Kinetic Measurements Using Flow Cytometry: New Methods for Monitoring Intracellular Processes

Abstract

The aim of our work was to establish flow cytometry methods for the characterization of mitochondrial Ca²⁺ levels, plasma membrane potential, and superoxide generation and to relate kinetics to that of cytoplasmic Ca²⁺ levels during short-term activation of T-lymphocytes. We monitored the change of fluorescence absorbance of sequentially measured Jurkat cells for 12 min. The cells were stained with the fluorescent dyes Fluo3-AM, Rhod2/AM, di-BA-C4-(5), or dihydroethidium and then were stimulated with increasing doses of phytohemagglutinin (PHA) or were treated with rotenone. Double-logistic function was fitted to cytoplasmic Ca²⁺ signal and mitochondrial Ca²⁺ levels, whereas logistic function was fitted to plasma membrane potential and superoxide levels. The calculated function parameters were area under the curve (AUC), maximum (Max), time to reach maximum (t_max ), slope at the first 50% value of Max (Slope), and ending (End) values, respectively. We found significant dose–response relationship between PHA dose and cytoplasmic Ca²⁺ signals (AUC, Max, Slope: P<0.05), mitochondrial Ca²⁺ levels (AUC and Max: P<0.05), and plasma membrane potential (AUC and End values: P<0.05). In rotenone-treated cells, superoxide generation increased in a dose-dependent manner (P<0.05 for AUC and End values, respectively). The present methodology provides an opportunity for monitoring and characterizing mitochondrial Ca²⁺ levels, plasma membrane potential, and superoxide generation in PHA-activated or rotenone-treated Jurkat cells with flow cytometry.

Introduction

Flow cytometry provides an opportunity for determining fluorescence emission intensities within sequentially measured cells. This approach may give an insight into kinetic alterations of the intracellular analytes tested.

For the identification of intracellular analytes, specific fluorochromes are used. These include Fluo-3-AM, Rhod2/AM, bis-(1,3-dibutylbarbituric acid)pentamethine oxonol [di-BA-C4-(5)], and dihydroethidium (DHE) for the detection of cytoplasmic Ca²⁺, mitochondrial Ca²⁺, plasma membrane potential (Δψ), and superoxide (O₂ ⁻), respectively. They are routinely used in fluorescent microscopy and fluorimetry. For flow cytometry, their use is rather limited to determine intracellular conditions at a given time point only; their application for kinetic analysis is restricted because of the absence of standardized approaches to evaluate sequentially obtained flow cytometry data.

Recently, our team introduced a novel algorithm to characterize cytoplasmic Ca²⁺ signal after short-term activation of lymphocytes that allows a description of kinetic changes in an objective manner.¹ Based on this approach, we developed flow cytometry methods that may be useful for characterization of other intracellular processes such as mitochondrial Ca²⁺ level, alteration of plasma membrane potential, and generation of reactive oxygen species. We established the optimal conditions for each analyte in Jurkat cells, a T-lymphocyte cell line. In this article, we present the methodology along with the results of some characteristic experiments.

Materials and Methods

Cell Cultures and Viability Assay

Jurkat leukemic T-cell line (generous gift from E. Buzás, Department of Genetics, Cell- and Immunobiology, Semmelweis University, Budapest, Hungary) was cultured in a complete medium (RPMI 1640 with 10% heat inactivated fetal calf serum [FCS] and 2 mM l-glutamine; all from Sigma–Aldrich, St. Louis, MO) and kept at 37°C humidified atmosphere (5% CO₂ in air). The cells were collected during the log phase of growth and washed in phosphate-buffered saline (2 mM KH₂PO₄, 9.5 mM Na₂HPO₄·2H₂O, and 136.7 mM NaCl). The viability of cells was assessed by trypan blue exclusion (viability of at least 90% was required before each experiment was started).

The cells were suspended in modified RPMI medium and analyzed by flow cytometry. The Ca²⁺ concentration of the modified RPMI medium was set to 2 mM by the addition of crystalline CaCl₂.

