Monitoring of real changes of plasma membrane potential by diS-C₃(3) fluorescence in yeast cell suspensions

Abstract

The fluorescent dye 3,3′-dipropylthiadicarbocyanine, diS-C₃(3), is a suitable probe to monitor real changes of plasma membrane potential in yeast cells which are too small for direct membrane potential measurements with microelectrodes. A method presented in this paper makes it possible to convert changes of equilibrium diS-C₃(3) fluorescence spectra, measured in yeast cell suspensions under certain defined conditions, into underlying membrane potential differences, scaled in the units of millivolts. Spectral analysis of synchronously scanned diS-C₃(3) fluorescence allows to assess the amount of dye accumulated in cells without otherwise necessary sample taking and following separation of cells from the medium. Moreover, membrane potential changes can be quantified without demanding calibration protocols. The applicability of this approach was demonstrated on the depolarization of Rhodotorula glutinis yeast cells upon acidification of cell suspensions and/or by increasing extracellular K⁺ concentration.

Appendix

Theory of diS-C₃(3) fluorescence response to membrane potential changes in yeast

Implications of the Nernst equation (Plasek et al. 1994)

The standard interpretation of the relationship between membrane potential and the uptake of small lipophilic ions by cells and/or cell organelles is based on the assumption that the equilibrium concentrations of unbound ions in cells, c_in, and cell suspension medium, c_out, obey the Nernst equation (Ehrenberg et al. 1988; Krasznai et al. 1995; Loew et al. 1993; Lolkema et al. 1982; Plasek and Hrouda 1991; Ross et al. 2005). For diS-C₃(3) assays relying on the spectral unmixing of free- and bound-dye contributions from complex fluorescence spectra of cell suspensions it is essential to use dye concentrations low enough to guarantee that the amount of intracellular bound dye is directly proportional to the intracellular free dye concentration resulting from the Nernstian accumulation of diS-C₃(3) in the cells (Plasek et al. 1994), see Methods.

Moreover, we will assume that the complex fluorescence of diS-C₃(3) stained cell suspensions is dominated by the fluorescence of free dye dissolved in the cell suspension medium, F_f, and the fluorescence of dye molecules bound to cytosolic proteins, or other intracellular substances, F_b, while both the contribution of fluorescence of unbound dye molecules present in the cell cytosol, F_in, and fluorescence of dye molecules bound to the outer cell surface, F ^out_b , is negligible in properly designed experiments. In particular, F_in is negligible compared to F_f if the partial volume of cells in measured suspensions is sufficiently low, what is the case in present experimental protocols. The amount of dye bound to the outer cell surface can also be neglected compared to the amount of the dye bound to cytosolic proteins. This was proved by fluorescence microscopy, which revealed no bright rim around the fluorescent images of stained cells.

Then, following the model presented in (Plasek et al. 1994), the ratio of bound and free-dye fluorescence intensities measured with spectral unmixing, i.e., F_b and F_f, respectively, can be combined with the Nernst equation to yield

$$ \frac{{{F_b}}}{{{F_f}}} = \frac{{{V_c}{Q_b}{S_b}{k_b}}}{{{V_s}{Q_f}{S_f}}}{e^{{ - \frac{F}{{RT}}\Delta \Psi }}} = {K_{{bf}}}{e^{{ - \frac{{\Delta \Psi }}{{{U_{{RTF}}}}}}}} $$

(A1)

where the physical constants R, T and F have their usual meaning, ΔΨ denotes the membrane potential. A number of multiplicative factors involved in this equation represents following experimental parameters: partial volume of cells (V_c) and cell suspension medium (V_s) within an effective fluorescence detection volume inside a cuvette, fluorescence quantum yields (Q_b and Q_f) and instrument sensitivity factors (S_b and S_f) of bound and free-dye fluorescence, respectively, and k_b is a proportionality constant of the ratio between the bound and free-dye concentrations inside the cells. For simplicity, the pre-exponential fraction in Eq. A1 can be represented by a single constant K _bf . To highlight quantitative aspects of this equation, we have introduced a voltage factor U_RTF = RT/F, the value of which is 25.7−26.9 mV at physiological temperatures between 25 and 37 °C.

