0 1994 Wiley-Liss, Inc. Cytometry 17:173-178 (1994)

BRIEF REPORT

Flow Cytometry Distinction Between Adherent and Phagocytized Yeast Particles

Jean Giaimis, Yves Lombard, Philippe Poindron, and Christian D. Muller' Departement dImmunologie, Immunopharmacologie et Pathologie (J.G., Y.L., P.P.) and Laboratoire de Biophysique

U.A. n"491 du CNRS (C.D.M.), Universite Louis Pasteur de Strasbourg, Centre de Recherches Pharmaceutiques, Illkirch, France

Received for publication July 2, 1993; accepted April 27, 1994

Our laboratory recently developed a light microscopy staining technique that provides a mean to distinguish between yeast that are simply bound to the sur- face of macrophages and yeast that have actually been phagocytized by macro- phages (7). We adapted this technique by using fluorescent probes in order to test phagocytic activity by flow cytometry. Thus we are able to distinguish unambig- uously extracellular from intracellular yeast during phagocytosis with the fast rate of flow cytometry (-200 cells/s). The fluorescence quenching induced by a 1% tannic acid solution (wlv) can be applied to any FITC-labeled, heat-killed yeast

cell or bacteria. The yeast cells already engulfed in the macrophage remain with their native fluorescence (internal and external pH equilibrated by 50 pM mon- ensin 30 minI4"C) protected from the ac- tion of tannic acid, a nonmembrane per- meable molecule. The results presented here validate this new technique. An ap- plication is presented showing the inhibi- tion of endocytosis by cytochalasin-B. 0 1994 Wiley-Liss, Inc.

Key terms: Macrophage, fluorescent particles, fluorescence extinction, phago- cytosis, tannic acid, flow cytometry

The first step in phagocytosis is adherence of a par- ticle onto a phagocytic membrane. The following step is internalization of the particle and fusion of the surface membrane (19-22). Various methods have been used to study and measure phagocytosis (18,11,16). We re- cently described a new and simple method for quanti- fying the binding and internalization steps during phagocytosis of yeast by macrophages (7). Tannic acid (TA) was used to change the tinctorial properties of heat-killed yeast stained by May-Griinwald-Giemsa. The change in colors allows, by light microscopy, dis- crimination between adherent and phagocytized yeast. The procedure described here is an extension of the former method and is dedicated to flow cytometry ie . , cells in suspension. To discriminate between adherent and internalized yeast, one can exploit extinction of the fluorescence of the noningested yeast. Originally, Hed (9) described the use of crystal violet as a fluorescence- quenching dye to differentiate between attached and

ingested particles due to dye exclusion by viable cells. Subsequently, a flow cytofluorimetric technique with crystal violet as a quenching agent was developed to quantify ingestion and attachment (2,3). Yet, crystal violet belongs to the lysosomotropic dyes, i.e., inducing a 45-70% quenching effect on the fluorescence of in- gested particles in granulocytes and macrophages (9). Another dye described as a f luorescence-quenching molecule is trypan blue (10). At a concentration of 0.25

'Address reprint requests to C.D. Muller, Laboratoire de Biophy- sique U.A. n"491 du CNRS, Universitb Louis Pasteur de Strasbourg, Centre de Recherches Pharmaceutiques, 67401 Illkirch, France.

Abbreviations: A.U., arbitrary units; FACS, fluorescent activated cell-sorter; FCS, fetal calf serum; FITC, fluorescein 5-isothiocyanate; PBS, phosphate buffer saline; TA, tannic acid.

174 GIAIMIS ET AL

mg/ml or higher, trypan blue showed a plateau effect in fluorescence quenching, suggesting an effect on at- tached but not ingested particles. The assay has to be run in acidic conditions (pH 4.5) leading to a dimin- ished ligand-receptor binding (17). Recently, an ele- gant fluorescence wavelength shift from green to red was used to discriminate between internalized and membrane-bound particles (5). Yet that technology is based on the ability of the utilized FACS device to dif- ferentiate two emission wavelengths. The method de- scribed here allows unambiguous discrimination of ex- tra- and intracellular yeast with only one emission wavelength and at physiological pH.

MATERIALS AND METHODS Tannic acid was obtained from Merck (Darmstadt,

Germany). Tannic acid stock solution (10% [w/v]) was prepared in freshly distilled water and sterilized by filtration using Millipore filters (pore size: 0.40 bm; Dachstein, France).