Solutions and Reagents

Fluo-3-AM, di-BA-C4-(5), Rhod2/AM, and DHE dyes as well as Pluronic F-127 were purchased from Molecular Probes (Karlsbad, CA) and they were freshly dissolved in dimethyl sulfoxide or ethanol according to the manufacturer's protocol before the measurements. Phytohemagglutinin (PHA), rotenone, RPMI-1640, l-glutamine, and FCS were purchased from Sigma–Aldrich.

Equipment

All flow cytometric measurements were performed on a BD FACSAria flow cytometer (BD Biosciences, San Jose, CA) equipped with 488- and 633-nm excitation lasers and data were processed using the FACSDiVa software.

Software developed at our laboratory was used for the evaluation of data from kinetic flow cytometry measurements. At first, Jurkat cells were gated according to forward scatter and side scatter characteristics (Fig. 1A), and then the gated events were plotted with the measured fluorescent channel against time (Fig. 1B).

Fig. 1.

(A) The population of Jurkat cells was gated according to FSC and SSC. (B) Dot plot acquired from a kinetic flow cytometry measurement. The subpopulation (gated in A) used for the calculation of functions was gated from the Jurkat population according to one fluorescence channel (in this case, Fluo-3-AM; band pass filter: 530/30) and the time. FSC, forward scatter characteristic; SSC, side scatter characteristic.

Simultaneous Measurement of Cytoplasmic Ca²⁺ Signal and Plasma Membrane Potential

Cytoplasmic free Ca²⁺ level was detected by loading the cells with Fluo-3-AM (emission maximum, 526 nm, recorded with a 530/30 nm band pass [BP] filter). The cells were loaded with 2.6 μM Fluo-3-AM and 0.02% Pluronic F-127 for 20 min at 37°C. The cells were washed once before measurements.² At the end of the measurements, ionomycin was added to the samples to verify the specificity of the signal.

For the analysis of membrane potential, the fluorescent dye oxonol di-BA-C4-(5) was used (emission maximum, 616 nm, recorded with 610/20 nm BP filter). The cells were loaded with 300 nM di-BA-C4-(5) for 8 min at 37°C.³ Positive control at the end of the measurements is not necessary as this dye is not compartmentalized and not extruded by the glycoprotein efflux pump either³ (for details, see Table 1).

Table 1.

Simultaneous Measurement of Cytoplasmic Ca²⁺ Signal and Plasma Membrane Potential

Step	Parameter	Value	Description
1	Cell solution	0.5 mL	1–2×10⁶ Jurkat cells
2	Loading with Fluo-3-AM and Pluronic F-127	5 μL	20 min at 37°C in dark
3	Washing	400 g	6 min suspending in modified RPMI 1640 medium
4	Loading with di-BA-C4-(5)	1.3 μL	8 min at 37°C in dark
5	Measurement
6	Activation process	0, 2.5, 5, 10, and 15 μg/mL in final concentration	Phytohemagglutinin
7	Measurement	12 min	37°C in dark
8	Signal detection:	Band pass filters:	BD FACSAria
	Fluo-3-AM	530/30 nm	Excitation with 488 nm blue laser line
	di-BA-C4-(5)	610/20 nm
9	Verifying	1 μL	Ionomycin

Step Notes

1. Jurkat cells in modified RPMI 1640 medium with 10% heat inactivated fetal calf serum, 2 mM l-glutamine, and 2 mM crystalline CaCl₂

2. Fluo-3-AM 2.6 μM and Pluronic F-127 0.02% in final concentration; 10 μg Fluo-3-AM dissolved in 1 μL DMSO+4 μL 10% Pluronic F-127

4. di-BA-C4-(5) 300 nM in final concentration; 115 μM stock concentration in DMSO

6. 1 mg/mL stock concentration of phytohemagglutinin

8. Compensation was 6.56% for Fluo-3-AM–di-BA-C4-(5) and 17.69% for di-BA-C4-(5)–Fluo-3-AM

9. 1 mg/mL stock concentration of ionomycin

DMSO, dimethyl sulfoxide.

