Cytosolic Free Calcium Measurements in Single Cells Using Calcium-Sensitive Fluorochromes

Abstract

Despite a 10,000-fold gradient of Ca²⁺ across the cell membrane, the concentration of cytosolic free calcium, [Ca²⁺]_i, is regulated with remarkable constancy. A combination of mechanisms precisely regulate [Ca²⁺]_i at nanomolar levels. These include influx of Ca²⁺ via plasma membrane calcium channels, release and redistribution of Ca²⁺ from internal stores, and efflux of Ca²⁺ by the action of ATP-driven calcium pumps.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

1.1 Background

Despite a 10,000-fold gradient of Ca²⁺ across the cell membrane, the concentration of cytosolic free calcium, [Ca²⁺]_i, is regulated with remarkable constancy. A combination of mechanisms precisely regulate [Ca²⁺]_i at nanomolar levels. These include influx of Ca²⁺ via plasma membrane calcium channels, release and redistribution of Ca²⁺ from internal stores, and efflux of Ca²⁺ by the action of ATP-driven calcium pumps.

A critical evaluation of the role of calcium as an intracellular messenger requires quantitative measurements of [Ca²⁺]i, and comparisons of varied stimuli and cell responses. During the past decade, the most popular method for measuring [Ca²⁺]i has been to monitor the fluorescence of an indicator, quin 2. However, this dye has some major and acknowledged disadvantages (Table 1). In 1985, Grynkiewicz et al.) synthesized a new generation of Ca²⁺ indicators, including fura 2 and indo 1, with greatly improved fluorescence properties. More recently, a third generation of calcium indicators has been developed that permits the use of long-wavelength excitation sources, including argon lasers for the determination of [Ca²⁺]_i. These developments have allowed extensive biochemical studies on the physiological role of cytosolic free calcium.

Table 1 Disadvantages of Using Quin 2

1.2 Chemistry

The chemical structures of two popular members (fura 2 and indo1) of this family of highly fluorescent indicators is shown in Fig. 1. Briefly, these compounds combine an 8-coordinate tetracarboxylate chelating site with stilbene chromophores. The combination of the ethylene-linked stilbene with a heterocyclic ring enhances the quantum efficiency and photochemical stability of the fluorophore.

Fig. 1.

Chemical structures of (A) indo 1/AM and (B) fura 2/AM.

1.3 Fluorescent Properties

Compared to quin 2, fura 2 and indo 1 have up to a 30-fold brighter fluorescence. Binding of Ca²⁺ to these indicators produces a change in wavelength, as well as a change in intensity of fluorescence. Furthermore, fura 2 and indo 1 possess lower affinities for Ca²⁺, have slightly longer wavelengths of excitation, and have a considerably improved selectivity for Ca²⁺ over other divalent cations. These optical properties have made these dyes the preferred fluorescent indicators for many intracellular applications, especially in adherent cell layers, bulk tissues, or single cells either viewed under a microscope or in a flow cytometer. The excitation and emission wavelengths of a number of newly available ion-sensitive fluorochromes are shown in Table 2.

Table 2 Ion-Sensitive Fluorochromes for Microspectrofluorimetry

1.4 Experimental Strategy

1.4.1 Principle of Ratiometric Measurements

The authors have used indo 1 as an example to illustrate changes in the fluorescence spectra on combination of the fluorochrome with Ca²⁺ (Fig. 2). Indo 1 is a dual-emission dye, the emission spectrum of which is affected by changes in [Ca²⁺],. The emission spectrum of 1 μM indo 1 has been obtained using a Perkin Elmer LS-5 spectro-fluorimeter. The excitation wavelength is 340 nm, and the emission spectra between 400 and 500 nm have been obtained at 20°C. As the [Ca²⁺] progressively increases from 100 to 1000 nM, the peak emission decreases and shifts to lower wavelengths. Thus, by determining the position of the emission spectrum, it is possible to derive the [Ca²⁺]_i in the environment of the dye. This is done by measuring the intensity of emission at two wavelengths, 405 and 480 (470–490) nm, and calculating the ratio of intensities at these wavelengths. By using ratiometric methods, the measurement of [Ca²⁺ _i], is not influenced greatly by changes in dye concentration or cell volume (see Notes 1 – 5 ).