Fluorescein 5-isothiocyanate (FITC), mannan, lami- narin, cytochalasin B and monensin (25 mM in ethanol prepared and stored a t -2OOC) were purchased from Sigma Chimie S.a.r.1. (La Verpilliere, France).

Cells The MALU macrophage cell line was used through-

out this study. This cell line has been established in our laboratory (13) from C57BL/6 lpr/lpr mouse lung, as previously described (12). Cells were grown in RPMI 1640 medium (Gibco BRL, Cergy-Pontoise, France), containing 10% heat-inactivated (30 min, 56°C) fetal calf serum (FCS; Gibco), 2 x lo5 UiL sodium benzyl- penicillinate, and 40 mg/L streptomycin sulfate. The absence of mycoplasma contamination was assessed by electron microscopic examination. The culture medium contained <1 ng/ml endotoxin as assayed by the Lim- ulus test. The cells, which has been shown to bear man- nose receptors on their surface (15), were regularly pas- saged as described by Lombard (13).

Yeast The strain of Saccharomyces cerevisiae used through-

out this study was a local strain grown in Sabouraud's broth for 48 h a t 28°C with agitation. The yeast were killed by autoclaving for 45 min at 120°C in the culture medium. After autoclaving, the suspension was washed three times in calcium- and magnesium-free phosphate-buffered saline (Ca' +- Mg+ '-free PBS) (4). The suspension was then aliquoted (1 ml) and stored at +4"C. Just before use, the suspension of killed yeast was gently sonicated (water bath apparatus, Bran- sonic, OSI, Paris, France) to disrupt clumps, and it was then diluted as needed in RPMI 1640 medium. The yeast particles were labeled with FITC according to Gefland (6), and the stock solution was kept into PBS- Dulbecco (SeromedB, Berlin, Germany) a t 4°C.

Fluorospectrometry and Flow Cytofluorometry Analysis

The fluorescence spectra of FITC-labeled yeast treated with or without increasing concentrations of TA and then washed three times in PBS were exam- ined using a SLM 4800 Spectrofluorometer (SLM, Aminco, IL) with the following settings: exc. 488 nm, 4 nm slit-width on excitation and emission monochroma- tors. The yeast were maintained in suspension by agi- tation in Ca' +-Mg+ +-free PBS (pH 7.4). The spectra was read against a suspension of autoclaved yeast par- ticles at the same concentration (20 x lo6 particlesiml) in the same buffer.

Fluoresceinated yeast were centrifuged (1,000 g, 15 rnin), resuspended in RPMI 1640 (lo8 particlesiml), and added to 1 ml of nonadherent cells (0.5 x lo6 cells/ ml) harvested from MALU culture flasks (12,131. After incubation for 60 min at 37°C in RPMI 1640 (serum free) in 5 ml polypropylene culture tubes (12 x 75mm PPN-rohrchen-Greiner, Dutscher, Brumath, France), the cells were aliquoted into two samples without any further washing step. For half the cell suspension, flow cytometric analysis was done on a FACS analyzer ATC 3000 (Brucker Spectrospin, Wissembourg, France) us- ing 400 mW of 488 nm light from an argon ion laser. Sort windows were used on forward and side scatter to eliminate debris and yeast not associated to cells (see Fig. 2C). Size and fluorescence (em. 515 nm) intensity were recorded at a rate of =200 cells/s. The other half of the cell suspension was incubated for 1 min in PBS containing 1% tannic acid prior to FACS analysis.

Calibration of the FACS by Quantum'" 25 beads (Flow Cytometry Standards Corporation, Research Tri- angle Park, NC) made i t possible to estimate the num- ber of FITC molecules associated with each yeast cell. The range covered was 6.3 x lo4 to 1.41 x lo6 FITC molecules using beads of 8.6 pm diameter.

RESULTS AND DISCUSSION FITC-Labeled-Yeast Fluorescence Quenching by

Tannic Acid (TA) Heat-killed yeast particles can easily be made fluo-

rescent by FITC to monitor the phagocytosis of such particles by fluorescent microscopy or flow cytometry (6). In preliminary spectrophotometric studies we no- ticed that the FITC fluorescence of labeled yeast is quenched by the addition of tannic acid (Fig. 1A) and that a 1% final concentration (w/v) of TA is sufficient to completely abolish the fluorescence of FITC-labeled yeast. The fluorescence extinction persisted when the yeast cells were washed three time in PBS, suggesting that TA is coating the yeast. The fluorescence intensity recovered up to the control levels when the yeast cells were washed once in a sodium carbonate buffer (pH 9.6) and then resuspended in PBS (data not shown), suggesting the ionic nature of the coating.

