Flow cytometric analysis of cell suspensions exposed to shock waves in the presence of the radical sensitive dye hydroethidine

Ultrasound

in Med.

Pergamon

& Biol.. Vol. 21, No. 4, pp. 569-577, 1995 Copyright 6 I995 Elsevier Science Ltd Printed in the USA. All rights reserved 0301-5629/95 $9.50 + .OO

0301~5629( 94)00133-2

*Original

Contribution

FLOW CYTOMETRIC ANALYSIS OF CELL SUSPENSIONS EXPOSED TO SHOCK WAVES IN THE PRESENCE OF THE RADICAL SENSITIVE DYE HYDROETHIDINE ELMAR ENDL, FIA STEINBACH

and FERDINAND

HOFSTADTER

Institut fiir Pathologie, Universitgt Regensburg, D-93042 Regensburg, Germany (Received

20 May

1994; in jinal form

22 August

1994)

Abstract-The occurrence of intracelhdarly and extracelhdarly generated free radicals during shock wave exposure on an experimental Siemens Mhotripter was tested with the radical sensitive dyes hydroethid& and dicblorofluoresciu (DCFH). DCFH, a notiuorescent compound, is oxidised to dicldorfluoresceb~ (DCF) by hydrogen peroxide in the presence of peroxidase. DCF green fluorescence intensity was used for fluorescence spectrometric measurement of hydrogen peroxide generated during shock-wave treatment of cellfree dye solutions. The fluorescence intensity of ethidium, the oxidised form of hydroethidine, was used for the flow-cytometric measurement of intracellular oxidkdng reagents present in RT4 tumour cells during shock-wave exposure. Changes in membrane permeability, which influence the intraceMar content of ethidium, were controlled by counterstaining the cells with propidium iodide, an indicator for membrane integrity. We observed no increase in intracellular ethidium fluorescence intensity after shock-wave treatment of single ceil suspensions and therefore no indication for shock-wave-induced intracellular free radicals. Key Wonkr: Shock waves, Lithotripsy, cells, Flow cytometry, HydroeHddine,

Cavitation, Free radicals, Hydrogen peroxide, Cell damage, Tumour Dichlorofluorescin-diacetate.

INTRODUCTION Ultrasound fields of high intensity are assumed to cause various effects on biological material. The widespread use of lithotripters for the disintegration of stones (e.g., Folberth 1989) and the discussion about using their damaging potential in tumour therapy (Dellian et al. 1993; Gamarra et al. 1993a, 1993b) brought about an increasing effort to examine the interaction between shock waves and tissue or cells. The temporal average intensity reached by the sound pulses of a lithotripter is comparable to diagnostic ultrasound, and the occurrence of gross heating is therefore negligible (Coleman and Saunders 1989; Folberth et al. 1992). If continuous heating is excluded, cavitation is thought to be the major cause for biological effects of ultrasound at therapeutic levels (Fischer et al. 1988; Williams et al. 1989; Zeman et al. 1990a, 1990b). Considering the large time intervals between the single pulses ( 1 to 2 Hz in common applications), cavitation is most probably of the transient type (Church 1989). Transient cavitation, however, can result in temperatures Address correspondence to: Elmar End].

up to several thousand degrees and pressures up to several MPa ( Apfel 198 1; Crum 1982; Mason and Lorimer 1988; Neppiras 1980). Therefore, the damaging potential of cavitation is high. The violent collapse of the cavity can also result in the formation of free radicals due to the high temperatures and high pressures reached in the hot spot of the collapse (Gambihler and Delius 1992b; Kondo et al. 1988; Riesz 1990). Radicals of this kind are identical to those produced in water by ionizing radiation (Morgan et al. 1988). They are powerful oxidising agents. Extensive studies on the effects of ionizing radiation on tissue and cells have shown that most of these effects are related to intracellularly generated radicals (Ward 1988). Most of the experiments concerning free radicals generated by a shock-wave device were performed on single cell suspensions. The influence of shock waves on cell survival was tested by treating the cells in the presence and absence of radical scavengers (Gambihler et al. 1992b; Suhr et al. 1991). Apart from this indirect method radical sensitive dyes were used to evaluate the production of free radicals. Evidence of

570

Ultrasound in Medicine and Biology

the intracellular production of free radicals after treatment of cell suspensions with electrohydraulically generated shock waves was published by Suhr et al. ( 1991) . The cells were treated in a solution containing the radical sensitive dye hydroethidine, and the increase of the intracellular fluorescence of the oxidative product of hydroethidine, ethidium, was taken as a measure for intracellularly generated free radicals. Destroyed cells and debris were excluded by measuring the volume by means of electrical resistance pulse sizing. Only intact cells were counted. The decision whether free radicals appear intraor extracellularly is a fundamental problem in this experimental setup. The distinction is necessary since the cell membrane is a primary target of shock waves (Gambihler et al. 1992a; Kondo et al. 1988; Steinbach et al. 1992). This article presents our results of the application of electromagnetically generated shock waves to bladder cancer cells in hydroethidine solution. To determine the influence of an increased permeability of the cell membrane we simultaneously recorded the ability of the cells to exclude propidium iodide. Thus, we could definitely decide whether the oxidation of hydroethidine occurred within an intact cell, or if an alteration of the ethidium fluorescence was a secondary effect due to a shock-wave-induced permeabilisation of the cell membrane.

