Mar 25, 1986 - marker has been localized to the subcellular fraction(s) containing specific (and probably tertiary) granules.'6l'7. Following stimulation with the ...
Expression of a Granule Membrane Marker on the Surface of Neutrophils Permeabilized With Digitonin Correlations With Ca 2+-Induced Degranulation
JAMES E. SMOLEN, PhD, ROBERT F. TODD Ill, MD, PhD, and LAURENCE A. BOXER, MD
From the Division of Pediatric Hematology-Oncology, C. S. Mott Children's Hospital, and the Division of Hematology-Oncology, Department of Internal Medicine, University of Michigan Medical School, Ann Arbor, Michigan
The authors have previously shown that human neutrophils can be permeabilized with the cholesterolcomplexing agent digitonin and that these cells can be induced to secrete granule contents by increasing free Ca2 concentrations. In the studies reported here, the authors wished to determine whether secretion of granule constituents correlated with the appearance of an immunologic marker for granule membranes on the surface of the permeabilized neutrophils. For this purpose, we used flow cytometry and two fluorometrically identifiable markers, Mol (a granule membrane marker) and IB2-microglobulin (,B2m) (a plasma membrane marker). It was found that the ratio of Mol/02m increased for perme-
abilized neutrophils which were exposed to micromolar concentrations of free Ca2". This increase in the detectible surface concentration of Mol was accompanied by the release of lysozyme, vitamin B12 binding protein, and I-glucuronidase into the medium. Statistical analysis showed a very strong correlation between granule secretion and the Mol/,B2m ratio, These data thus suggest that granule membrane components were being introduced into the plasma membrane during Ca2 -induced granule discharge; this in turn suggests that secretion by permeabilized neutrophils represents true degranulation. (Am J Pathol 1986, 124:281-285)
A NUMBER of lines of evidence suggest that intracellular free Ca2l plays a second messenger role in human neutrophils. We have reported that extracellular chelators inhibit, while intracellular calcium antagonists block, neutrophil responses. ' Mobilization of intracellular Ca2l appears to be rapid when measured as a disappearance from putative membrane-bound deposits with chlortetracycline2-5 or as the appearance of the free cytosolic species with Quin 2.6-10 We have recently shown that the cells can be permeabilized with the cholesterol-complexing agents saponin or digitonin and induced to release granule contents by exposure to Ca2l I2 These data thus complement the results obalone."" tained with the fluorescent probes by showing that direct application of the putative second messenger results in the release of granule contents. While we have obtained substantial biochemical"'2 and morphologic'3 evidence that the release of granule contents from permeabilized neutrophils represents true degranulation (actual fusion of the plasma and granule membranes), direct support for this mechanism has
been lacking. In order to provide stronger evidence, we wished to see whether a granule membrane component was "translocated" to the cell surface during Ca2linduced granule discharge. For this purpose, we decided to examine Mol, which is a membrane antigen detectable on human monocytes, neutrophils, and null cells,'4 and which is believed to be a determinant on the C3bi receptor.'5 With the use of immunoprecipitation, this marker has been localized to the subcellular fraction(s) containing specific (and probably tertiary) granules.'6l'7 Following stimulation with the ionophore A23187, Mol has been shown to be "translocated" to the plasma memSupported by NIH Grants AM 32471 (J.E.S.), CA-39064 (R.F.T.), HL-31963 (L.A.B.) and AI-20065 (L.A.B.). Dr. Smolen is a Senior Fellow of the Arthritis Foundation. Accepted for publication March 25, 1986. Address reprint requests to James E. Smolen, PhD, Division of Pediatric Hematology-Oncology, C. S. Mott Children's Hospital, University of Michigan Medical School, Ann Arbor, MI 48109.