Compensation was 6.56% for Fluo-3-AM–di-BA-C4-(5) and 17.69% for di-BA-C4-(5)–Fluo-3-AM.

Measurement of Superoxide Generation

DHE, a redox-sensitive probe, reacts with superoxide and results in the formation of a fluorescent dye (emission maximum, 605 nm, recorded with 610/20 nm BP filter). The cells were loaded with 1 μM DHE for 18 min at 37°C⁴ (for details, see Table 2).

Table 2.

Measurement of Superoxide Generation

Step	Parameter	Value	Description
1	Cell solution	0.5 mL	1–2×10⁶ Jurkat cells
2	Loading with dihydroethidium	1 μL	18 min at 37°C in dark
3	Measurement
4	Activation or inhibition process	0, 2.5, 5, 10, and 15 μg/mL or 0.002, 0.02, 0.2, 2, 20, and 200 μg/mL in final concentration	Phytohemagglutinin or rotenone
5	Measurement	12 min	37°C in dark
6	Signal detection	Band pass filter: 610/20 nm	BD FACSAria Excitation with 488 nm blue laser line

Step Notes

1. Jurkat cells in modified RPMI 1640 medium with 10% heat inactivated fetal calf serum, 2 mM l-glutamine, and 2 mM crystalline CaCl₂

2. 500 μM stock concentration of dihydroethidium in DMSO

4. 1 mg/mL stock concentration of phytohemagglutinin

Measurement of Mitochondrial Ca²⁺ Level

Mitochondrial Ca²⁺ level was monitored using Rhod2/AM (emission maximum, 581 nm, recorded with a 575/26 nm BP filter). The cells were loaded with 2.5 μM Rhod2/AM and 0.02% Pluronic F-127 for 20 min at 30°C and washed once before measurements.⁵ At the end of the measurements, carbonylcyanide-4-(trifluoromethoxy)-phenylhydrazone was added to the sample to verify the specificity of the signal (for details, see Table 3).

Table 3.

Measurement of Mitochondrial Ca²⁺ Level

Step	Parameter	Value	Description
1	Cell solution	0.5 mL	1–2×10⁶ Jurkat cells
2	Loading with Rhod2/AM and Pluronic F-127	5.25 μL	20 min at 30°C in dark
3	Washing	400 g	6 min suspending in modified RPMI 1640 medium
4	Measurement
5	Activation process	0, 2.5, 5, 10, and 15 μL	Phytohemagglutinin
6	Measurement	12 min	30°C in dark
7	Signal detection	Band pass filter: 575/26 nm	BD FACSAria Excitation with 488 nm blue laser line
8	Verifying	1 μL	FCCP

Step Notes

1. Jurkat cells in modified RPMI 1640 medium with 10% heat inactivated fetal calf serum, 2 mM l-glutamine, and 2 mM crystalline CaCl₂

2. 2.5 μM Rhod2/AM and 0.02% Pluronic F-127 in final concentration; 1.25 μL of 1 mM stock concentration of Rhod2/AM in DMSO+4 μL of 10% Pluronic F-127

5. 1 mg/mL stock concentration of phytohemagglutinin; 1 mM stock concentration of FCCP

FCCP, carbonylcyanide-4-(trifluoromethoxy)-phenylhydrazone.

Cell Activation and Mitochondrial Inhibition

At the beginning of measurements, different amounts of PHA (0, 2.5, 5, 10, and 15 μg/mL in final concentration) or rotenone (0.002, 0.02, 0.2, 2, 20, and 200 μg/mL in final concentration) were added to the cell suspension. At each experiment, fluorescence emission of sequentially measured cells was monitored for 12 min. In average, about 1×10⁶ cells were measured. Five independent measurements were made for each assay condition.

Statistical Analysis

We fitted the following functions on median fluorescence values obtained from different types of experiments.

