Anal. Chem. 2005, 77, 6999-7004
Determination of Single-Cell Oxygen Consumption with Impedance Feedback for Control of Sample-Probe Separation Damon M. Osbourn, Richard H. Sanger, and Peter J. S. Smith*
Program in Molecular Physiology, Marine Biological Laboratory, BioCurrents Research Center, Woods Hole, Massachusetts 02543
The ability to measure chemical gradients surrounding single cells provides novel insights into several areas of cell dynamicssparticularly metabolism. Detection of metabolic oxygen consumption can be achieved from a single mammalian cell using a modulated amperometric sensor in a self-referencing mode. To date, however, apart from visual cues, we do not have a reliable and cell-compatible method for determining and stabilizing the position of such probes. In this paper, we report on having successfully measured the increase in the uncompensated resistance of an electrochemical cell upon approach to single, living, biological cells, while simultaneously measuring the metabolic oxygen consumption. This was accomplished by applying an ac and a dc excitation signal to the electrode. The applied ac waveform was a 100-kHz sine wave with an amplitude of 10 mV rms, while the dc voltage applied was -600 mV. The two signals were shown not to interfere with one another. Furthermore, it is shown that the sample-probe distance can be measured for approach to single cells on the order of 10-15-µm diameter and 5-µm height, with 100-nm resolution. Several groups have previously demonstrated that electrochemical microsensors have the required properties for successful single-cell studies. In addition to good spatial resolution, these microsensors exhibit high temporal resolution enabling the investigation of relatively fast processes. For example, manually positioned carbon-fiber sensors have been employed to study the time course of the release of neurotransmitters.1,2 With the same method, hydrogen peroxide flux and corresponding oxygen concentration changes, from a single human fibroblast in response to mechanical stimulation, have been monitored.3 Oxygen consumption and photosynthetic production at substructures of cells have been demonstrated with isolated retinal photoreceptors4 and the alga Spirogyra greveilina.5 It has also been shown that the * To whom correspondence should be addressed. Phone: (508) 289-7241. E-mail:
[email protected]. (1) Cahill, P. S.; Walker, Q. D.; Finnegan, J. M.; Mickelson, G. E.; Wightman, R. M. Anal. Chem. 1996, 68, 3180-3186. (2) Bruns, D.; Jahn, R. Nature 1995, 377, 62-65. (3) Arbault, S.; Pantano, P.; Jankowski, J. A.; Vuilaume, M.; Amatore, C. Anal. Chem. 1995, 67, 3382-3390. (4) Malchow, R. P.; Land, S. C.; Patel, L. S.; Smith, P. J. S. Biol. Bull. 1997, 193, 231-232. 10.1021/ac050326w CCC: $30.25 Published on Web 10/01/2005
© 2005 American Chemical Society
modulation of neuronal oxygen flux can be measured using electrochemical microsensors.5 To measure oxygen flux in the latter experiments,4,5 the electrode was operated in a modulation format, stepping over a known distance (∆x) usually 10-20 µm, between near and far positions with respect to the cell at ∼0.3 Hz. The procedure, termed self-referencing, has been previously described for both potentiometric and amperometric sensors.6,7 This technique is particularly useful in measuring small extracellular gradients, such as oxygen consumption, in the presence of a large background concentration. The electrode spends ∼1.5 s at each position, allowing a large number of data points to be averaged, and then the difference in signal between the two positions is recorded. Because this is a relatively slow technique, error due to noise and drift is reduced and small differences in current of tens to hundreds of femtoamperes can be determined. The difference current obtained by an electrode with a step distance ∆x represents the slope of the extracellular diffusion gradient (∆C/∆x) created through the metabolic processes of the cell. Alternatively, this gradient can be represented as a flux (J, pmol/cm2‚s) through application of Fick’s first law, -J(x,t) ) (D)δC(x,t)/δx, where D is the diffusion coefficient of the electroactive species. Diffusion-based signals detected by the self-referencing technique are inevitably distance dependent. This is a result of the gradient created by the cell as it releases analyte into, or consumes analyte from, the bulk solution. The placement of the electrode near pole relative to the cell has an effect on the measured concentration change and thus flux. Although this distancedependent variation does not affect the determination of cellular responses to perturbation in a qualitative fashion, it does prevent the direct comparison of measurements taken at multiple points or multiple cells. Presently, electrode positioning is achieved by manually adjusting the probe tip in the cellular near-field. This has problems. First, it is difficult and requires operator experience to avoid touching the cell. Second, the measurements are limited to the x-y plane where the high contrast boundary of the cell can be seen. Third, it is not possible to observe small changes in (5) Land, S. C.; Porterfield, D. M.; Sanger, R. H.; Smith, P. J. S. J. Exp. Biol. 1999, 202, 211-218. (6) Smith, P. J. S. Nature 1995, 378, 645-646. (7) Smith, P. J. S.; Haydon, P. G.; Hengstenberg, A.; Jung, S. K. Electrochim. Acta 2001, 47, 283-292.
Analytical Chemistry, Vol. 77, No. 21, November 1, 2005 6999
the z axis of the electrode position that may arise while adding or removing solution to the sample dish. Fourth, the resolution of electrode placement is limited to 1 µm, even for experienced analysts. Finally, the repeatability of electrode placement cannot be determined. For these reasons, impedance measurements of the electrochemical cell will be used as positional feedback for amperometric electrodes. Previous uses of an impedance feedback approach have been largely limited to nonbiological systems with two exceptions. Kashyap and Gratzl employed impedance feedback for the placement of large (g250 µm) electrodes next to a monolayer of cells on a solid support.8 In this report a 20-mV ac signal was applied to large electrodes at a frequency of 250 Hz.8 Since the solution resistance does not change, the resistive component of the impedance can be related to the proximity of obstructions in the immediate vicinity of the probe tip. This method reports a dependence of the distance measurement on the size of the probe tip. A maximum distance of 300 µm was measured for a 250-µm probe, while a maximum distance of 400 µm was measured for a 500-µm probe. In this application, it is not obvious if the change in the ac signal is due to the cell monolayer or tp the solid support. Alpuche-Aviles and Wipf also detailed the use of impedance feedback, using 2- and 10-µm platinum disk electrodes with a 14-mV ac signal at