When assessed by flow cytofluorometry, the same quenching could be observed, as shown by the histo-

PHAGOCYTOSIS: ADHERENTiINTERNALIZED PARTICLES 175

FIG. 1. A. FITC yeast fluorescence spectra. The fluorescence quenching is obtained with 1-min incubation in increasing concentra- tions of TA (wiv). The fluorescence intensity spectra of FITC yeast washed three times in PBS (pH 7.4) were measured with an SLM 4800

grams in Figure 1B. The cytogramme and histo- grammes obtained for quenched FITC-labeled yeast (Fig. 1B) mask the data of unlabeled yeast particles. The coating of the yeast surface by TA can be detected by the slightly changed refringence of yeast so that the forward scatter pattern is shifted as seen in Figure 1B. We assessed by microscopy that no aggregation of the particles was induced by 1% TA. By fluorescence cali- bration of the FACS, we were able to determine that in our preparation each yeast particle was labeled by 4.26 x lo5 k 0.66 x lo5 FITC molecules (n = lo4). After fluorescence extinction by 1% TA, the value was less than 6.3 x lo4 molecules FITC, the lowest point of the standard curve. Thus TA causes a t least an 85% ex- tinction of fluorescence.

Fluorescence and FALS Distribution Histograms of MALU Cells after Incubation With

FITC-Yeast Particles Flow cytometry allows discrimination of cells and

yeast particles in suspension by setting a gate on the forward and right scattered light parameter (Fig. 2C). In this way, fluorescence of yeast particles not associ- ated to cells is completely separated from cell-associ- ated fluorescence, as shown in Figure 2A,B.

Assay Optimization Increasing concentrations of FITC yeast particles

were incubated for 1 h (37°C) with MALU suspension (Fig. 2). A saturation plateau was reached when the concentration of incubated particles was -20 per cell. The fluorescence intensity of cells incubated with FITC yeast does not reflect the actual amount of cell-associ- ated yeast in that the fluorescence of FITC-labeled pro-

Spectrofluorometer (excitation 488 nm). B. FITC yeast fluorescence by FACS. The fluorescence of FITC yeast (in black) is quenched by the addition of 1% TA. Quenched FITC yeast and native unstained yeast cytograms (in gray) are superimposed.

teins is highly pH dependent (14), i.e., adherent FITC- labelled yeast fluoresces normally, whereas labeled yeast in the phagolysosomes do not due to low pH. The addition of 50 FM monensin (30 min, 4”C), a proton sodium ionophore known to equilibrate internal and external pH (171, reveals that a proportion of the phagocytized particles is associated with endosomes and lysosomes (Fig. 2A).

Contact time between cells and TA was tested for up to 30 min, and no changes in cell viability or FITC- yeast-cell-associated fluorescence were observed (data not shown). Thus no evidence was obtained to suggest that the macrophage was permeable to the TA. With the previously determined 20 yeast particleskell ratio, the kinetic study of phagocytosis by MALU cells showed that the initial rate of mean fluorescence prior to TA addition (i.e., “external + ingested” yeast), be- gins to decline after 20 min. A plateau is reached after 50 min (the longest incubation time tested was 90 min). Thus a 60-min yeast-cell incubation time was used for further experiments.

Inhibition Assays by Cytochalasin B To illustrate the capabilities of our technique of flu-

orescence quenching by tannic acid, we choose an ex- perimental approach that was to inhibit the role of the cytoskeleton in mediating yeast internalization. Cy- tochalasin B, an inhibitor of micro-filament polymer- ization ( l ) , blocked the internalization of the yeast, but it did not prevent their adherence to the cell surface (Fig. 3). The quenching by tannic acid demonstrates that most of the particles were not surrounded by mem- branes, i.e., not internalized.