Fig. 2.

Emission spectra for the Ca²⁺-sensitive fluorescent dye indo 1 (1 μM) obtained using a Perkin Elmer LS-5 spectrofluorimeter. The dye was dissolved in K⁺-substituted balanced salt solution with HEPES (10 mM and EGTA (2.5 mM), and varying [Ca²⁺]. The excitation wavelength was 340 nm, and the emission spectra from 350 to 550 nm were obtained at room temperature (20°C). The ordinate shows relative fluorescence.

1.4.2 Dye Loading

Indo 1 is loaded into intact cells by incubating them with a membrane-permeant acetoxymethyl ester, indo 1/AM. Cytosolic esterases split off the ester groups and leave the membrane impermeant indo 1 tetra-anion trapped in the cytosol. There are two basic problems dependent on dye loading that can lead to either an underestimation or an overestimation oftCa²⁺ See Note 6).

Incomplete loading in some cells, where indo 1 and other fluoro-chromes compartmentalize in noncytoplasmic compartments, might result in the underestimation of [Ca²⁺],. The addition of pluronic acid (Molecular Probes, Eugene, OR) helps to overcome this artifact.

Problems of quantitation also arise when the dye is incompletely hydrolyzed intracellularly, as is often noted in endothelial cells (2). The fluorescence emitted by the uncleaved ester (indo 1/AM) at 480–500 nm interferes with the fluorescence owing to Ca²⁺-bound indo 1. Also, the emission of the latter complex measured at 400 nm influences the fluorescence signal of the ester. Thus, although a basal [Ca²⁺], is measurable with time, the measurement of [Ca²⁺], increases can be overestimated because of the presence of an undefined amount of ester, in combination with a secondary fluorescence that cannot be quantitated.

1.4.3 Intracellular Calibration

For every Ca²⁺ chelator exhibiting 1:1 binding stoichiometry, the ratio of the [Ca²⁺] bound to indo 1 (bound) to that of free dye (free) is related to [Ca²⁺],:

(1)

where K _d is the dissociation constant. For indo 1, K _d is 250 nM as measured in the presence of a solution resembling intracellular milieu: 115 mM KC1, 20 mM NaCl, 10 mM K-MOPS, pH 7.05, and 1 mM Mg²⁺, at 37°C (1).

The fluorescence of indo 1 may be named “Faa” if an emission wavelength is chosen where the fluorescence of the Ca²⁺-saturated dye is greater than that of the free dye and may be “G” when the reverse is true. Thus, at an excitation wavelength of 355 nm, when the isosbestic emission wavelength is 450 nm, emitted fluorescence becomes “F” at a wavelength <450 nm and “G” at a wavelength >450 nm. Then the actual fluorescence intensities, F_act and G_act, indicate [Ca²⁺]₁.

(2)

(3)

The subscripts “max” and “min” denote the maximal and minimal values, respectively, at a given dye concentration, that result if the dye is in the Ca²⁺-bound (F_max, G_mtn) or free (F_min, G_max) form.

A combination of the Eqs. (2) and (3) yields:

(4)

Substituting F/G for R yields:

(5)

Thus, in contrast to fluorescence, F and G, R (ratio) changes only with [Ca²⁺]_i, but not with the concentration of indicator. Equations (4) and (5) allow calculation of [Ca ²⁺]_i in indo 1-loaded cells when emitted fluorescence is simultaneously recorded at two wavelengths (405 and 480 nm). The new software from Newcastle Photonics allows this. According to Eq. (5), therefore, the ratio R_act = F_act/G_act should be a measure of the [Ca²⁺]_i that is independent of the total amount of intracellular dye, provided the proper corrections are applied.

