillars as well as strong interaction between sulfide linkages of the insulin molecules and the gold nanoparticles produced highly sensitive and reliable insulin measurements down to 100 pM. The sensors exhibited a dynamic range of 100 pM to 50 nM with an estimated detection limit of 35 pM, which covers the reported concentration range of insulin observed in pancreatic cell secretions. The sensitivity of this approach is approximately 4 orders of magnitude greater than previously reported results using label-free optical approaches, and it is much more cost-effective than immunoassay-based insulin detection widely used in clinics and laboratories. These promising results may open up new opportunities for insulin sensing in research and clinical applications.