Single molecule detection of double-stranded DNA in poly(methylmethacrylate) and polycarbonate microfluidic devices

• Published in: Electrophoresis Vol. 22; no. 18; pp. 3939 - 3948
• Main Authors: Wabuyele, Musundi B., Ford, Sean M., Stryjewski, Wieslaw, Barrow, James, Soper, Steven A.
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 01.10.2001
• Subjects: Bacteriophage lambda - chemistry; DNA, Viral - analysis; Double-stranded DNA; Electrophoresis, Capillary - instrumentation; Equipment Design; Fluorescent Dyes - analysis; Fluorescent Dyes - chemistry; Fluorometry - instrumentation; Intercalating Agents - analysis; Intercalating Agents - chemistry; Lasers; Microchemistry - instrumentation; Microfluidic devices; Molecular Weight; Poly(methylmethacrylate); Polycarboxylate Cement - chemistry; Polymethyl Methacrylate - chemistry; Quinolines - analysis; Quinolines - chemistry; Rheology; Sensitivity and Specificity; Thiazoles - analysis; Thiazoles - chemistry; Time Factors
• ISSN: 0173-0835; 1522-2683
• DOI: 10.1002/1522-2683(200110)22:18<3939::AID-ELPS3939>3.0.CO;2-9

Single photon burst techniques were used to detect double‐stranded DNA molecules in poly(methylmethacrylate) (PMMA) and polycarbonate (PC) microfluidic devices. A confocal epi‐illumination detection system was constructed to monitor the fluorescence signature from single DNA molecules that were multiply labeled with the mono‐intercalating dye, TOPRO‐5, which possessed an absorption maximum at 765 nm allowing excitation with a solid‐state diode laser and fluorescence monitoring in the near‐infrared (IR). Near‐IR excitation minimized autofluorescence produced from the polymer substrate, which was found to be significantly greater when excitation was provided in the visible range (488 nm). A solution containing ‐DNA (48.5 kbp) was electrokinetically transported through the microfluidic devices at different applied voltages and solution pH values to investigate the effects of polymer substrate on the transport rate and detection efficiency of single molecular events. By applying an autocorrelation analysis to the data, we were able to obtain the molecular transit time of the individual molecules as they passed through the 7 νm laser beam. It was observed that the applied voltage for both devices affected the transport rate. However, solution pH did not alter the transit time for PMMA‐based devices since the electroosmotic flow of PMMA was independent of solution pH. In addition, efforts were directed toward optimizing the sampling efficiency (number of molecules passing through the probe volume) by using either hydrodynamically focused flows from a sheath generated by electrokinetic pumping from side channels or reducing the channel width of the microfluidic device. Due to the low electroosmotic flows generated by both PMMA and PC, tight focusing of the sample stream was not possible. However, in PMMA devices, flow gating was observed by applying field strengths >–120 V/cm to the sheath flow channels. By narrowing the microchannel width, the number of molecular events detected per unit time was found to be four times higher in channels with 10 νm widths compared to those of 50 νm, indicating improved sampling efficiency for the narrower channels without significantly deteriorating detection efficiency. Attempts were made to do single molecule sizing of ‐DNA, M13 (7.2 kbp) and pUC19 (2.7 kbp) using photon burst detection. While the average number of photons for each DNA type were different, the standard deviations were large due to the Gaussian intensity profile of the excitation beam. To demonstrate the sensitivity of single molecule analysis in the near‐IR using polymer microfluidic devices, the near‐IR chromophore, NN382, was analyzed using our confocal imager. A detection efficiency of 94% for single NN382 molecules was observed in the PC devices.