281
282
SMOLEN ET AL
brane of intact neutrophils, increasing surface expression by 5-10-fold.16'18 In order to follow the disposition of Mol using flow cytometry, we also measured the expression of 02microglobulin (,B2m), which is a plasma membrane marker and which is not "translocated" during degranulation.19 We found that the ratio of Mol/032m increased following Ca2+-induced granule discharge from permeabilized neutrophils and that this increase was closely correlated with secretion. These results thus provide strong evidence that the secretion observed in this model system represents true degranulation.
Materials and Methods Reagents Digitonin was purchased from Sigma Chemical Company, St. Louis, Missouri. All other materials were reagent grade. Monoclonal Antibodies Mouse ascites containing monoclonal antibodies specific for Mol14 (IgG-2a subclass), 132m20 (IgG-2b subclass, a gift of Dr. Lee Nadler, Boston, Mass), and Ia21 (IgG-2a subclass) were used in indirect immunofluorescence studies. Preparation of Cell Suspensions Heparinized (10 U/ml) venous blood was obtained from healthy adult donors. Purified preparations of neutrophils were isolated from this blood by means of Hypaque-Ficoll gradients22 followed by standard techniques of dextran sedimentation and hypotonic lysis of erythrocytes.23 This allowed studies of cell suspensions containing 98%o ± 2%o neutrophils with few contaminating platelets or erythrocytes. The cells were washed and finally suspended in a buffer consisting of 138 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4, mM CaCl2, and 0.6 mM MgCl2, pH 7.4 (PiCM). For permeabilization studies, the cells were instead washed once and resuspended in Buffer K (100 mM KCl, 20 mM NaCl, 1 mM EGTA, and 30 mM HEPES, pH 7.0).
Neutrophil Permeabilization Permeabilization was performed as previously described.12 In essence, stock solutions of digitonin (1 mg/ml in buffer K) were prepared daily. Neutrophils (25 x 106/ml) suspended in Buffer K were preincubated for 10 minutes at 37 C. Incubation was at 37 C for 25
AJP * August 1986
minutes with 10 gg/ml digitonin; the cell suspension was mixed every 5 minutes. Lysosomal Enzyme Release After permeabilization, neutrophils were suspended in Buffer K at a concentration of 5 x 106/ml and then incubated with the desired concentration of Ca2l at 37 C for 10 minutes; free Ca2+ was determined for the buffer by using a Ca2+ electrode. Cells incubated without calcium, and unpermeabilized neutrophils were employed as routine controls. The cell suspensions were then centrifuged at 750g for 10 minutes. Aliquots of the supernatants were taken for standard determinations of f1-glucuronidase24 (an enzyme found exclusively in azurophil granules), lysozyme25 (an enzyme found in both specific and azurophil granules), vitamin B12 binding protein26 (a component of specific granules alone), and LDH activities" (a cytosolic enzyme).
Immunofluorescent Analysis of Membrane Antigens Indirect immunofluorescence of Mol, Ia, and I2m were measured by flow cytometry as described by Todd et al.16 The channel number (log scale) measuring the mean fluorescence intensity of 5000 cells, which had been exposed to either experimental or a negative control antibody (anti-Ia) and then to a fluoresceinconjugated goat anti-mouse immunoglobulin, was determined. Logarithmic channel numbers were converted to linear values,28 and the specific mean fluorescence intensity for cells stained by experimental antibodies was calculated by subtracting the mean channel number of cells exposed to the negative control antibody from that of cells stained by the experimental antibodies.