Volume 21, Number 4, 1995

and Brandt 1965). Dichlorofluorescin (DCFH), the hydrolysed form of DCF-DA, becomes fluorescent (dichlorofluorescein DCF) when treated with hydrogen peroxide in the presence of peroxidase. DCFH-DA (Molecular Probes, Eugene, OR, USA) was stored as a 10 mM stock solution in acetone at -20°C in the dark. For each experiment, a working solution was prepared by dissolving 10 PL of the stock solution in 990 PL PBS. To obtain complete hydrolysis of DCFH-DA to DCFH, 140 ,uL of a O.lN NaOH solution was added to 60 PL of the working solution and incubated for 2 min at room temperature. The solution was neutralized by adding 18 mL PBS and stored in the dark ( Burow and Valet 1987 ) . This corresponds to a final concentration of 300 nM DCFH. Horseradish peroxidase (Sigma, Mtinchen, Germany) was added immediately before shock-wave exposure (20 units/ml). Three milliliters of the dye solution was completely oxidised by adding 10 pg/mL hydrogen peroxide (Merck, Darmstadt, Germany). This 300 nM DCF standard served as a reference to the fluorescence intensity of DCF generated during shock-wave treatment and was measured with each set of samples. The stoichiometric reaction for the catalysed oxidation of the hydrogen donor DCFH by hydrogen peroxide is shown in the equation: Pemxidase

2DCFH + H202 MATERIALS

-+

2DCF + 2Hz0.

AND METHODS

Cell culture The experiments were performed with the human bladder cancer cell line RT4, which represents the phenotype of a well-differentiated Gl papillary carcinoma (Rigby and Franks 1970). Cells were serially passaged as monolayer cultures in RPMI-1640 medium (Biochrom, Berlin, Germany) containing 10% fetal calf serum, 1% penicillin/streptomycin, 1% sodium pyruvate and 1% L-glutamine (all from Gibco, Eggenstein, Germany). The cell culture flasks (Greiner, Frickenhausen, Germany) were incubated in a humidified atmosphere containing 5% carbon dioxide at 37°C. Cells grown to subconfluence were washed with phosphatebuffered saline (PBS, pH 7.4, Biochrom) and harvested by a 3-min treatment with 0.25% trypsin/0.02% EDTA (Gibco) in PBS. After centrifugation, the cells were resuspended in PBS for further processing (Steinbach et al. 1992; Wiirle et al. 1994). Reagents The fluorimetric determination of hydrogen peroxide in ultramicro amounts was performed using a stable nonfluorescent reagent, dichlorofluorescin-diacetate (DCFH-DA) (Black and Brandt 1974; Keston

For 1 mol of hydrogen peroxide, 2 mol of the oxidised form DCF are produced. The amount of hydrogen peroxide per pulse was recalculated with linear regression curve fitting (Sigma Plot, Jandel Scientific, Erkrath, Germany) in a two-parameter display DCF fluorescence intensity versus number of shock waves. Fluorescence intensity was calculated as percentage of fluorescence intensity of the 300 nM DCF standard minus fluorescence intensity of a sham-treated control. Hydroethidine is an uncharged blue fluorescent compound produced by the reduction of ethidium bromide. It is incorporated into viable cells where it is enzymatically dehydrogenated or oxidised by free radicals (Bucana et al. 1986). Due to its cationic nature the product ethidium becomes trapped intracellularly. There it intercalates into DNA showing reddish fluorescence when excited by visible light. Hydroethidine was used primarily as a vital stain (Olive 1989). Its staining properties are known to depend on the metabolic activity of cells (Rothe and Valet 1990). Hydroetbidine (Molecular Probes, Eugene, OR, USA) was dissolved in dimethyl-sulphoxide at a concentration of 10 n&l. Aliquots were stored as a stock solution at -20°C in the dark. For each experiment a 1 mM working solution in PBS was prepared.