1. Logistic function: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6} \begin{document} \begin{align*}{y} = {y} _0 + \frac {y_2 - y _0} {1 + \ exp \left( ( - x + {x_1}) \times 4 \times \frac{m}{y_2 - {y}_0} \right) }\end{align*} \end{document}

where y ₀, y ₂, x ₁, m ≥ 0 and y ₀ < y ₂.

2. Double-logistic function: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6} \begin{document} \begin{align*}{y} = {y}_0 + \frac {y_1 -y_0} {1 + \left( \frac {x_1 - x} {x_d0} \right) \frac {4 \times {x_d0} \times {m_0}} {y_1 -y_0}} \end{align*} \end{document}

if x<x ₁ \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6} \begin{document} \begin{align*} y = y_2 + \frac {y_1 - y_2} {1 + \left (\frac {x - x_1} {x_d2} \right) \frac {4 \times {x_d2} \times {m_2}} {y_2 - y_1}} \end{align*} \end{document}

otherwise,

where y ₀, y ₁, y ₂, m ₀, x ₁, x _d0, x _d2≥0 and m ₂≤0 and x _d0≤x ₁ and y ₁>y ₀ and y ₁>y ₂.

The logistic function was used for the characterization of continuously increasing fluorescence values, whereas the double-logistic function was used to describe measurements that have an increasing phase, a peak, and a decreasing phase as time passes. The parameters used to describe these functions are area under the curve (AUC) and ending (End) value at 12 min for the logistic function (calculated from y ₀, y ₂, x ₁, and m), and AUC, maximum value (Max), time to reach maximum (t _max), and the slope at the first 50% value of Max (Slope) for the double-logistic function (calculated from y ₀, y ₁, y ₂, m ₀, x ₁, x _d0, and x _d2) (Fig. 2).¹ Logistics and double-logistic functions were fitted to the data, and then the goodness of fit was evaluated using 10-fold cross validation approach; the function with the better fit was used later on.

Fig. 2.

Parameter values calculated for kinetic measurements. Max, maximum value (X); t _max, time to reach maximum value (Y); AUC, area under the curve; End, ending value.

Polynomial contrast tests were used to detect trends of changes in the parameters of functions after treatment of cells performed with different doses of PHA or rotenone. When necessary, values were log-transformed in order to reach normal distribution. Two-tailed P values of <0.05 were considered significant. We report here nominal P values without adjustment for multiplicity. The study was not adjusted for the multitude of statistical tests we performed, and thus, some significance could occur by chance alone.

Student's t-test was used to test the difference between untreated and treated samples. Statistics were calculated using the R software (R Development Core Team, R Foundation for Statistical Computing, Vienna, Austria).

Results

First, we stimulated Jurkat cells with increasing doses of PHA (an aspecific T-lymphocyte activator) and monitored the fluorescence emission of sequentially measured cells for 12 min.

For cytoplasmic Ca²⁺ signal and mitochondrial Ca²⁺ level measurements, double-logistic functions resulted in the best fit. Trend analysis revealed a significant relationship between AUC, Max, and Slope parameters, respectively, and PHA dose (Table 4A, B). Even at the lowest PHA doses, significant alterations in cytoplasmic Ca²⁺ and mitochondrial Ca²⁺ kinetics were observed in both cases in AUC and Max compared with untreated cells.

Table 4.

