Christopher L. Davey

The dielectric properties of biological cells at radiofrequencies: applications in biotechnology

The study of the dielectric properties of cells in the radiofrequencies is increasingly leading to new practical applications, including online techniques for biomass measurements and novel techniques for the electrokinetic separation, manipulation, and characterization of single cells. In this review, we will discuss the dielectric properties of cells and their components and the electrical techniques that use them. This will be done mainly in the context of biotechnology but some applications in medicine will also be highlighted.

Real-time monitoring of cellular biomass : methods and applications

Abstract We review physical approaches to the problem of devising a real-time biomass probe. Direct measurement of the dielectric permittivity of cell suspensions at radio frequencies provides one possible solution to this problem.

DIELECTRIC PROPERTIES OF HUMAN BLOOD AND ERYTHROCYTES AT RADIO FREQUENCIES(0.2-10 MHZ) ; DEPENDENCE ON CELL VOLUME FRACTION AND MEDIUM COMPOSITION

The dielectric properties of human erythrocytes (red blood cells) suspended in whole blood and in isotonic media at various volume fractions (haematocrits) have been studied in the frequency range 0.2–10 MHz, in which the so-calledβ-dispersion due to the Maxwell-Wagner effect is known to occur. The capacitance and conductance at 25 °C were measured by an instrument interfaced to a computer. The rectangular sample cavity (1 ml volume) contained four pure gold electrode pins, and the sample could be circulated by a roller pump. The frequency-dependence of the permittivity and conductivity were fitted by non-linear least squares regression. Corrections were applied for non-linearity in the dielectric increment at high haematocrit, and for electrode polarisation when diluting the blood in saline. Data were interpreted in terms of a simple equivalent resistor-capacitor circuit. From the measured haematological values the specific membrane capacitance (Cm) and the conductivities internal and external to the cells (σ′i and σ′o respectively) were estimated. The conductivities behaved in a predictable manner with a mean of 0.458 S · m−1 (s.d. ± 0.044) for σ′i, whereas the value of Cm (and indeed the actual capacitance of the suspension) was dependent on the amount of plasma present. Hence, in stationary normal (anticoagulated) whole blood samples, Cm was as high as 2.98 μF · cm−2 (s.d. ± 0.40), in contrast to about 0.9 μF · cm−2 in blood diluted more than two-fold (to less than 20% hct) in isotonic media. The high value remained when the diluent was plasma. The Cm value returned to a high value when washed erythrocytes were reconstituted with plasma, provided that this was present at above a critical or threshold concentration of about 30 vol % in the medium, irrespective of the haematocrit in the range studied (15–44%). The Cm remained low in serum. When added to washed cells in saline, purified fibrinogen had no effect. However, high Cm values were obtained by fibrinogen supplementation to serum and diluted plasma. Applying moderate flow to whole blood approximately halved its high Cm value in an exponential manner with flow rate, whilst the Cm of washed cells (31–67% hct) slightly increased, and converged to the value for whole blood under flow. We interpret the highapparent Cm value in stationary samples to be a result of rapid cell aggregation in the presence of plasma, where rouleaux formation takes place before visible sedimentation sets in.

The permittistat: a novel type of turbidostat

Summary: Bakers yeast was grown in a novel type of turbidostat in which the steady-state biomass level was controlled not by the optical turbidity but by the dielectric permittivity of the suspension at appropriate radio frequencies. Dry weight, fresh weight, the optical density at 600 nm, percentage viability (from methylene blue staining), bud count and ethanol concentration were measured off-line and the cell size distribution was recorded using flow cytometry. Any changes in the physiological properties of the yeast had a negligible effect on the ratio between the permittivity set (and measured) and the steady-state dry weight, fresh weight or optical density of the cultures. The permittistat was found to provide an extremely convenient means for carrying out turbidostatic culture.

On-line, real-time measurements of cellular biomass using dielectric spectroscopy.

Introduction All else being equal, the productivity of a biological process is determined by the quantity of biomass present. There is therefore a major requirement for the accurate measurement and control of the biomass within fermentors, at both laboratory and industrial scales. Presently the range of sensors available that can be used in situ and reliably for the monitoring and regulation of biotechnological processes in general is rather limited. These sensors normally rely upon physical (e.g. optical, mechanical and electrical) or chemical variables (e.g. pH and concentration) rather than biological ones per se (Sarra et al., 1996; Pons, 1991). However only physical methods allow the on-line, real-time estimation of biomass (Harris and Kell, 1985). As well as physical methods, any easily determinable chemical that is produced or consumed by cells at an essentially constant rate during cell growth may also be used to assess biomass, e.g. carbon dioxide evolution and oxygen consumption. In these indirect methods biomass is then calculated based upon mass balances, stoichiometric relationships or empirical constants. However, this type of approach has the great disadvantage that it does not generally discriminate between biomass and necromass (Kell et al., 1990). Even if biomass was easily measurable there is still the question of what is biologically relevant information for fermentation control and how can one define and quantify it (e.g. metabolism, viability, vitality, morphology) (Kell et al., 1987; Kell, 1987a;

Real-time monitoring of the accretion of Rhizopus oligosporus biomass during the solid-substrate tempe fermentation.

