polymer-based dense fluidic networks for high throughput

POLYMER-BASED DENSE FLUIDIC NETWORKS FOR HIGH THROUGHPUT SCREENING (HTS)

WITH ULTRASENSITIVE FLUORESCENCE Paul I. Okagbare1, Jost Gottert4, Proyag Datta4, Varshni Singh4 and

Steven A. Soper1,2,3 Department of Chemistry,1 Department of Mechanical Engineering,2 Center for

BioModular Multi-Scale Systems,3 and Center for Advanced Microstructures and Devices4

Louisiana State University, Baton Rouge, Louisiana

ABSTRACT High density fluidic networks were constructed and used as platforms for High Throughput Screening (HTS) with ultrasensitive fluorescence readout. The fluidic system consisted of a series of parallel processors (1 µm width x 1 µm pitch) replicated in PMMA via hot embossing using metal electroforms. A frame transfer charge coupled device (CCD) was used to detect single fluorescent molecules in the fluidic assembly when the molecules were electrokinetically driven through the device, which was mounted on a large Field-of-view (FoV = 200 µm) imaging system. Two different illumination geometries were investigated; (1) epi-illumination and (2) embedded waveguide with evanescent excitation. KEYWORDS: HTS, Microfluidics, CCD and Ultrasensitive Detection INTRODUCTION HTS of elements from combinatorial libraries represents the first step in the drug discovery pipeline. Microfluidics can provide a viable platform for performing HTS due to its ability to automate fluid handling and generate fluidic networks with high numbers over small footprints appropriate for optical imaging. Unfortunately, few efforts have been invested into developing microfluidics with high information content appropriate for HTS [1]. While most HTS campaigns depend on fluorescence, readers typically use point detection and serially address the assay results, significantly lowering throughput [2]. To address these challenges, we present here the fabrication of high density microfluidic vias packed into the imaging area of a large field-of-view (FoV) ultrasensitive fluorescence detection system. THEORY Two different fluidic architectures were evaluated for providing an optical system with single-molecule sensitivity and a large FoV: (1) Epi-illumination interrogation with beam shaping optics to provide a large FoV (Figure 1A). Fluidic channels are 1 μm (width and depth) with a pitch of 1 μm. Using a 40X objective (numerical aperture = 0.75), the FoV is 200 μm providing the ability to interrogate ~100 vias. (2) Air-embedded waveguide situated orthogonal to the fluidic vias, which defines the excitation volume (Figure 1B) and is produced using double-sided hot embossing. Fluorescence excitation is accomplished using an evanescent field

978-0-9798064-1-4/µTAS2008/$20©2008CBMS 594

Twelfth International Conference on Miniaturized Systems for Chemistry and Life SciencesOctober 12 - 16, 2008, San Diego, California, USA

(penetration depth ~300 nm) with extremely shallow channels to keep the sampling efficiency high (~60% for 500 nm deep channels). EXPERIMENTAL The fluidic structures were fabricated using UV-LiGA to produce Ni electroforms [3] that were subsequently used for hot embossing. Embossing of the structures was accomplished with a JenOptik HEX02 high-precision hot embossing system to create high fidelity in the features over large areas (Figure 2A). The fluorescence from the fluidic vias was monitored using a CCD operated in a frame transfer mode to track molecules moving through the illumination field.

Figure 1. (A) Schematic of the optical system possessing a large FoV for imaging. Using epi-illumination, large numbers of fluidic vias for HTS could be imaged simultaneously. (B) Fluidic channels poised over an air-embedded waveguide for interrogating large numbers of fluidic vias. Both devices are produced by hot embossing PMMA from metal electroforms.

Figure 2. (A) SEM of 0.5 μm (width) fluidic vias embossed into PMMA. (B) Schematic of fluidic network for HTS (1, 2 & 3 are SEMs of a Ni electroform related to the indicated sections of the device) using reactor channels for incubating the enzyme target with the substrate and potential drug inhibitor. The dense network of channels for imaging evaluates the efficiency of drug inhibition by counting single molecule fluorescence signatures.

Input Reservoirs

Reaction Zone (HTS)

1 µm x 1 µm x 0.5 µmchannels (detection region)

(B)

1

2

3

595


RESULTS AND DISCUSSION The integrated system is depicted schematically in Figure 2B and provides reactor networks for incubation of the enzyme target and drug inhibitor along with quantitative assessment of inhibition using single molecule counting. The utility of these multichannel networks for HTS with an optical system for producing the prerequisite sensitivity was demonstrated by monitoring single fluorescent tags (AF-660) using a series of microchannels in a test device (30 µm x 20 µm; pitch = 25 µm) with epi-illumination (Figure 3). The single molecules were electrokinetically driven through the fluidic channels (E = 80 V/cm, ν = 0.01 cm/s). The number of single molecules detected was linear with concentration.

Figure 3. (A) CCD image of single molecules of AlexaFluor 660 migrating through a multichannel chip. The fluorescence was collected with a 40x/0.75 objective using a 100 ms CCD exposure time. (B) 3-D image showing the intensity distribution of the single dye molecules. CONCLUSIONS We have developed a high density fluidic system for HTS. The fluidic chip provides automated fluidic handling and a high density of vias for the simultaneous high sensitivity readout of drug inhibition of selected targets. The system will be further evaluated by screening potential therapeutic agents for L1-Endonuclease (L1-EN), which induces DNA double-strand breaks and is associated with 45 different diseases, including aging [4]. ACKNOWLEDGEMENTS The authors acknowledge support of this work through the National Institute of Health (EB-006639) and the National Science Foundation (EPS-0346411). REFERENCES [1] Khandurina J, Guttman, A. (2002). Microchip-based high-throughput screening

analysis of combinatorial libraries. Curr. Opion. Chem. Biol. 6, 359-366. [2] www.evotec.com [3] Ford S.M, et al. (1999). Micromachining in plastics using X-ray lithography for

the fabrication of micro-electrophoresis devices. J. Biomech. Eng.-Trans. ASME 121, 13-21.

[4] Gasior S.L, et al. (2006). The human LINE-1 retrotransposon creates DNA double-strand breaks. J. Molc. Biol. 357, 1383-1393.

(B) (A)

596


polymer-based dense fluidic networks for high throughput

Documents