Experiments with Phytohemagglutinin

	Function Parameters
PHA Concentration (μg/μL)	AUC (U)	Slope	Max (rpv)	t _max (s)
(A) Cytoplasmic Ca²⁺ Signal (Fitted Function: Double Logistic)
0	637.1 (531.4–742.7)	0.161 (0.005–0.993)	1.094 (0.816–1.373)	45.5 (11.4–184.96)
2.5	959.5^* (853.8–1065.1)	0.006 (0.002–0.020)	1.626^* (1.347–1.904)	286.2^* (146.7–425.7)
5	1390.8^* (1285.1–1496.4)	0.016 (0.005–0.050)	2.414^* (2.135–2.693)	267.4 (127.9–406.9)
10	1500.4^* (1394.8–1606.0)	0.022^* (0.008–0.090)	2.593^* (2.314–2.871)	177.9 (38.4–317.3)
15	1506.8 (1401.2–1612.5)	0.025 (0.007–0.080)	2.658^* (2.379–2.937)	125.0 (14.5–264.5)
Trend analysis (P values)	l<0.01	q<0.01	l<0.01	NS
(B) Mitochondrial Ca²⁺ Level (Fitted Function: Double Logistic)
0	622.95 (566.64–679.26)	0.289 (−0.013–0.591)	0.911 (0.772–1.050)	241.7 (109.8–532.0)
2.5	703.11^* (641.42–764.79)	0.175 (−0.128–0.477)	1.049^* (0.910–1.188)	149.8 (68.1–329.6)
5	803.23^* (741.55–864.92)	0.027 (−0.275–0.329)	1.192^* (1.053–1.331)	159.9 (72.7–351.9)
10	860.47^* (798.78–922.15)	0.005 (−0.298–0.307)	1.358^* (1.219–1.497)	206.2 (93.7–453.7)
15	938.72^* (869.75–1007.7)	0.006 (−0.325–0.337)	1.521^* (1.368–1.673)	199.3 (84.0–472.9)
Trend analysis (P values)	l<0.01	NS	l<0.01	NS

	Function Parameters
PHA Concentration (μg/μL)	AUC (U)	End (rpv)
(C) Plasma Membrane Potential (Δψ) (Fitted Function: Logistic)
0	843.0 (675.6–1010.4)	1.341 (1.102–1.580)
2.5	1084.8 (917.4–1252.2)	1.723 (1.491–1.968)
5	1157.4^* (990.0–1324.8)	1.561 (1.323–1.800)
10	1159.9^* (992.5–1327.3)	1.747^* (1.508–1.985)
15	1286.5^* (1119.1–1453.0)	1.907^* (1.668–2.146)
Trend analysis (P values)	q<0.01	l=0.02
(D) Superoxide Generation (O₂ ⁻) (Fitted Function: Logistic)
0	875.6 (841.3–909.9)	2.090 (1.560–2.621)
2.5	867.0 (832.7–901.3)	1.945 (1.415–2.475)
5	872.4 (838.0–906.7)	1.878 (1.348–2.408)
10	884.4 (850.1–918.8)	1.955 (1.425–2.486)
15	881.3 (847.0–915.7)	2.217 (1.687–2.747)
Trend analysis (P values)	NS	NS

Effects of PHA stimulation in Jurkat cells on fluorescence intensity of dyes specific for (A) cytoplasmic calcium, (B) mitochondrial calcium, (C) plasma membrane potential, and (D) superoxide, respectively. The measured parameter values are AUC (area under the curve in units [U]), Max (maximum value in relative parameter value [rpv]), t _max (time to reach maximum value in seconds [s]), slope at the first 50% value of Max (Slope), and ending value (End). l=linear trend; q=quadratic trend. Mean values (95% confidence intervals) are shown.

PHA-treated samples were compared with samples without treatment (P<0.05).

PHA, phytohemagglutinin; NS, not significant.

Plasma membrane potential was described by logistic functions. Trend analysis detected a significant association between AUC and End value parameters and PHA dose. Higher AUC and End values were measured in cells treated with ≥5 μg/mL PHA compared with untreated cells (Table 4C).

No effect of PHA on superoxide generation was observed (Table 4D). Therefore, to determine the suitability of our method for the monitoring of superoxide generation, we used rotenone, a known inhibitor of complex I in mitochondria. In this case, the logistic function was fitted to the data. Trend analysis detected a significant association between AUC and End value parameters and rotenone dose. Rotenone increased the superoxide generation at the 2 μg/mL dose as it is reflected by an increase of AUC and End values compared with untreated cells (Table 5).

Table 5.