We describe a novel method for the real-time estimation of the accretion of blomass during the solid-substrate tempe fermentation of soy beans, lupins and quinoa by Rhizopus oligosporus Salto. The method is based on measurements of the dielectric permittivity at radio-frequencies, using a four-terminal instrument (the Bugmeter). In all cases, excellent Ilnearity is observed during the growth phase between the dielectric permittivity and the hyphal length as determined microscopically.

Introduction to the dielectric estimation of cellular biomass in real time, with special emphasis on measurements at high volume fractions

The equations that describe the magnitude of the β-dielectric dispersion of biological cell suspensions are introduced. It is then demonstrated how this magnitude can be used to monitor cellular biomass concentrations in real time. These equations are then shown accurately to describe experimental data obtained over a wide range of cell sizes and volume fractions.

Dielectric Spectroscopy: a Rapid Method for the Determination of Solvent Biocompatibility During Biotransformations

Dielectric spectroscopy provides a convenient means of determining the degree of intactness of biological cells. 4-terminal dielectric measurements of suspensions of Saccharomyces cerevisiae at 0.4 MHz show that, as with all other biological cells, these organisms possess a substantial β-dispersion. The additional of octanol to such suspensions causes a rapid decrease in the electrical capacitance of the suspension, which parallels the cellular viability as determined by methylene blue staining. The kinetics of cell death are determined in part by the rate of dissolution of the organic solvent in the aqueous phase. The toxicity of several organic solvents to S. cerevisiae is studied using this technique, and is found to be dependent upon the polarity of the solvent. The present method provides a simple and rapid means for assessing the biocompatibility of solvents used in biotransformations.

On the audio- and radio-frequency dielectric behaviour of anchorage-independent, mouse L929-derived LS fibroblasts☆

Abstract The utility of dielectric methods as a means for estimating the biomass of animal cells in suspension culture was assessed, using mouse L929-derived LS fibroblasts. The dielectric increment of the β-dispersion was found to be a linear function of both cell number (30–70 permittivity units per 10 6 cells/ml, depending on the batch of cells) and volume fraction in the range measured (up to 1.28 × 10 8 cells/ml, volume fraction = 0.14). The notional distribution of relaxation times as encompassed in the Cole/Cole α (0.13 ± 0.03 SD) was rather modest. If the cells were treated as spherical shell capacitors of their observable diameter and number, the apparent capacitance of the plasma membrane was some 1.9–4.0 μF/cm 2 . This value significantly exceeded those (0.5–1 μF/cm 2 ) usually encountered or claimed, due predominantly to the possession by these cells of numerous plasma membrane protrusions. As the osmolarity of the suspending medium was increased using the non-permeant solute sorbitol, the apparent specific capacitance of the plasma membrane and the Cole/Cole αa were increased, whilst the dielectric increment per 10 6 cells/ml was unchanged. In addition, a secondary β-dispersion, with a characteristic frequency greater than that of the main β-dispersion, became increasingly prominent as the medium osmolarity was increased. It is proposed that this β 2 -dispersion is dominated by a Maxwell-Wagner mechanism taking place in the region of the plasma membrane protrusions of these cells.

Substitution and spreadsheet methods for analysing dielectric spectra of biological systems

Abstract1. Two major problems are encountered when one wishes to fit audio- and radio-frequency dielectric spectra of biological cell suspensions (or other materials): (a) changes in the apparent frequency-dependent permittivity of the system due to the phenomena of electrode polarisation can dominate those due to the biological system, and (b) because of the overlap of different dispersions it may be very difficult to deconvolute the individual contributions of the underlying biophysical mechanisms. 2. The extent of electrode polarisation depends substantially upon the conductivity of the medium surrounding the cells, but only marginally on the nature of the ions of a given valency contribution to it. 3. This, and the fact that the apparent time constants of the phenomena contributing to electrode polarisation are much greater than those of biological dielectric dispersions, permits one to use a simple substitution method to extract the latter in the presence of the former. This is shown both by simulation and by experiments using suspensions of human erythrocytes. 4. A spreadsheet method is described for the display of dielectric data and their conformance to the double Cole-Cole equation. The method provides a rapid and convenient approach, based on interactive graphical outputs, for the fitting of dielectric data to this equation. 5. Estimates derived from the spreadsheet program may be used in a BASIC program to arrive at the optimal fit. 6. The method is applied to the strongly-overlapping α- and β-dispersions of erythrocytes, permitting their deconvolution and providing a high level of accuracy.