Results The purpose of this work was to examine the appearance of Mol, a granule membrane component, on the surface of permeabilized neutrophils after exposure to
micromolar concentrations of free Ca2+. For these studies, we routinely employed Ia as a control antibody for nonspecific binding, which was invariably low on intact cells. However, once the neutrophils were permeabilized with digitonin, detection of Ia increased almost nine-fold (Figure 1). The high background of Ia was attributable to increased access of the antibody to the cytosol, wherein it could be detected nonspecifically; this latter contribution decreased as Ca2 concentration increased (Figure 1), reflecting partial resealing of the cells. 13 This resealing had the fortuitous advantage of minimizing the possibility that the specific antibodies
MO1 EXPRESSION ON PERMEABILIZED NEUTROPHILS
Vol. 124 * No. 2
could gain access to intracellular pools of antigen after stimulation with Ca2'. Because of the high background, it was necessary to subtract the values measured for the control antigen from all of the experimental measurements. Once this correction was made, expression of the putative plasma membrane marker 32m did not significantly increase after permeabilization (Figure 1). This suggests that no substantial pool of this antigen was exposed after permeabilization. Expression of I2m also remained constant once these cells were exposed to Ca2+, as would be expected for a plasma membrane marker. In contrast, expression of Mol tripled following permeabilization (Figure 1), which reflects, perhaps, some activation of the neutrophils by digitonin. 12 Detectable Mol increased following exposure to micromolar levels of
12 F I-
10
z z
w
B Cla n
w6 0
_
o -J L 4 I-i 2 0
0
0 U
13.w
a
*oA (INTACT)
-1'*
0
1
3
10
30
283
°3 100
Ca2+(a.M M) Figure 1-The effect of free Ca2+ concentrations on the detection antidetecton of Ofantlgenic markers on intact and permeabilized human neutrop)hils. Human neutrophils were permeabilized with digitonin as outlined in Materials and Methods and then exposed to the indicated concentrat ions of Ca2+. Following incubation for 10 minutes at 37 C, the cell suspeilnsions were centrifuged and the supernatants were removed. The ccell pellets were resuspended and analyzed for expression of the antigenil markers by flow cytometry. To correct for nonspecific binding, the fluores;cence intensities measured for la were subtracted from those obtained wi,ith both 132m and Mol before any further calculations were performed. In oirder to accommodate changes in instrument calibration and gain, the mean channel numbers for intact cells were used as a standard and given the value of 1.0; all other samples are expressed as ratios of this value (ccorrected fluorescence intensity). The results are presented as the mesans (+| SD) for 5 separate experiments. Solid points represent those which were significantly different from the permeabilized, 0 Ca2+ control on the basis of a paired Student t test.
Ca
2+
.
As reported before,12 and as shown in Figure 2, increasing concentrations of Ca2+ result in enhanced release of vitamin B12 binding protein (a component of
~c
specific granules), lysozyme (a component of both specific and azurophil granules) and ,-glucuronidase (an enzyme found in azurophil granules). Release of vitamin B12 binding protein and lysozyme was most sensitive to Ca2 , with substantial release occurring at 7-13 gM; release of 1-glucuronidase required 13-30 gM. Secretion of the cytosolic enzyme lactate dehydrogenase 2
60 LAJ LJ
w 0
9
cr
-J
2.0
40
E
=
trJ
ae
z
20
0
[CC2+1 FREE (>M)
10
[Co24]FREE
30
100
(M)
Figure 2-The effect of free Ca2+ concentrations on the release of cellular constituents from and expression of Mol/02m on permeabilized human neutrophils. Human neutrophils were permeabilized with digitonin as outlined in Materials and Methods and then exposed to the indicated concentrations of Ca2+. After incubation for 10 minutes at 37 C, the cell suspensions were centrifuged and aliquots of the supernatants taken for determinations of released LDH, vitamin B,2 binding protein, lysozyme, and j-glucuronidase. The cell pellets were resuspended and analyzed for expression of the antigenic markers by flow cytometry. The results are presented as the means (± SD for Mol/132m) for five separate experiments; SDs for the cellular constituents were omitted for the sake of clarity. Solid points represent those which were significantly different from the permeabilized, 0 Ca2+ control on the basis of a paired Student t test.