Experiments with Rotenone

Superoxide Generation (O₂ ⁻) with Rotenone Treatment (Fitted Function: Logistic)
	Function Parameters
Rotenone Concentration (μg/μL)	AUC (U)	End (rpv)
0	583.6 (580.9–607.0)	1.192 (1.185–1.266)
0.002	576.6 (573.4–577.1)	1.150 (1.140–1.660)
0.02	593.6 (585.4–594.3)	1.206 (1.199–1.216)
0.2	595.9 (591.2–597.7)	1.308^* (1.226–1.339)
2	623.2^* (597.7–633.7)	1.308^* (1.226–1.339)
20	619.9^* (617.9–629.8)	1.295^* (1.292–1.325)
200	654.9^* (652. 9–665.9)	1.464^* (1.425–1.469)
Trend analysis (P values)	l<0.01	l<0.01

Effects of rotenone on superoxide generation (O₂ ⁻) parameter values (area under the curve [AUC] in units [U] and ending value [End] in relative parameter value [rpv]) in Jurkat cells. l=linear trend. Mean values (95% confidence intervals) are shown.

Rotenone-treated samples were compared with samples without treatment (P<0.05).

Discussion

In this work, we established flow cytometry methods that enable the users to monitor intracellular processes related to short-term T-lymphocyte activation or free-radical generation. We optimized incubation times, temperature, and dye concentrations for Jurkat cells. However, one should emphasize that these experimental conditions may be different for other cell types.

We demonstrated that the cytoplasmic Ca²⁺ signal, the kinetics of membrane potential, mitochondrial Ca²⁺ levels, and superoxide generation can also be successfully analyzed using flow cytometry.

Our findings demonstrate a significant difference in the kinetics of measured parameters between untreated cells and cells exposed to the lowest PHA (2.5 μg/mL) doses (generally, published experiments apply a 2–10-fold higher PHA dose to evoke T-lymphocyte activation).⁶ This might indicate that our system is sensitive even to these low PHA levels. However, direct comparisons of our method with other fluorescence techniques are required to establish whether it is more sensitive than traditional methods.

As an immediate response to T-lymphocyte activation, cytoplasmic Ca²⁺ levels immediately increase and then, after a peak, decrease (Fig. 1A). Major mechanisms contributing to this process are the endoplasmic reticulum (ER) calcium release, the opening of calcium-release activated calcium channels (responsible for Ca²⁺ influx), opening of Kv1.3 and IKCa1 potassium channels (responsible for the maintenance of electrochemical potential gradient), and different Ca²⁺ clearance mechanisms (such as plasma membrane Ca²⁺ ATPase and sarco-endoplasmic reticulum Ca²⁺ ATPase).⁷ These processes can be tested by specific inhibitors.⁸

T-lymphocyte activation triggers a transient increase in mitochondrial Ca²⁺ levels.⁹ The kinetics observed in our system are comparable to those reported with other methods in other cell types.^10,11 Of note, mitochondria act as reservoirs for Ca²⁺ and may contribute to Ca²⁺ clearance from the cytoplasm.¹² Indeed, as it is supported by our observations summarized by Fig. 3A and B, mitochondria immediately take up Ca²⁺ released upon activation, as no time lag between maximum Ca²⁺ levels of the cytoplasm and mitochondria was observed (Fig. 3, on the function of cytoplasmic Ca²⁺ and mitochondrial Ca²⁺ the parameters t _max are comparable). Mitochondria contain calcium at micromolar concentrations; however, the major fraction of Ca²⁺ is complexed with phosphate and ATP or bound to matrix proteins. Therefore, under resting conditions the concentration of ionized calcium in the matrix is comparable to that in the cytoplasm (∼0.1 μM). For entering the mitochondrial matrix, Ca²⁺ has to penetrate the outer and inner mitochondrial membranes. The amount of Ca²⁺ stored in mitochondria is relatively low compared with that of the cytoplasm and the ER; therefore, any change in mitochondrial Ca²⁺ stores upon cell activation is more rapidly reflected. In addition, the fast decrease of mitochondrial Ca²⁺ levels following the peak supports recent observations that mitochondrial Ca²⁺ stores may be directly connected to other intracellular stores (i.e., those in the ER).¹²

Fig. 3.