R_max is obtained by measuring, at the end of the experiment, the fluorescence in the presence of the calcium ionophore, ionomycin, which permits free entry of Ca²⁺ into the cells.

R_min can be defined in relation to R_max from the changes in fluorescence that results if Ca²⁺ is stripped from Ca²⁺-saturated indo 1 by EGTA and alkalinization measured in series of parallel experiments carried out under the same conditions.

The background fluorescence, which needs to be subtracted from all test measurements, is also obtained at the end of the experiment. Mn²⁺ is added to quench the dye fluorescence subsequent to the addition of ionomycin.

2 Materials

2.1 Reagents

• 1. Balanced salt solution (BSS): 140 mM NaCl, 5 mM KC1,1.25 mM CaCl₂, 1 mM MgCl₂, 10 mM Na₂HPO₄, 5 mM Na₂CO₃, 10 mM glucose, 1 g/L bovine serum albumin, and 10 mM HEPES-NaOH buffered to pH 7.4.

• 2. K⁺-substituted solutions: Na⁺ is replaced by K⁺ to the required molarity.

• 3. Ca²⁺-free solutions: 1.25 mM CaCl₂ is substituted with 2.5 mM ethyleneglycol-bis-(aminoethyl ether) tetraacetic acid (EGTA).

• 4. Calibrating solution: 150 mMKCl, 10 mMHEPES, and 2.5 mMEGTA, containing different [Ca²⁺] (see Table 3).

• 5. Indo-1-acetoxymethyl ester (indo 1/AM).

• 6. Indo-1.

Table 3 Extracellular Calibration Solutions ^a

2.2 Sample System

• 1. A perspex (lucite) bath, about 2.2 cm in diameter, incorporating inlet and outlet ports for perfusing cells mounted on a cover slip, with a thermo-couple device for regulating the temperature of the perfusate. The bath is mounted on the microscope stage (see Section 2.3.).

• 2. A pump for delivering the perfusate into the bath at 1–2 mL/min.

2.3 The Microspectrofluorimeter

A dual-emission microspectrofluorimeter (Fig. 3) is constructed using the following components (3). Figure 4 shows its optical path.

• 1. Microscope: an inverted phase-contrast (Diaphot, Nikon) fitted with epifluorescence using a 100-W Xenon lamp source. A variable aperture, a shutter, and a beam splitter containing a dichroic mirror (455nm) are attached to the sideport of the microscope.

• 2. Interference filters: Excitation filter 340 nm and emission filters (470–490 and 405 nm), in order to filter light transmitted and reflected, respectively, by the dichroic mirror.

• 3. Photomultiplier tubes: Separate photomultiplier tubes (PM28B, Thorn EMI) to record the intensity of fluorescence at the two wavelengths. Single-photon currents in each tube are converted to 5V/25 ns TTL (Transistor-Transistor Logic) pulses.

• 4. Computer-based photon counting system: A dual-photon counter (Newcastle Photometric) to count the TTL pulses, and IBM microcomputer to record photon counts s^-1 from each channel and to calculate and display the ratio of intensities (405/480 nm).

Fig. 3.

The dual-emission microspectrofluorimeter constructed from a Nikon Diaphot inverted phase-contrast microscope. To the sideport is attached a variable aperture, a beam splitter, and a dichroic mirror (455 nm). The two ports of the beam splitter are covered by barrier filters (405 and 480 [470–490] nm) and attached to separate photomultiplier tubes (Thorn EMI) that feed via British National Connectors (BNC) into two separate channels of a dual-photon counting system (Newcastle Photometric Systems), The latter is controlled by an IBM microcomputer. See Fig. 4 for more information.

Fig. 4.

The optical path of the microspectrofluorimeter constructed from the Nikon Diaphot microscope

3 Methods

3.1 Extracellular Calibration

• 1. Determine the dye concentration that will produce the optimal response of your fluonmeter by perfusing the bath with various concentrations (1-20 μM) of indo 1 dissolved in the calibrating solution.