284
SMOLEN ET AL
(LDH) was diminished under these conditions, which reflects partial resealing of the cells."2-"3 It should be noted that the concentrations of Ca2l required for secretion by this permeabilized cell system are substantially greater than the intracellular level reported for stimulated neutrophils by Quin-2.i1 In order to control for changes in detectibility of membrane antigens after treatment with digitonin and/or Ca2+, we compared Mol expression with that of 32m. By calculating the ratio of Mol/12m, an evaluation of the relative surface concentration of Mol should be obtained. The Mol/032m ratio increased when permeabilized neutrophils were treated with Ca2l (Figure 2). As can be seen, this ratio increased substantially at 4-7 gM, thereby paralleling release of lysozyme and vitamin B12 binding protein. Substantial day-to-day variability in the extent of permeabilization was reflected in variability of release of cellular constituents and expression of Mol (apparent in Figures 1 and 2). In order to more closely examine the correlation between detection of Mol and secretion of granule contents, we constructed scattergrams and analyzed them by regression analysis. The data included those used for Figure 2 as well as those from the intact, resting control samples prior to permeabilization. As can be seen in Figure 3, the Mol/0i2m ratio correlated moderately well with secretion of all three granule constituents. The correlation coefficients (r) ranged from only 0.55 to 0.68, reflecting the substantial day-to-day variability of the data. However, the significance of the trends was very high, with P values from 0.01 to 0.0005.
Discussion Our data show that expression of the granule membrane antigen Mol increased after exposure of permeabilized neutrophils to micromolar concentrations of free Ca2+. This expression was accompanied by release of granule contents such as vitamin B12 binding protein, lysozyme, and I-glucuronidase. A close correlation between the Mol/,B2m ratio and granule discharge was also observed. These data thus suggest that granule membrane components were incorporated into the surface membrane during the release reaction. There is substantial evidence indicating that Mol is a granule membrane component. It is expressed on the cell surface following exposure to degranulating stimulation, such as A2318716; stimulation that does not produce degranulation does not increase the expression of Mol. Furthermore, immunoprecipitation studies with anti-Mol antibody have localized the antigens to a subcellular fraction containing specific granules.16 However, it should be noted that this latter evidence does not prove a specific granule localization; they could also
AJP * August 1986 LDH r- 0.1472 (N. S.)
*0 40 _ wL CO
20
wx
0
.8- Glucuronidose
30 r
r 0.596 (p< 0.002)
20 F
: .0*
0
10F
I
10
0
0
-
11
0~~~~~~~~ Vit. 812 BP
Lysozynse
30 40
wL
30
. 0*
_
:
20 30
.*
20 10
*.
.*
20 20
1.0
2.0
3.0
*
0
r~0.550
,
.
(p < 0.0005)
-*
0
r z0.680
0
.
-Q5
.
1.0
(p < 0.01) 2.0
3.0
Mo l/A2m RATIO
Figure 3-Correlations between the Mol/132m ratio and release of cellular constituents. The data from Figure 2 as well as those from intact neutrophils were used for construction of the scattergrams shown. Regression analysis was applied for each pair of cellular constituents and the Mol/12m ratio. Correlation coefficients (r) and probability values are given.
be contained in other organelles of similar sedimentation behavior, such as tertiary granules. I2m can probably be considered a plasma membrane marker. This antigen does not increase in expression after exposure to degranulating stimuli"9 or to permeabilization (Figure 1). Thus, ,B2m was useful for normalizing our Mol data with respect to detectable plasma membrane. This permitted us to minimize uncertainties with regard to antigen detection which might arise in the permeabilized neutrophil system. The Mol/0i2m ratio also should provide a measure of the relative concentration of Mol in the surface membrane. One of the difficulties of using the permeabilized cell system was that the extent of permeabilization varied substantially from day to day. This is a recognized, inescapable limitation,11 13 which led in turn to variability in detection of all of the markers examined. Nonetheless, the results for any given experiment were quite consistent in showing the association between Mol expression and secretion. Even when the raw data were employed, a very strong correlation was found between these two events (Figure 3). Our data suggest that Mol is localized in a subcellular pool behaving like the specific granules. This antigen was expressed at the same Ca2l concentrations as caused release of lysozyme and vitamin B12 binding protein. However, this does not prove a specific granule localization. Indeed, we have evidence that at least some of the Mol is located in tertiary granules (manuscript
Vol. 124 * No. 2
MO1 EXPRESSION ON PERMEABILIZED NEUTROPHILS
submitted). It is of greater import that our data simply indicate that Mol was expressed by the same concentrations of free Ca2l which resulted in granule secretion. This in turn suggests that granule membrane components were being introduced into the surface membrane, as an accompaniment of secretion. These data thus provide evidence that fusion between the granule and plasma membranes was taking place and that the secretion being measured in the permeabilized neutrophil system represents true degranulation.