Analysis

of cell suspensions

Propidiumiodide (PI) carries two positive charges and is therefore excluded from membrane-intact cells (Waggoner 1990). When the permeability properties of the membrane changes, PI enters the cell and intercalates into DNA where it fluoresces red (Dive and Watson 1990; Macklis and Madison 1990). In this respect, PI is similar to the acid dye trypan blue, which is commonly used in dye exclusion tests to determine cell viability (Ross et al. 1989). A stock solution of PI (Sigma, Mtinchen, Germany) was prepared at a concentration of 10 mg/mL in PBS and stored in the dark at 4°C. For each experiment, a working solution at a concentration of 1 mg/ mL was prepared. All working solutions were freshly prepared and removed after each experiment. Hydroethidine stock solutions show a conversion of the dye into ethidium when stored at 4°C for several days. Conversion can be recognized by an increasing pink shimmer of the dye solution and was controlled by fluorescence spectrometric measurements. Prior to shock-wave treatment RT4 cell suspensions ( lo6 cells/ml in PBS) were stained for 20 min at room temperature with hydroethidine at a final concentration of 10 yM. After shock-wave treatment samples were split up into two aliquots. One aliquot was counterstained with PI at a final concentration of 10 pg/rnL for 5 min. The other sample remained as a reference for pure ethidium fluorescence. All samples were measured within 10 min after shock-wave exposure. Shock-wave

exposure

Shock waves were generated by an electromagnetic shock-wave source kindly provided by Siemens Co., Erlangen, Germany. Source and focusing lens are identical to those used in the commercially available lithotripter Lithostar Plus (Siemens). The maximum positive pressure is 60 MPa, and the corresponding energy density 0.6 mJ/mm2 (pressure maximum). Negative pressures reach up to 10 MPa. The focal area (defined as pressure 6-dB zone) is 4 x 40 mm (lateral x vertical) at maximum setting. The experimental setup is described in detail by Steinbach et al. ( 1992)) Folberth et al. (1992) and Wijrle et al. (1994). Polyethylene vials (70 X 9 mm, Nunc, Wiesbaden, Germany) containing 3 mL of a tumour cell suspension or DCFH dye solution were positioned in the focus of the shock-wave device. The bottom of the vial was placed at the intersection of two crossing laser beams marking the vertical and lateral center of the focal area. Controls were left outside the focal region but within the temperature-controlled water bath. All experiments

0 E. ENDL et al.

571

were carried out at 21°C. Treatment modalities were 250, 500 and 750 pulses with an energy density of 0.6 m.I/mm2 and a repetition frequency of 1 Hz. Flow cytometry

Flow cytometry was accomplished with a Becton Dickinson FACStar”“” dual-argon laser cell sorter. Hydroethidine uptake was examined using UV-light excitation (argon ion laser, emission wavelength 365 nm, output power 200 mW), with emission monitored at 425 nm (bandpass filter, bandwidth 44 nm, detector FL4). A combination of a 575-nm bandpass filter (bandwidth 30 nm, detector FL2) and a 630-nm bandpass filter (bandwidth 22 nm, detector FL3) was used for simultaneous measurement of PI and ethidium fluorescence. The difference in the emission spectra of PI (emission maximum at 623 nm) and ethidium (emission maximum at 610 nm) is small compared to the filter settings listed above (Dive and Watson 1990; Waggoner 1990). The emission of both dyes interferes at the FL2 and FL3 detectors. The ratio of the fluorescence intensity recorded at the FL2 and FL3 detector, however, is different for cells that reveal PI uptake, which adds to the fluorescence of ethidium and for cells that show distinct ethidium fluorescence. According to the different ratios of their fluorescence intensities PI- or ethidium-stained cells could be analysed separately in a two-parameter display. Fluorescence together with forward and 90” scattered light was recorded for 10,000 cells per sample. Data analysis was performed on a Hewlett Packard 9000/340 computer using LYSIS II software (Becton Dickinson, Heidelberg, Germany). Fluorescence

spectroscopy

Fluorimetric measurements were made with a Hitachi F-2000 fluorescence-spectrometer (Colora Messtechnik, Larch, Germany). Spectra were measured in PBS, pH 7.4. Slit setting was 10 nm for excitation and emission. The excitation wavelength was 501 nm and DCF-fluorescence was recorded at 521 nm using quartz cuvettes with lo-mm lightpath. Data analysis

Experiments were performed on n different days. Data are represented as geometric mean values of the n different experiments. Errors are represented as standard deviations. Relative fluorescence intensities were calculated as percentage of the fluorescence intensity of a sham-treated control (Fig. 9). Differences were tested for significance with the two-sided t test. RESULTS Extracellular generated hydrogen peroxide

Fluorescence data from DCFH probes treated with 250, 500 and 750 shock waves are shown in Fig. 1.

572

Ultrasound in Medicine and Biology

Volume 2 I, Number 4, 1995

-2.- -..’ i FSC/Forw.rd scattered light

number of shock waves Fig. 1. Two-parameterdisplay of relative DCF fluorescence versusnumber of appliedshock waves. Fluorescencedata werecalculatedaspercentageof fluorescenceintensity from a 300 nM DCF standardminusfluorescenceintensity of a sham-treatedcontrol (n = 5, mean+- standarddeviation).

Fig. 2. Two-parameterdisplayof forward-scatteredlight intensity versussideward-scattered light intensity (linear amplification, arbitrary units) of the sham-treatedcontrol. Region 1 wasdefinedasthe main populationcontainingmore than 90% of the total numberof events.