• 2. Construct calibration curves by perfusing the bath with dye-containing solutions of known [Ca²⁺]. Figure 5 shows the output of a micro-spectrofluorimeter as solutions containing indo 1(10 μmol/L) and different [Ca²⁺] (Table 3) are perfused through the experimental chamber. The ordinates show the ratio of intensities of emitted light at 405 and 480 nm and corresponding [Ca²⁺] values.

Fig. 5.

The output of a microspectrofluorimeter, as solutions containing indo 1 (unesterified) (10 μM) at different [Ca²⁺] values (Table 3) are perfused through the experimental chamber The ordinates show the ratio of intensities of the emitted light at 405 and 480 (470–490) nm. The inset shows a standard curve of [Ca²⁺] values plotted against the observed ratio. The different [Ca²⁺] values were, (nM)’ 1 5 (baseline); 187 (step 1); 321 (step 2); 499 (step 3); 749 (step 4); and 1700 (step 5). Temperature 20°C.

3.2 Measurement of[Ca²⁺]_i

Carry out all experiments in a dark room.

• 1. Allow the chosen cells to settle on glass cover slips (22 mm). Load cells with indo 1/AM (to the optimal dye concentration, determined in Section 3.1.) by incubating in serum-free BSS for 5, 10, or 20 min at 37°C. Rinse the cover slips gently three times with BSS, and store in the dark. Carry out morphological assessments and trypan blue exclusion assays to check for cell damage or deterioration (see Notes 6 and 7).

• 2. Transfer cover slips with dye-loaded cells to the bath on the stage of a microspectrofluorimeter.

• 3. Select a cell to he within the aperture of the sideport of the microspectrofluorimeter. Close the diaphragm to approximate the cell margin.

• 4. Switch the photomultiplier tubes on. Each photomultiplier tube has an optimal operating voltage. Check these voltages regularly for best functioning.

• 5. Quantitate the background fluorescence (autofluorescence) of the cover slip by removing the cell from the field and recording the counts per second. The computer takes the average of 10 such readings and subtracts the average from consecutive cell readings.

• 6. Bring the same cell back into the set field, and refocus at the sideport.

• 7. Assess dye loading visually by looking at the fluorescent cell, and then measure the fluorescence signal using the photomultipliers. Counts per second should be >50,000, if reliable estimates are to be made. Lower counts produce a low signal:noise ratio and, hence, less reliable baseline level. The background fluorescence should be <5% of the signal.

• 8. Record the emissions: Computer displays either the fluorescence intensity at 405 or 480 nm, or their ratio, or the calculated value of [Ca²⁺], (4,5). See Figs. 6 and 7 for examples.

• 9. After making baseline recordings over defined times, pass preheated agonist-containing solution into the chamber at a defined flow rate of 1.8 mL/min, and maintain the temperature of the bath at 37°C using a thermo-coupled device; continue to record the emissions.

• 10. At the end of each experiment, expose cells sequentially to 5 μM ionomycin and 1 mM Mn²⁺, recording the emissions after each addition.

Fig. 6.

A trace representing the biphasic effect of asusuberic^1–7-eel calcitonin (eCT; 1 nM) on cytosolic free calcium ([Ca²⁺],) levels in single isolated rat osteoclasts. Panels B and C show the changes in absolute fluorescence (counts s^-1) at sampling wavelengths of 405 and 480 (470–490) nn The cell was exposed to the peptide at zero-time. Medium [Ca²⁺] = 1.25 mM.

Fig. 7.

Cytosolic free calcium ([Ca²⁺]), levels of macrophages during phagocytosis of zymosan particles or during spreading on nonopsonized surface. Panel A represents a [Ca²⁺], trace of a fully spread macrophage not it contact with zymosan. Panel B shows two representative traces showing [Ca²⁺], pulses during the early phase of phagocytosis by macrophages that either contained (right trace) or did not contain (left trace) another zymosan particle. Panel C represents the oscillatory pattern of [Ca²⁺], increment during the spreading of a macrophage in the absence of zymosan. The left lower vertical axis represents the fluorescence ratio (405/480 nm), which corresponds to the logarithmic [Ca²⁺], scale on the right. Reproduced with permission from ref. 5

See Notes 6 and 7 for possible problems and Notes 8–11 for new developments.