Effect of different doses (at the beginning of the measurement 0, 2.5, 5, 10, and 15 μg/mL PHA doses were added) of PHA on (A) cytoplasmic calcium level, (B) mitochondrial calcium level, (C) plasma membrane potential, and (D) superoxide generation. The sample was monitored for 12 min. PHA, phytohemagglutinin. Color images available online at www.liebertonline.com/adt

Simultaneously with Ca²⁺ level fluctuations, marked alteration in transmembrane potential can be also observed (Fig. 3C), as fluorescence values of di-BA-C4-(5) are steadily increasing. (Note that because of the physical properties of this dye the decreasing potential difference produces increasing fluorescence emission.) Earlier it was found that T-lymphocyte membranes were depolarized within 10 min following exposure to phytohaemagglutinin.¹³ These results obtained by direct measurement of membrane potential with implanted microelectrodes are reinforced by our flow cytometry analysis. In our experiments, characteristics of functions fitted to transmembrane potential and Ca²⁺ signal in stimulated cells were markedly different (as logistic and double-logistic functions were fitted to transmembrane potential and Ca²⁺ signals, respectively). This is probably due to the fact that, besides intracellular Ca²⁺ levels, other factors (opening/closure of transmembrane cationic channels) may also regulate the transmembrane potential.

We also measured superoxide generation in Jurkat cells. Our experiments indicate no marked superoxide production in Jurkat cells during the first 12 min following activation with PHA (Fig. 3D). (This finding does not exclude an increase in superoxide generation in a longer term as was reported by others.⁴) Therefore, to develop a flow cytometry approach to monitor superoxide generation, we used rotenone, a known inhibitor of mitochondrial complex I. Our results obtained in Jurkat cells indicate that almost immediately after rotenone treatment there is a sudden increase in superoxide generation (Fig. 4). The kinetics of this process can be characterized by a logistic function without a definitive peak within 12 min, indicating that the effect of rotenone is not efficiently counterbalanced in the cell. (Of note, rotenone also induces apoptosis and this property is used in general to investigate apoptosis.¹⁴

Fig. 4.

Effect of different doses (0.002, 0.02, 0.2, 2, 20, and 200 μg/mL) of rotenone on superoxide generation. The sample was monitored for 12 min. Color images available online at www.liebertonline.com/adt

A clear advantage of flow cytometry measurement over single-cell measurements is that flow cytometry can efficiently handle a large number of different cell types simultaneously.¹⁵ An important advantage of flow cytometry compared with other fluorescence measurement technologies (e.g., spectrofluorometric cuvette–based approaches) is that flow cytometry is able to measure mixed cell populations simultaneously (e.g., peripheral blood mononuclear cells), identified with fluorescent antibodies. Our experiments clearly demonstrate the suitability of our methodology to monitor and describe intracellular processes in activated T-lymphocytes. Previous cytometric methods evaluated intracellular processes (alteration of membrane potential, mitochondrial Ca²⁺, and superoxide generation) mostly in a static phase; therefore, they did not exploit the possible benefit provided by the large number of sequentially measured cells. In addition, the majority of published experiments with flow cytometry was performed at some selected time points and, therefore, could have been biased by the possibility that a suboptimal time point was used for determinations.

Further, our technique also enables users to monitor and describe intracellular processes with fitting functions. Our approach also provides an opportunity to compare individual measurements in terms of specific parameters such as AUC, Max, t _max, Slope, and End.

In this work, we developed our method for Jurkat cells. Presumably, this technique may also be used for other cell types that present further perspectives for research with flow cytometry.

Footnotes

Acknowledgment

The authors thank Dr. Gergő Szanda (Semmelweis University—Institute of Physiology, Budapest) for his technical advice. This study was funded by grants OTKA 76316 and ETT05-180/2009.