The amount of DCF generated during shock-wave treatment was proportional to the number of applied pulses (n = 5, r2 = 0.84). Linear regression analysis revealed the generation of 0.36 ng hydrogen peroxide per pulse.

. .

-

--*y----‘.--:

.--r...*..*

Light scatter properties of shock-wave-treated

s.T-y,

-....

. ,.

..

.

:,,..

..

.:

-i.i

cells

The amount of light scattered in the forward direc-

tion is a function of both volume and refractive index. The intensity of light scattered in an angle of 90“ to the incident light beam depends on the inner structure of cells (Benson et al. 1984; Dubelaar et al. 1987; Loken and Herzenberg 1975). Preparations of RT4 single cell suspensions all yielded scatter diagrams similar to the one illustrated in Figure 2. The main population of cells in the two-parameter display forward-scattered light intensity (FSC ) versus sidewardscattered light intensity (SSC) was denoted as region 1. In contrast to the sham-treated control, shockwave-treated cell suspensions showed two distinct populations in the FSC-SSC dot-plot (see Fig. 3). Cells in region 1 had scatter properties similar to the control, whereas cells in region 2 showed a decrease of FSC signal to 37.6 ? 9.4% compared to the control. On the other hand, the SSC signal increased to 119.3

k 8.0%.

z

.

v

region 1

--ds-w3e-‘)

FSCIForward scattered light

lees

Fig. 3. Two-parameterdisplay of forward-scatteredlight intensity versussidewardscatteredlight intensity (linear amplification, arbitrary units) from RT4 singlecell suspension treatedwith 250 shockwaves.Two subpopulations could be identified by flow cytometry and were denotedas region 1 and region 2.

Analysis

8 autofluorescence Y s s = s1 ;?l,r*\,

of cell

suspensions

0 E. ENDI.

et a/.

573

control sham treated control

& d autofluorescence

FL4Dlydroethidine --

FL2/Ethidium fluorescence intensity

Fig. 4. Hydroethidine fluorescence intensity of RT4 cells stained with 10 PM hydroethidine (control) compared to autofluorescence intensity (logarithmic display, arbitrary units) (upper histogram). FL2 signal intensity illustrates the conversion of hydroethidine into ethidium due to metabolic activity of the cells (lower histogram).

Hydroethidine

and ethidium jluorescence

intensity

The properties of hydroethidine as a vital stain are illustrated in Fig. 4. Dye uptake could be controlled by measuring the increase in FL4 signal intensity (hy-

droethidine emission) compared to autofluorescence intensity. The increase in FL2 signal intensity (ethidium emission) indicated the conversion of hydroethidine into ethidium due to the metabolic activity of the cell. Cells that showed an increase of both hydroethidine and ethidium fluorescence intensity correspond to region 1 in the FSC-SSC dot-plot and were counted as vital. On the contrary, shock-wave-treated samples revealed two subpopulations of cells in the histogram display of the hydroethidine fluorescence intensity. The first subpopulation ( 1) had fluorescence intensities similar to the control, whereas cells out of the second subpopulation (2) showed a diminished fluorescence intensity compared to the sham-treated control. Cells belonging to this second subpopulation corresponded to region 2 in the FSC-SSC dot-plot (see Fig. 5). The existence of two subpopulations in the ethidium histogram display was not as obvious as it was for hydroethidine (see Fig. 6). On first sight, shockwave-treated

ethidium

cells showed

a uniform

distribution

fluorescence intensity, illustrated

in

in Fig. 6.

fluorescence intensity

Fig. 5. Comparison of hydroethidine fluorescence intensity of the RT4 single cell suspension treated with 250 shock waves and a sham-treated control (logarithmic display, arbitrary units). The curve of the shock-wave-treated sample reveals two peaks. Peak ( 1) corresponds to cells out of region 1 in the FSC-SSC dot-plot, whereas peak (2) corresponds to cells out of region 2 (see Fig. 3).

This single peak, however, consisted of two subsets (see Fig. 7). Cells belonging to region 1 in the scatter diagram showed ethidium fluorescence intensities ( 1)

FL2/Ethidium fluorescence intensity Fig. 6. Comparison of ethidium fluorescence intensity of the RT4 single cell suspension treated with 250 shock waves and the sham-treated control (logarithmic display, arbitrary units )

Ultrasound in Medicine and Biology

Volume 21, Number 4, 1995

shock wave treated

FL21 575 nm bandpass

FL2/Ethidium fluorescenceintensity Fig. 7. Histogram

display of ethidium

fluorescence intensity

for a sampletreated with 250 shock waves. Fluorescence data from membraneintact cells belongingto region 1 ( 1) andmembrane-damaged cellsbelongingto region2 (2) were displayed separatelyto illustrate the different fluorescence intensitiesof the two subpopulationswhich form the peak of the shock-wave-treatedsample.