The technique described here allows an easy and

176 GIAIMIS ET AL.

FIG. 2. Fluorescence and FALS distribution histograms of MALU cells after incubation with increasing amounts of FITC yeast. The events acquisition was gated on size and granularity to remove yeast particles that were not associated with macrophages (C). Fluores- cence of 2,000 gated events showing the fluorescence due only to cell-associated yeast particles. Bold numbers inside of the histograms represent the numbers of yeast particles added per cell, (B) prior to

monensin addition, (A) after monensin addition (50 +M, 30 mid4"C) to equilibrate internal and external pH. As expected, the cell size slightly increases after phagocytosis of a large number of yeast par- ticles (D). The autofluorescence signals of MALU cells not incubated with FITC yeast particles are off-scale (values too low) due to the limited range of our ATC 3000 cell sorter (equiped for a two-log scale only).

prompt way to obtain an accurate rate of phagocytosis of macrophages. The use of tannic acid as a fluores- cence-quenching molecule for any kind of FITC-la- beled, heat-killed particle is the key to discriminating between internalized and cell surface adherent ele- ments. The precise mechanism of the quenching in- duced by TA remains unclear. A change in local pH andlor an energy transfer from FITC to TA can be at the origin of quenching. When yeast cells are washed several times in PBS, the quenching remains, suggest- ing a coating of the yeast surface by TA. It is only by incubating the coated yeast in a sodium carbonate buffer (pH 9.6) that the TA can be washed off the par- ticles, suggesting a strong ionic interaction of TA with

the surface of the heat killed yeast. Experiments are underway to clarify the exact nature of the quenching induced by TA.

The difference and advantage of our technology com- pared to those previously described come mainly from the mild (pH 7.4) assay condition compared to trypan blue assay run at pH 4.5. Recently, a fluorescence wavelength shift from green to red was obtained by incubating FITC with ethidium bromide inducing an energy-transfer phenomenon between the two mole- cules (5). This elegant way to follow phagocytosis has one major constraint; the FACS used to do the assay needs to be equipped with three detectors to detect side scattering and the two fluorescence wavelengths. With

PHAGOCYTOSIS: ADHERENTiINTERNALIZED PARTICLES 177

FIG. 3. Fluorescence distribution histograms of MALU cells after phagocytosis of FITC yeast: Inhibition by Cytochalasin B. Increasing amounts of Cytochalasin B (Fgiml; bold numbers next to histograms) were co-incubated with the cells (0.5 x 106/ml) and yeast particles (20 yeasticell; 37°C; 60 m i d . Flow cytometry analyses of 2,000 cells are

run as described under Materials and Methods. Bold numbers next to histograms represent the cytochalasin B concentration: A control; B: after addition of 1% TA; C: control in presence of 50 pM monensin; D in presence of 50 pM monensin after addition of 1% TA.

the technology described here on a FACS device equipped with two detectors only (side scatter and FITC fluorescence), we were able to obtain a FITC- fluorescence extinction by 1% TA and so could descrim- inate adherent from internalized yeast cells.

At present, it is not possible using our batch of cali- brated FITC yeast to quantify the average uptake of yeast particles per cell. This is simply due to the lack of commercialy available calibration beads with more than one million FITC molecules per bead. With the currently existing calibration beads, only a maximum of three FITC yeast cells associated per MALU cell can be estimated. That value is too low in our case consid- ering that MALU cells cultured on glass coverslips are able to ingest a mean of 20.8 t 0.9 particles per cell (n = 200) (8). When fewer FITC molecules per yeast were coupled, a far less homogeneous population (f luores-

cence intensities) was obtained leading to a miscal- culation of the average uptake of yeast particles per cell.

The experiments reported here were performed using FITC yeast particles. We have additionally found that the general approach is also applicable to phagocytic studies that use Streptoccocal bacteria as test particles. TA allows, by FACS (i.e., fluorescence quenching) and light microscopy (i.e., tinctorial alteration), measure- ment of the ability of cells to phagocyte yeast, respec- tively in suspension and adherent culture conditions.

ACKNOWLEDGMENTS We are grateful to Dr. Guy Duportail (Laboratoire de

Biophysique, URA 491 CNRS) for hints on fluores- cence quenching, and to Dr. Paul Fonteneau (Departe- ment d'Immunologie, Immunopharmacologie et Pa-

178 GIAIMIS ET AL.

thologie, Universite Louis Pasteur) for his interest in this investigation and critical reading of the manu- script.

LITERATURE CITED 1. Axline SG, Reaven EP: Inhibition of phagocytosis and plasma

membrane mobility of the cultivated macrophage by cytochalasin B. Role of subplasmalemmal micro filaments. J Cell Biol62: 647- 659,1974.

2. Bjerknes R, Bass6e C-F: Human leukocyte phagocytosis of zymo- san particles measured by flow cytometry. Acta Pathol Microbiol Immunol Scand [C] 91: 341-348, 1983.