4 Notes

• 1. In contrast to indo 1, the excitation spectrum of fura 2 is affected by changes in [Ca²⁺],. Thus, when using fura 2, a wavelength splitter is required at the excitation end. This proves expensive and is technically more complicated than using indo 1, where two photomultiplier tubes measure fluorescence at the emission end.

• 2. Since the excitation spectrum of fura 2 is sensitive to changes in [Ca²⁺],, fast calcium measurements are best made using indo 1, for which excitation filters do not have to be changed in rapid succession.

• 3. The fluorescence quantum efficiency of indo 1 (free anion, 0.38; Ca complex, 0.56) is greater than that of fura 2 (free anion, 0.23; Ca complex, 0.49). Thus, lower concentrations are required to overcome autofluorescence.

• 4. Indo 1 has also been the dye of choice for flow cytometry where limitations in excitation wavelength availability makes the monitoring of emission more practical.

• 5. The K _d of fura 2 is affected by pH, ionic strength, temperature, and viscosity, so that a K _d in one type of cell, or even in vitro, outside the cell may not be valid for other cells

• 6. The cells may load poorly despite prolonged incubation times. This is commonly because of compartmentalization or incomplete intracellular hydrolysis (see Section 1.4.2.). It is also owing to hydrolysis of the ester during storage and because of poor handling procedures. It is recommended that the dye is aliquoted after purchase and frozen at -70°C. In case of poor loading, it is recommended that a fresh batch of dye be used. Unhealthy cells take up less dye, and this can be assessed by the trypan blue.

• 7. Photobleaching is the loss of fluorescence because of long exposures to bright light. In order to overcome this problem, a neutral density filter is often used at the excitation port. Bright UV light can also have adverse effects on cell function. Dyes may also leak, particularly from macrophages. If the counts are high, this will not pose a problem while ratiometnc methods are being used.

• 8. There have been three major areas of development since Tsien’s group first reported the synthesis of fura 2 and indo 1. These include the synthesis of new Ca-sensitive fluorochromes, fluo 3 and rhod 2 (6); the development of fluorochromes for other ions, including H⁺, Mg²⁺, Na⁺, K⁺, and Cl^- (Table 2); and more recently, and perhaps the most revolutionary, the development of a dye for measuring cyclic AMP concentrations in single cells (7).

• 9. Briefly, the properties of fluo 3 and rhod 2 are as follows. Both dyes permit the use of longer wavelengths and can be used in argon lasers. The K _d s are 0.45 μM and 1 μM, respectively, in buffers containing 0.1M KC1, pH 7.0, at 22°C. Their calibration protocols are similar to those described for quin 2 (xc8

• 10. Fluo 3.

  • a. Uses fluorescein filters.

  • b. Nonfluorescent, until it is hydrolyzed within the cell.

  • c. Alkali metal salts of fluo 3 are also nonfluorescent at resting [Ca²⁺].

  • d. Undergo a 40-fold enhancement of fluorescence on binding to Ca²⁺.

  • e. High K _d and thus weaker binding to Ca²⁺, permits the measurement of higher peaks of [Ca²⁺] than is possible with indo 1 or fura 2, making it the dye of choice for kinetic measurements of Ca²⁺ release.

  • f. No spectral shift, precluding its use in ratio imaging.

• 11. Rhod2.

  • a. Uses rhodamine filters.

  • b. Undergoes a 3.4-fold enhancement of fluorescence on binding to Ca²⁺. This lower enhancement is owing to a high quantum yield of its Ca²⁺-free fluorescence.

  • c. Less readily excited by argon lasers.