Fig. 8. Two-parameterdisplay of FL2 detectorsignalintensity (575nm bandpass)versusFL3 detectorsignalintensity (630-nm bandpass)from the RT4 single cell suspension treatedwith 250shockwaves.Cellswerestainedwith hydroethidinebeforeandwith PI after shock-wavetreatment.The two main populationscould be identified as cells showing PI uptake and cells having distinct ethidiumfluorescence.

similar to the control, while cells belonging to region 2 exhibited an increased ethidium fluorescence intensity (2) compared to the sham-treated control. Membrane permeability For the evaluation of the intracellular content of ethidium an alteration of the cell membrane permeability must be taken into account. Any increase in membrane permeability was therefore controlled by counterstaining shock-wave-treated cells with PI. In a FL2-FL3 two-parameter display (see Fig. 8) we could discriminate between cells which exhibited PI fluorescence and cells which showed distinct ethidium fluorescence. PI uptake was associated with cells belonging to region 2 in the FSC-SSC diagram. The mean fluorescence intensity of the membrane-intact subpopulation, i.e., PI-negative cells, was slightly but not significantly reduced as compared to the shamtreated control (see Fig. 9). There was a considerable amount of cells which indicated an increase in membrane permeability (PI uptake) after shock-wave treatment. The percentages of PI-positive cells and their dependence on the number of applied pulses is listed in Table 1. If those cells are included in the calculation of the ethidium fluorescence intensity the entire cell population, re-

.-g iri z

90 80 70

j 0

250

500

750

number of shock waves Fig. 9. Mean fluorescenceintensity of ethidiumversusnumber of shockwaves from RT4 cells after shock-wavetreatment. Relative fluorescenceintensitieswere calculated as percentageof the fluorescenceintensity of a sham-treated control. Curves representthe fluorescencedata of the subpopulationof cellswith intact membranes (0) andthe entire cell population(0) illustratedin Figs. 6 and7 (n = 8, mean 2 standarddeviation).

Analysis

of cell suspensions

0 E. ENDL

ef al.

515

Table 1. Flow-cytometric analysis of membrane-damaged (PI positive) cells after shock-wave exposure of single cell suspensions. Number

of shock

PI-positive Results

waves

cells

are represented

as geometric

0

250

500

150

3.9 -f 2.6%

21.2 + 7.4%

38.1 2 2.6%

49.4 t 7.1%

mean -+ standard

deviation.

gardless of their membrane permeability, revealed a significant increase (p < 0.05) in mean fluorescence intensity. DISCUSSION The term “acoustic cavitation” has been used widely to cover a variety of experimental results concerning ultrasound treatment of biological material (Crum 1982; Fischer et al. 1988; Williams et al. 1989; Zeman et al. 1990a; 199Ob). Free radicals formed during transient cavitation have been discussed to contribute to the biological effects of therapeutic ultrasound and lithotripter generated shock waves (Gambihler 1990, 1992a; Kondo et al. 1988; Suhr 1991). Since the diffusion length of free radicals in aqueous solutions is between 0.1 and 1 pm, due to their high reactivity, and the generation is concentrated in the centre of the collapsing cavity (Riesz 1990), a recombination of these radicals to hydrogen peroxide is very likely. We therefore used DCFH, an established test for the measurement of ultramicroquantities of hydrogen peroxide (Black and Brandt 1974; Ferrer et al. 1990; Keston and Brandt 1965) to estimate the amount of hydrogen peroxide generated during shock-wave treatment in cell-free assays. The estimated oxidation of DCFH to DCF suggests the formation of 0.36 ng hydrogen peroxide per pulse in 3 mL of a DCFH dye solution. In this report we concentrated on the detection of intracellularly generated free radicals during shockwave exposure. DCFH was previously used as a radical sensitive dye to analyse respiratory burst activity in phagocytes (Burow and Valet 1987; Rothe and Valet 1990). As fluorescein derivates show a relatively fast efflux, even through intact cell membranes (Dive et al. 1990; Suhr 1991), a quantitative analysis of intracellularly generated free radicals with DCFH is particularly difficult. We therefore used the radically sensitive dye hydroethidine for the flow-cytometric analysis of single cell suspensions after shock-wave treatment. Ethidium, the oxidative product of hydroethidine, becomes trapped inside the cell by intercalation into the DNA (Bucana et al. 1986). It can therefore serve as a measure for the total amount of radicals generated during metabolic processes inside the cell (Olive 1989; Rothe and Valet 1990) as well as for any additional

