(198a) Subcellular Fractionation in a Fluidic Microsystem by Dielectrophoresis (DEP)

In Proteome analysis the application of pre-fractionation techniques is a widely used strategy to reduce sample complexity and to enrich low abundance proteins. In this context, subcellular organelle fractionation is of particular importance, since functional units are isolated providing additional information with respect to the cellular localization of a particular protein (1, 2). Techniques currently employed for subcellular fractionation lack specificity to provide sufficiently pure organelle preparations for specific down-stream applications like comparative Proteome research. In order to overcome the above mentioned drawbacks of current organelle isolation techniques we have developed a micro fluidic system with integrated electrode arrays and a dedicated high through-put channel layout to isolate mitochondria from crude organelle samples based on dielectrophoretic sorting.

Mitochondria of a human lymphoblastoid cell line were stained with the fluorescent dye JC-1. Cells were homogenized with a douncer and the resulting cell homogenate was centrifuged at 1000xg for 10 min to remove cell debris and unbroken cells. The supernatant was centrifuged at 10.000xg for 10 min in order to obtain a crude mitochondrial preparation. The resulting pellet was suspended in a standard mitochondrial storage buffer at a concentration of 0.2 µg/µL and pumped through the microsystem at rates of 1 µl/min (Figure 1). Mitochondria were deflected by dielectrophoretic forces for 36 hours. Mitochondrial samples (before and after dielectrophoresis) were characterized either by 1-D electrophoresis followed by immunoblotting with organelle specific antibodies or by 2-D electrophoresis followed by mass spectrometric analysis.

Micro channels which contain deflector electrode arrays are generating inhomogeneous electric fields. Particles like cell organelles moving through the channels become polarized and experience dielectrophoretic forces whose magnitude depends on their size and dielectric properties. The small size of organelles of typically less than 1 µm in diameter require both high field strength and large degree of field inhomogeneity and consequently small electrode distances in the order of 10 ? 20 µm. By variation of buffer flow velocity, voltage amplitude and frequency, optimum parameters for manipulation of mitochondria were determined. Deflection of mitochondria at 400 kHz and 4kV/cm was successfully demonstrated. In a time frame of 20 hours ~100 µg mitochondrial protein was collected and further analyzed by 1-D immunoblotting with organelle specific antibodies. Data evaluation of the immunoblots showed that the ratio between antibody signal intensity and protein amount loaded per lane differs significantly between the crude starting material and DEP sample. For VDAC (mitochondrial marker protein) an enrichment factor of about 2.5 can be calculated. Concerning BiP (marker protein for ER) a sixfold decrease in signal intensity is obvious for the DEP sample. LAMP1 (lysosomal marker protein) immunoblot signals were not evaluated in a quantitative manner, but almost no LAMP1-signal was observed for the DEP sample. 2D-PAGE analysis followed by mass spectrometry confirmed the superior purity of the DEP sample in comparison to standard gradient centrifugation techniques for the isolation of mitochondria.

References (1) H. Lu et al., Anal.Chem. 76, 5705-5712 (2004). (2) S. Brunet et al., Trends in Biotechnology 13, 829-637 (2003).