Ultra-sensitive protein detection using droplet digital ELISA

Proteins are important biomarkers for disease detection, monitoring, and treatment. Many proteins are present at low levels in biological samples and therefore techniques for ultra-sensitive protein detection are necessary. To overcome limitations in analytical sensitivity, David Walt’s lab at BWH previously developed a single molecule protein detection method known as digital ELISA using Single Molecule Arrays (Simoa). Simoa immunoassays are 1000X more sensitive than the conventional ELISA. However, in many cases, the sensitivity of Simoa is still not sufficient. Some proteins, such as neurological proteins that are present at low levels in the blood due to difficulty crossing the blood brain barrier and cancer biomarkers are still unmeasurable due to inadequate sensitivity. Additionally, high sensitivity tools for protein detection are often costly and require complex devices with ultra-fine structures. As a result, high sensitivity tools for protein detection are generally not amenable to point-of-care applications (POC). Thus, protein detection tools that have improved sensitivity over the conventional Simoa while still being amenable to POC applications are needed.

We developed a novel approach for ultra-sensitive, single-molecule detection of proteins using digital ELISA and droplet microfluidics (Figure1). It is a bead-based digital immunoassay in which beads are isolated in pL-sized droplets and then loaded into a chamber, forming droplet arrays, for analysis. More specifically, antibody coated paramagnetic beads are added to a sample containing the target molecule. The target molecule is then labeled with a biotinylated detection antibody and streptavidin-β-galactosidase (SβG), forming an enzyme-labeled immunocomplex. The beads are then re-suspended in a small volume (2 ml) of substrate, fluorescein di-β-D-galactopyranoside (FDG) and the mixture is partitioned into pL droplets such that most droplets have zero beads and a small percentage has one bead. The droplets are then loaded into a chamber in a monolayer to form droplet arrays. Images in three channels are obtained to identify the droplets, the beads, and the fluorescent product and thus the “on droplets”.

This approach allows us to sample low numbers of molecules, thereby reducing measurement uncertainty and increasing sensitivity. Our approach demonstrates high sensitivity, with improvements of up to approximately 25-fold in sensitivity over Simoa, which is the current state-of-the-art approach for ultra-sensitive protein detection. Additionally, due to the simplicity of our device design, our method is promising for POC applications. 

This project is collaborated with Limor Cohen from David Walt’s lab at BWH.

Figure1 Ultra-sensitive protein detection using digital ELISA and droplet microfluidics.