n = 8

radical species, e.g., due to shock-wave exposure (Suhr et al. 1991). A precise measurement of intracellularly formed ethidium makes it necessary to control the integrity of the cell membrane. Light scatter measurements are a common way in flow cytometry to obtain useful information about changes in cellular and nuclear morphology. Light scattering itself is a complex phenomenon that depends on diffraction, reflection and refraction properties of the cell (Sharpless et al. 1977). Changes in size and extent of the reflecting and absorbing components within the cell will result in the variation of FSC and SSC signal intensity. Diminished FSC intensity can be associated with size, membrane alterations (Weston et al. 1994), loss of optical density (Dubelaar et al. 1987; McGann et al. 1988) and cell death (Loken and Herzenberg 1975; McGann et al. 1988). SSC signal intensity, on the other hand, depends on the granularity of the cell. Intracellular reflective surfaces, especially the nucleus, contribute to SSC signal intensity (Benson et al. 1984). The decreasing FSC intensity illustrated in Fig. 3 suggests changes in cell size and/or loss of intracellular components with higher optical density resulting from an increased membrane permeability. The correlation between FSC intensity and the uptake of the polar dye PI or FSC intensity and the loss of hydroethidine fluorescence intensity supports the suggestion of alterations in membrane permeability. SSC signal intensity, on the other hand, shows a slightly increasing intensity. Changes in membrane folding and nuclear size and shape are known to influence the SSC signal intensity (Benson et al. 1984; McGann et al. 1988). A more precise method of controlling membrane integrity, apart from light scatter measurements, is the use of a dye-exclusion test (Gorer and O’Gorman 1982). Thus, we counterstained the cell samples after shock-wave treatment with PI, a polar and charged fluorescent compound. The counterstaining of hydroethidine-stained cell samples with PI is particularly difficult since PI and the oxidative product, ethidium, have similar emission spectra and are therefore hard to discriminate in a one-parameter display of red fluorescence intensity (FL2 or FL3 signal intensity). In a two-parameter display (FL2 vs. FL3 ), however, we

576

Ultrasound in Medicine and Biology

were able to distinguish between PI positive, i.e., membrane-damaged, and PI negative, i.e., membrane-intact cells (see Fig. 8). As far as intracellularly generated free radicals during shock-wave treatment are concerned, we could not detect any sign of a shock-wave-dependent increase in the ethidium fluorescence intensity of cells that have an intact cell membrane. Regarding the entire cell population, however, the ethidium fluorescence linearly increased with the number of shocks. This suggests that the increase is primarily caused by those cells that were permeable for PI. Cells which were subject to an intracellular conversion of hydroethidine to ethidiume.g., by intracellular cavitation-will have damaged intracellular structures, will appear in region 2 and will be stained by PI. On the other hand, ethidium is a polar molecule with properties similar to PI (Dive et al. 1990). Therefore, the increased ethidium fluorescence in PI-stained cells may be caused by the penetration of extracellular ethidium molecules into the cell rather than by intracellular oxidation. Extracellular ethidium molecules may result from impurities of the dye solution or by oxidation due to extracellular radical formation (Rothe and Valet 1990; Suhr et al. 1991) . A conversion of hydroethidine to ethidium in stock solutions stored at 4°C for several days can be discernible by pure eyesight (increasing pink shimmer). Experiments performed with hydroethidine dye solutions that showed an increased amount of already formed ethidium revealed an up to fivefold increase in ethidium fluorescence intensity of the lysed fraction of cells belonging to region 2 in the FSC-SSC dot-plot (data not shown). The decrease of hydroethidine fluorescence intensity can be interpreted in more technical terms. In a FACStar P’“” flow cytometer the sample stream is injected into a nozzle containing a stream of sheath fluid which consists of an electrolyte with physiologic osmolarity. A laminar stream emerges from this nozzle and is intersected by the laser beam just below the orifice. Throughout this procedure small molecules can be exchanged between the cells and the sheath fluid. An additional effect is the deformation of the cells during the process of hydrodynamic focusing (Kachel 1990). This deformation can accelerate the exchange of intracellular hydroethidine between cells and sheath fluid, especially when the membranes of these cells are lysed. Interpreting the increase of ethidium fluorescence in membrane-damaged cells seems therefore very difficult. The PI-negative (membrane intact) subpopulation, on the other hand, showed a decrease of the ethidium fluorescence intensity after shock-wave treatment.

Volume 21, Number 4, 1995

As mentioned above, hydroethidine is oxidised by radicals that are generated during cellular metabolism. Therefore, a decrease of the oxidative product indicates a reduced metabolic activity. An impaired function of mitochondria, which has been described as a consequence of shock-wave treatment, might account for this phenomenon (Seitz et al. 1993). The results obtained in this study are contradictory to the results by Suhr et al. ( 1991) who found a shockwave-dependent increase of ethidium fluorescence in membrane-intact cells after shock-wave treatment. They excluded membrane damaged cells from analysis by electrical resistance measurement (Coulter volume determination). Since it is possible that this method did not detect the entire fraction of cells that were permeable for polar dyes (Brtimmer et al. 1989; McGann et al. 1988), cells which were penetrated by extracellularly generated ethidium, or ethidium already present in the dye solution, could influence analysis and interpretation. This would explain the discrepancy. Our experiments, however, were performed on a different shock-wave device and with a different cell line, which, of course, limits the comparability. Nevertheless, our results do not support the evidence of intracellularly generated free radicals during shock-wave exposure in the fraction of membrane-intact cells. Acknowledgement-This work was supported by Siemens Medizintechnik, Erlangen, Germany.

