Automated Microfluidic Droplet-Based Sample Chopper for Detection of Small Fluorescence Differences Using Lock-In Analysis.

ABSTRACT

Fluorescence is widely used for small-volume analysis and is a primary tool for on-chip detection in microfluidic devices, yet additional expertise, more elaborate optics, and phase-locked detectors are needed for ultrasensitive measurements. Recently, we designed a microfluidic analog to an optical beam chopper (µChopper) that alternated formation of picoliter volume sample and reference droplets. Without complex optics, the device negated large signal drifts (1/f noise), allowing absorbance detection in a mere 27 µm optical path. Here, we extend the µChopper concept to fluorescence detection with standard wide-field microscope optics. Precision of droplet control in the µChopper was improved by automation with pneumatic valves, allowing fluorescence measurements to be strictly phase locked at 0.04 Hz bandwidth to droplets generated at 3.50 Hz. A detection limit of 12 pM fluorescein was achieved when sampling 20 droplets, and as few as 310 zeptomoles (3.1 × 10-19 mol) were detectable in single droplets (8.8 nL). When applied to free fatty acid (FFA) uptake in 3T3-L1 adipocytes, this µChopper permitted single-cell FFA uptake rates to be quantified at 3.5 ± 0.2 × 10-15 mol cell-1 for the first time. Additionally, homogeneous immunoassays in droplets exhibited insulin detection limits of 9.3 nM or 190 amol (1.9 × 10-16 mol). The combination of this novel, automated µChopper with lock-in detection provides a high-performance platform for detecting small differences with standard fluorescence optics, particularly in situations where sample volume is limited. The technique should be simple to implement into a variety of other droplet fluidics devices.