Despite the impossibility to determine the actual value of the K_bf coefficient in real experiments Eq. A1 can be still used for the assessment of membrane potential differences in certain defined situations. In particular, this is the case, when it is justified to postulate that the cell-related parameters V_c and k_b remain constant on going from a cell physiological state 1 (characterized by membrane potential ΔΨ₁) to another state 2 (characterized by membrane potential ΔΨ₂). Upon a simple logarithmic transformation of Eq. A1 we get finally

$$ \Delta {\Psi_1} = - 2.3{U_{{RTF}}}\left[ {\log \frac{{{F_{{b1}}}}}{{{F_{{f1}}}}} - \log {K_{{bf}}}} \right] $$

(A2a)

$$ \Delta {\Psi_2} = - 2.3{U_{{RTF}}}\left[ {\log \frac{{{F_{{b2}}}}}{{{F_{{f2}}}}} - \log {K_{{bf}}}} \right] $$

(A2b)

These two equations can be combined into the following simple formula, in which the unknown constant K_bf has been cancelled out:

$$ \Delta {\Psi_1} - \Delta {\Psi_2} = - 2.3{U_{{RTF}}}\left[ {\log \frac{{{F_{{b2}}}}}{{{F_{{f2}}}}} - \log \frac{{{F_{{b1}}}}}{{{F_{{f1}}}}}} \right] $$

(A3)

In this way, the F_b/F_f fluorescence ratios obtained by the spectral unmixing of diS-C₃(3) fluorescence spectra measured in cell suspensions can be used to quantify membrane potential differences in the absolute scale of mV, without necessity to perform any a priori calibration experiment.

However, the Eq. A3 is obviously an oversimplification aimed merely to illustrate the idea of a quantitative link between the diS-C₃(3) fluorescence spectra and membrane potential changes of examined cells. To match real experimental conditions, the above model must be amended by including i) the uptake of diS-C₃(3) by mitochondria, ii) the effect of yeast cell wall and surface potential, and iii) the extrusion of diS-C₃(3) from yeast cells by MDR pumps, responsible for multi drug resistance of yeast cells.

Contribution of probe accumulation in mitochondria

The treatment of the contribution of mitochondria to the diS-C₃(3) accumulation in yeast cell follows also the model already used for animal cells (Plasek et al. 1994), which leads to a conclusion that the correction of Eq. A1 for the contribution of mitochondria can be represented by a multiplicative factor M(V_m, ΔΨ_m), which is the function of both the partial volume of mitochondria in cells, V_m, and mitochondrial membrane potential, ΔΨ_m.

$$ \frac{{{F_b}}}{{{F_f}}} = {K_{{bf}}}M\left( {{V_m},\Delta {\Psi_m}} \right){e^{{ - \frac{{\Delta \Psi }}{{{U_{{RTF}}}}}}}} $$

(A4)

This means that Eq. A3 can be still used for assessment of membrane potential differences whenever the postulation is justified that neither the partial volume of mitochondria in the cells, nor the mitochondrial membrane potential vary in the course of the experiment. The role of mitochondria-related artefacts in membrane potential assays can also be significantly lessened by working with fermenting yeast cells in which the function of the mitochondria is highly reduced (Criddle and Schatz 1969). On the other hand, experimental protocols based on uncoupling yeast mitochondria with protonophores are not recommended since these protonophores, such as carbonyl cyanide m-chlorophenylhydrazone (CCCP), collapse inevitably the proton electrochemical gradient across the plasma membrane as well (Macey et al. 1978; Pereira et al. 2008; Stevens and Nichols 2007).

The role of cell wall and surface potential

The effect of cell wall, which is specific for yeast and bacteria, has not been treated in models published in our preceding papers. Since the cell wall space of yeast possess negative Donnan potential, a correction should be made for the accumulation of cationic species in the close proximity of cell membrane (Borst-Pauwels 1989; Theuvenet and Borst_Pauwels 1983), albeit under some conditions the role of cation accumulation in the cell wall might be rather small (Gage et al. 1985).

As the Nernstian accumulation of diS-C₃(3) in yeast cells is concerned, a true dye concentration at the periplasm/plasmalemma interface, c_ppi, must be therefore inserted into the Nernst equation instead of its bulk concentration, c_out, in the cell suspension medium. The former concentration is proportional to the bulk diS-C₃(3) concentration through two distinct exponential factors that arise from i) the partition coefficient of diS-C₃(3) between the aqueous suspension medium and the less polar cell wall space, and ii) the series of Boltzmann equilibriums related to the Donnan potential of cell wall and the surface potential of plasmalemma. It seems therefore practical to introduce a certain “effective“negative surface potential ψ^*, which makes it possible to express the increase of local diS-C₃(3) concentration at membrane surface c_ppi with respect to its bulk value c_out in terms of a product c_out exp(ψ^*/U_RTF). Then the Nernst equation used to describe the accumulation of diS-C₃(3) in animal cells (Plasek et al. 1994), i.e.,

$$ {c_{{in}}} = {c_{{out}}}{e^{{ - \frac{F}{{RT}} \ \Delta \Psi }}} $$

(A5)

must be replaced by

$$ {c_{{in}}} = {c_{{out}}}{e^{{ - \frac{F}{{RT}}{\psi^{*}}}}}{e^{{ - \frac{F}{{RT}} \ \Delta \Psi }}}, $$