REFERENCES Apfel, R. E. Acoustic cavitation. In: Edmonds, P., ed. Methods in experimental physics 19. New York Academic Press; I981 :355411. Benson, M. C.; McDougal, D. C.; Coffey, D. S. The application of perpendicular and forward light scatter to assess nuclear and cellular morphology. Cytometry 5:515-522; 1984. Black, M. J.; Brandt, R. B. Spectrofluorometric analysis of hydrogen peroxide. Anal. Biochem. 58:246-254; 1974. Brtimmer, F.; Brenner, J.; Brauner, T.; Htilser, D. F. Effect of shock waves on suspended and immobilized L1210 cells. Ultrasound Med. Biol. 15:229-239; 1989. Bucana, C.; Saiki, I.; Nayar, R. Uptake and accumulation of the vital dye hydroethidine. J. Histochem. Cytochem. 34:11091115; 1986. Burow, S.; Valet, G. Flow-cytometric characterization of stimulation, free radical formation, peroxidase activity and phagocytosis of human granulocytes with 2,7-dichlorofluorescein (DCF). Eur. J. Cell Biol. 43:128-130; 1987. Church, C. C. A theoretical study of cavitation generated by an extracorporal shock wave lithotripter; J. Acoust. Sot. Am. 86:215-227; 1989. Coleman, A. J.; Saunders, J. A survey of the acoustic output of commercial extracorporeal shock wave lithotripters. Ultrasound Med. Biol. 15:213-227; 1989. Crum, L. A. Acoustic cavitation. Proceedings of the 1982 IEEE Ultrasonic Symposium. New York IEEE 1982:1-l 1. Dellian, M.; Walenta, S.; Gamarra, F.; Kuhnle, G. E. H.; MuellerKlieser, W.; Goetz, A. E. Ischemia and loss of ATP in tumours following treatment with focused high energy shock waves. Br. J. Cancer 68:26-31; 1993. Dive, C.; Watson, J. V.; Workman, P. Multiparametric analysis of

Analysis

of cell suspensions

cell membrane permeability by two colour flow cytometry with complementary fluorescent probes. Cytometry 11:244-252; 1990. Dubelaar, G. B. 1.; Visser, J. W. M.; Donze, M. Anomalous behaviour of forward and perpendicular light scattering of a cyanobacterium owing to intracellular gas vacuoles. Cytometry 8:405412; 1987. Ferrer, A. S.; Santema, J. S.; Hilhorst, R.; Visser, A. J. W. G. Fluorescence detection of enzymatically formed hydrogen peroxide in aqueous solution and in reversed micelles. Anal. Biochem. 187:129-132; 1990. Fischer, N.; Muller, H. M.; Gulhan, A.; Sohn, M.; Deutz, F. J.; Rubben, H.; Lutzeyer, W. Cavitation effects: Possible cause of tissue injury during extracorporal shock wave lithotripsy. J. Endourol. 2:215-220; 1988. Folberth, W. Non-invasive treatment of urinary and biliary stones with extracorporal shockwave lithotripsy. Kemtechnik 53: l-5; 1989. Folberth, W.; Kohler, G.; Rohwedder, A.; Matura, E. Pressure distribution and energy flow in the focal region of two different electromagnetic shock wave sources. J. Stone Dis. 4: l-7; 1992. Gamarra, F.; Spelsberg, F.; Dellian, M.; Goetz, A. E. Complete remission after therapy with extra-corporeally applied high-energy shock waves (HESW). Int. J. Cancer 55:153-156; 1993a. Gamarra, F.; Spelsberg, F.; Kuhnle, G. E. H.; Goetz, A. E. High energy shock waves induce blood flow reduction in turnours. Cancer Res. 53:1590-1595; 1993b. Gambihler, S.; Delius, M.; Brendel, W. Biological effects of shock waves: Cell disruption; viability; and proliferation of L12 10 cells exposed to shock waves in vitro. Ultrasound Med. Biol. 16:587594; 1990. Gambihler, S.; Delius, M. Influence of dissolved and free gases on iodine release and cell killing by shock waves in vitro. Ultrasound Med. Biol. 18:617-623; 1992a. Gambihler, S.; Delius, M. Transient increase in membrane pcrmeability of Ll210 cells upon exposure to lithohipter shock waves in vitro. Naturwissenschaften 79:328-329; 1992b. Gorer, P. A.; O’Gorman, P. The cytotoxic activity of iso-antibodies in mice. Transplant Bull. 3: 142-143; 1982. Kachel, V. Hydrodynamic properties of flow cytometry instruments. In: Melamed, M. R.; Mendelsohn, M. L. eds., Flow cytometry and sorting. New York: Wiley; 1990. Keston, A. S.; Brandt, R. The fluorometric analysis of ultramicro quantities of hydrogen peroxide. Anal. Biochem. 11: l-5; 1965. Kondo, T.; Gamson, J.; Mitchell, J. B.; Riesz, P. Free radical formation and cell lysis induced by ultrasound in the presence of different rare gases. Int. J. Radiat. Biol. 54:955-962; 1988. Loken, M. R.; Herzenberg, L. A. Analysis of cell populations with a fluorescence activated cell sorter. Ann. NY Acad. Sci. 254:163-171; 1975. Macklis, J. D.; Madison, R. D. Progressive incorporation of propidium iodide in cultured mouse neurons correlates with declining electrophysiological status: A fluorescence scale of membrane integrity. J. Neurosci. Meth. 31:43-46; 1990. Mason, T. J.; Lorimer, J. P. Sonochemistry; theory; applications and uses of ultrasound in chemistry. Chichester, U.K.: Wiley; 1988. McGann, L. E.; Walterson, M. L.; Hogg, L. M. Light scattering and