3. Cantinieaux B, Hariga C, Courtoy P, Hupin J, Fondu P: Staphy- lococcus aureus phagocytosis: A new fluorometric method using FITC and paraformaldehyde. J Immunol Methods 121: 203-208, 1989.

4. Dulbecco R, Vogt M: Plaque formation and isolation of pure lines with poliomyelitis viruses. J Exp Med 99: 167-182, 1954

5. Fattorossi A, Nisini R, Pizzolo JG, DAmelio R New, simple flow cytometry technique to discriminate between internalized and membrane-bound particles in phagocytosis. Cytometry 10: 320- 325, 1989.

6. Gefland JA, Fauci AS, Green I, Frank MM: A simple method for the determination of complement receptor-bearing mononuclear cells. J Immunol 116: 595-599, 1976.

7. Giaimis J , Lombard Y, Makaya-Kumba M, Fonteneau P, Poin- dron P: A new and simple method for studying the binding and ingestion steps of lectinophagocytosis of yeast. J Immunol Meth- ods 154: 185-193, 1992.

8. Giaimis J, Lombard Y, Makaya-Kumba M, Fonteneau P, Muller CD, Levy R, Makaya-Kumba M, Lazdins J, Poindron P: Both mannose and P-glucan receptors are involved in phagocytosis of unopsonized, heat-killed Saccharomices cerevisiae by murine macrophages. J Leuk Biol54: 564, 1993.

9. Hed J : The extinction of fluorescence by crystal violet and its use to differentiate between attached and ingested microorganisms in phagocytosis. FEMS Microbiology Letters 1: 357-361, 1977.

10. Hed J , Hallden G, Johansson SGO, Larsson P: The use of fluorescence quenching in flow cytometry to measure the attach- ment and ingestion phases in phagocytosis in peripheral blood

without prior cell separation. J Immunol Methods 101: 119-125, 1987.

11. Lehrer RI: Ingestion and destruction of Candzda albicans. In: Methods for Studying Mononuclear Phagocytes. Adams DO, Edel- son JP , Koren H (eds). Academic Press, New York, 1981, pp 693- 708.

12. Lombard Y, Ulrich B, Poindron P: In vitro multiplication and apparently indefinite subcultures of normal mouse peritoneal res- ident macrophages. Biol Cell 53: 219-230, 1985.

13. Lombard Y, Bartholeyns J, Chokri M, Illinger D, Hartmann D, Dumont S, Kauffmann SH, Landmann R, Loor F, Poindron P: Establishment and characterization of long-term cultured cell lines of murine resident macrophages. J Leukocyte Bio 44: 391- 401, 1988.

14. Mollenhauer HH, Morre DJ, Rowe LD: Alteration of intracellular traffic by monensin: Mechanism, specificity and relationship to toxicity. Biochim Biophys Acta 1031: 225-246, 1990.

15. Muller CD, Lombard Y, Bartholeyns J , Poindron P, Schuber F: Characterization of surface markers of continuously growing rnu- rine resident macrophages. J Leukocyte Biol 49: 165-171, 1988.

16. Ogle JD, Noel JG, Stramkoski RM, Ogle CK, Alexander JW: Phagocytosis of opsonized fluorescent micro spheres by human neutrophils: A two color flow cytometry method for the determi- nation of attachment and ingestion. J Immunol Methods 115: 17-29, 1988.

17. Roche A-C, Midoux P, Bouchard P, Monsigny M: Membrane lec- tins on human monocytes. FEBS Letters 193: 63-68, 1985.

18. Shaw DR, Griffin FM: In: Methods for Studying Mononuclear Phagocytes, Adams DO, Edelson JP, Koren H (eds). Academic Press, New York, 1981, pp 511-527.

19. Silverstein SC, Steinman RS, Cohn ZA: Endocytosis. Ann Rev Biochem 46: 699-722, 1977.

20. Stahl PD, Schlessinger PH, Sigardson E, Rodman JS, Lee YC: Receptor mediated pinocytosis of mannose glycoconjugates by macrophages: Characterization and evidence for receptor recy- cling. Cell 19: 207-215, 1980.

21. Stahl PD: The macrophage mannose receptor. Am J Respir Cell Mol Biol 2: 317-318, 1990.

22. Ueda MJ, Ito T, Ohnishi SI, Okada TS: Phagocytosis by macro- phages: Kinetics of adhesion between particles and phagocytes. J Cell Scien 51: 173-188, 1981.