Author Disclosure Statement

No competing financial interests exist.

Abbreviations

References

Kaposi

, Veress

, Vásárhelyi

, Macardle

, Bailey

, Tulassay

et al. Cytometry-acquired calcium-flux data analysis in activated lymphocytes. Cytometry A, 2008; 73A:246–253.

June

, Rabinovitch

. Flow cytometric measurement of intracellular ionized calcium in single cells with indo-1 and fluo-3. Methods Cell Biol, 1990; 33:37–58.

Mello de Queiroz

, Ponte

, Bonomo

, Vianna-Jorge

, Suarez-Kurtz

. Study of membrane potential in T lymphocytes subpopulations using flow cytometry. BMC Immunol, 2008; 9:63.

Nagy

, Koncz

, Perl

. T cell activation-induced mitochondrial hyperpolarization is mediated by Ca2+- and redox-dependent production of nitric oxide. J Immunol, 2003; 171:5188–5197.

Hajnóczky

, Robb-Gaspers

, Seitz

, Thomas

. Decoding of cytosolic calcium oscillations in the mitochondria. Cell, 1995; 82:415–424.

Launay

, Cheng

, Srivatsan

, Penner

, Fleig

, Kinet

. TRPM4 regulates calcium oscillations after T cell activation. Science, 2004; 306:1374–1377.

Lewis

. Calcium signaling mechanisms in T lymphocytes. Annu Rev Immunol, 2001; 19:497–521.

Toldi

, Vásárhelyi

, Kaposi

, Mészáros

, Pánczél

, Hosszufalusi

et al. Lymphocyte activation in type 1 diabetes mellitus: the increased significance of Kv1.3 potassium channels. Immunol Lett, 2010; 133:35–41.

Quintana

, Schwindling

, Wenning

, Becherer

, Rettig

, Schwarz

, Hoth

. T cell activation requires mitochondrial translocation to the immunological synapse. Proc Natl Acad Sci USA, 2007; 104:14418–14423.

10.

Babcock

, Herrington

, Goodwin

, Park

, Hille

. Mitochondrial participation in the intracellular Ca2+ network. Cell Biol, 1997; 136:833–844.

11.

Szanda

, Koncz

, Várnai

, Spät

. Mitochondrial Ca2+ uptake with and without the formation of high-Ca2+ microdomains. Cell Calcium, 2006; 40:527–537.

12.

Duchen

. Mitochondria in health and disease: perspectives on a new mitochondrial biology. Mol Aspects Med, 2004; 25:365–451.

13.

Taki

. Studies on blastogenesis of human lymphocytes by phytohemagglutinin, with special reference to changes of membrane potential during blastoid transformation. Mie Med J, 1970; 19:245.

14.

de Lima

, Yamada

, da Costa

, Pessoa Cdo

, Rabenhorst

, Bahia Mde

, Cardoso

, Santos

, Smith Mde

, Burbano

. Genotoxic effects of rotenone on cultured lymphocytes. Genet Mol Res, 2005; 4:822–831.

15.

Baumgarth

, Roederer

. A practical approach to multicolor flow cytometry for immunophenotyping. J Immunol Methods, 2000; 243:77–97.

Kinetic Measurements Using Flow Cytometry: New Methods for Monitoring Intracellular Processes

Abstract

Introduction

Materials and Methods

Cell Cultures and Viability Assay

Solutions and Reagents

Equipment

Simultaneous Measurement of Cytoplasmic Ca2+ Signal and Plasma Membrane Potential

Step Notes

Measurement of Superoxide Generation

Step Notes

Measurement of Mitochondrial Ca2+ Level

Step Notes

Cell Activation and Mitochondrial Inhibition

Statistical Analysis

Results

Discussion

Footnotes

Acknowledgment

Author Disclosure Statement

Abbreviations

References

Simultaneous Measurement of Cytoplasmic Ca²⁺ Signal and Plasma Membrane Potential

Measurement of Mitochondrial Ca²⁺ Level