0 E. ENDL

er (II.

571

cell volumes in osmotically stressed and frozen-thawed cells. Cytometry 9:33-38; 1988. Morgan, T. R.; Laudone, V. P.; Heston, D. W.; Z&z, L.; Fair, W. R. Free radical production by high energy shock wavescomparison with ionizing irradiation. J. Urol. 139: 186- 189; 1988. Neppiras, E. A. Acoustic cavitation. Phys. Rep. 61:160-251; 1980. Olive, P. L. Hydroethidine: A fluorescent redox probe for locating hypoxic cells in spheroids and murine tumours. Br. J. Cancer 60:332; 1989. Riez, P. Free radical generation by ultrasound in aqueous solutions of volatile and nonvolatile solutes. In: Mason, T. J., ed. Advances in sonochemistry. London: JAI Press Ltd.; 1990. Rigby, C. C.; Franks, L. M. A human tissue culture cell line from a transitional cell tumour of the urinary bladder: Growth, chromosome pattern and ultrastructure. Br. J. Cancer 24:746-754; 1970. Ross, D. D.; Joneckis, C. C.; Ordonez, J. V.; Sisk, A. M.; Wu, R. K.; Hamburger, A. W.; Nora, R. E. Estimation of cell survival by flow cytometric quantification of fluorescein diacetate/propidium iodide viable cell number. Cancer Res. 49:3776-3782; 1989. Rothe. G.; Valet, G. Flow cytometric analysis of respiratory burst activity in phagocytes with hydroethidine and 2 ‘,7 ‘-dichlorofluorescin. J. Leukocyte Biol. 47:440-448; 1990. Seitz, R.; Seidl, M.; Worle, K.; Steinbach, P.; Hofst;idter, F. The effects of high energy shock waves on cell membranes and mitochondria. Ultrasonics Int. Conf. Proc. 1993643646. Sharpless, T. K.; Bartholdi, M.; Melamed, M. R. Size and refractive index dependence of simple forward angle scattering measurements in a flow system using sharply focused illumination. I. His&hem. Cytochem. 7:845-856; 1977. Steinbach, P.; Hofstadter, F.; Nicolai, H.; Riissler, W.; Wieland, W. In vitro investigations on cellular damage induced by high energy shock waves. Ultrasound Med. Biol. 18:691-699; 1992. Suhr, D.; Brttmmer, F.; Httlser, D. F. Cavitation-generated free radicals during shock wave exposure: investigasons with cell-free solutions and susoended cells. Ultrasound Med. Biol. 17:761768; 1991. Waggoner, A. S. Fluorescent probes for cytometry. In: Melamed, M. R.: Mendelsohn, M. L., ed. Flow cytometry and sorting. New York: Wiley; 1990. Ward, J. F. DNA damage produced by ionizing radiation in mammalian cells: Identities, mechanisms of formation, and reparability. Prog. Nucl. Acid Res. Molec. Biol. 35:95- 123; 1988. Weston, K. M.; Alsalami, M.; Raison, R. L. Cell membrane changes induced by the cytolytic peptide, melittin, are detectable by 90” laser scatter. Cytometry 15: 141- 147; 1994. Williams, A. R.; Delius, M.; Miller, D. L.; Schwarze. W. Investigation of cavitation in flowing media by lithotripter shock waves both in vitro and in vivo. Ultrasound Med. Biol. 15:53-60; 1989. Worle, K.; Steinbach, P.; Hofstsdter, F. The combined effects of high-energy shock waves and cytostatic drugs or cytokines on human bladder cancer cells. Br. J. Cancer 69:58-65; 1994. Zeman, R. K.; Davros, W. J.; Goldberg, J.; Garra, B. S.; Hayes, W. S.; Cattau, E. L.; Horri, S. C.; Cooper, C. J.; Silverman, P. M. Cavitation effects during litbotripsy: Part II. Clinical observations. Radiology 177: 163- 166; 1990a. Zeman. R. K.; Divros. W. J.; Goldberg, J.; Garra, B. S.; Horri, S. C. Cavitation effects during lithotripsy: Part I. Results of in vitro experiments. Radiology 177: 157- 161; 1990b.