(A6)

and Eq. A3 modified accordingly to yield a following equation

$$ - 2.3{U_{{RTF}}}\left[ {\log \frac{{{F_{{b2}}}}}{{{F_{{f2}}}}} - \log \frac{{{F_{{b1}}}}}{{{F_{{f1}}}}}} \right] = \Delta {\Psi_2} - \Delta {\Psi_1} + \left[ {\psi_2^{*} - \psi_1^{*}} \right], $$

(A7)

which reflects a fact that the intracellular accumulation of small lipophilic cations does not respond correctly to membrane potential changes if the surface potential of examined cells varies in the course of the experiment. This is a very important, thought often ignored, aspect of monitoring cell membrane potentials by using lipophilic cations. In particular, the use of different buffers may result in serious discrepancies between diS-C₃(3) fluorescence intensities measured in apparently parallel assays, just because the surface potential of cell membranes is extremely sensitive to the ionic strength of cell suspension medium. On the other hand, when the ionic strength is maintained constant during the experiment, [ψ2*− ψ1*] becomes zero and thus can be omitted

Contribution of MDR pumps

Voltage-sensing fluorescent dyes are possible substrates of MDR pumps responsible for multi drug resistance of yeast cells. When considering the effect of these pumps on dye redistribution it is necessary to move from the thermodynamic equilibrium model to a steady-state equilibrium, under which the dye uptake by cells is counter balanced by the MDR pump mediated outward flux J_MDR (in mol m⁻¹ s⁻¹). The transmembrane inward flux of ions J (in mol m⁻¹ s⁻¹) that is driven by the membrane potential can be expressed

$$ J = \frac{G}{{zF}}\left( {{U_{{RTF}}}\ln \frac{{{c_{{out}}}}}{{{c_{{in}}}}} - \Delta \Psi } \right) $$

(A8)

where G is the membrane conductivity, z the ion valence, and other symbols as above (Friedman 2008). In the steady state this flux is offset by J_MDR, so that J = J_MDR. Under these circumstances Eq. A8 should be replaced by the following formula

$$ - 2.3{U_{{RTF}}}\log \frac{{{c_{{out}}}}}{{{c_{{in}}}}} = \Delta \Psi - \frac{{zF}}{G}{J_{{MDR}}} $$

(A9)

This implicates that the cationic dye redistribution reflects false membrane potentials that are reduced with respect to their true values by a term $\Delta {\Psi^{{MDR}}} = \frac{{zF}}{G}{J_{{MDR}}}$ and thus, Eq. A3 must be extended to a more complex formula

$$ - 2.3{U_{{RTF}}}\left[ {\log \frac{{{F_{{b2}}}}}{{{F_{{f2}}}}} - \log \frac{{{F_{{b1}}}}}{{{F_{{f1}}}}}} \right] = \Delta {\Psi_2} - \Delta {\Psi_1} - \left[ {\Delta \Psi_2^{{MDR}} - \Delta \Psi_1^{{MDR}}} \right] $$

(A10)

Unfortunately, little is known about the size of MDR mediated currents. This hampers both the qualitative and quantitative evaluation of the membrane potential dependent response of slow dyes. Therefore, Eq. A10 can be used for the assessment of membrane potential changes only in a few special cases, such as when working with MDR pump-deficient yeast mutants and/or with cells for which it was demonstrated that the actual contribution of MDR pumps to membrane potential, ΔΨ ^MDR, stays constant and independent of the rate of diS-C₃(3) accumulation in cells (see Methods). Under the latter conditions, the term $\left( {\Delta \Psi_2^{{MDR}} - \Delta \Psi_1^{{MDR}}} \right)$ of Eq. A10 is cancelled out and thus, true ΔΨ₂ – ΔΨ₁ differences can be obtained.

Important protocol requirements for quantifying membrane potential changes

The most important requirements for the quantitative evaluation of diS-C₃(3) fluorescence response to membrane potential changes in mV are:

• a) It is essential to use dye concentrations low enough to guarantee direct proportionality between the amount of intracellular bound dye and the intracellular free dye concentration resulting from the Nernstian accumulation of diS-C₃(3) in the cells.

• b) Any experimental protocol aimed at the assessment of membrane potential changes between two physiological states of examined cell suspension must guarantee that the steady-state dye redistribution between cells and their medium was reached. This does not apply when the time-course of membrane potential changes is monitored.

• c) All details of the experiment must be carefully controlled so that no variations occur in the density of cells in cell suspension or in ionic strength of the suspension medium.

• d) Ideal situation for the quantitative evaluation of diS-C₃(3) fluorescence in terms of underlying membrane potential changes is represented by experiments in which the cells respond to any stimulus in periods short enough from the point of view of possible changes in cellular parameters such as metabolism, protein synthesis, macromolecular composition of cytosol as well as partial volume of mitochondria and mitochondrial membrane potential.