References 1. Smolen JE, Korchak HM, Weissmann G: The roles of extracellular and intracellular calcium in lysosomal enzyme release and superoxide anion generation by human polymorphonuclear leukocytes. Biochim Biophys Acta 1981, 677:512-520 2. Naccache PH, Showell HJ, Becker EL, Sha'afi RI: Involvement of membrane calcium in the response of rabbit neutrophils to chemotactic factors as evidenced by the fluorescence of chlorotetracycline. J Cell Biol 1979, 83:179-186 3. Naccache PH, Volpi M, Showell HJ, Becker EL, Sha'afi RI: Chemotactic factor-induced release of membrane calcium in rabbit neutrophils. Science 1979, 203:461-463 4. Smolen JE, Weissmann G: The effect of various stimuli and calcium antagonists on the fluorescence response of chlorotetracycline-labelled human neutrophils. Biochim Biophys Acta 1982, 720:172-180 5. Smolen JE, Eisenstat BA, Weissmann G: The fluorescence response of chlorotetracycline-loaded human neutrophils: Correlations with lysosomal enzyme release and evidence for a "trigger" pool of calcium. Biochim Biophys Acta 1982, 717:422-431 6. White JR, Naccache PH, Molski TFP, Borgeat P, Sha'afi RI: Direct demonstration of increased intracellular concentration of free calcium in rabbit and human neutrophils following stimulation with chemotactic factor. Biochem Biophys Res Commun 1983, 113:44-50 7. Pozzan T, Lew DP, Wollheim CB, Tsein RY: Is cytosolic ionized calcium regulating neutrophil activation? Science 1983, 221:1413-1415 8. Korchak HM, Rutherford LE, Weissmann G: Stimulus response coupling in the human neutrophil: I. Kinetic analysis of changes in calcium permeability. J Biol Chem 1984, 259:4070-4075 9. Korchak HM, Vienne K, Rutherford LE, Wilkenfeld C, Finkelstein MC, Weissmann G: Stimulus response coupling in the human neutrophil: II. Temporal analysis of changes in cytosolic calcium and calcium influx. J Biol Chem 1984, 259:4076-4082 10. Naccache PH, Molski TFP, Borgeat P, Sha'afi RI: Intracellular calcium redistribution and its relationship to fMet-Leu-Phe, leukotriene B4, and phorbol ester induced rabbit neutrophil degranulation. J Cell Physiol 1985, 122:273-280 11. Smolen JE, Stoehr SJ: Micromolar concentrations of free calcium provoke secretion of lysozyme from human neutrophils permeabilized with saponin. J Immunol 1985, 134:1859-1865
285
12. Smolen JE, Stoehr SJ, Boxer LA: Human neutrophils permeabilized with digitonin respond with lysosomal enzyme release when exposed to micromolar levels of free calcium. Biochim Biophys Acta 1986, 886:1-17 13. Smolen JE, Stoehr SJ, Boxer LA: Partial resealing of neutrophils previously permeabilized with digitonin (Abstr). J Cell Biol 1984, 99:331a 14. Todd RF III, Schlossman SF: Analysis of antigenic determinants on human monocytes and macrophages. Blood 1983, 59:775-786 15. Arnaout MA, Todd RF III, Dana N, Melamed J, Schlossman SF, Colten HR: Inhibition of phagocytosis of complement C3 or immunoglobulin G coated particles and of C3bi binding by monoclonal antibodies to a monocytegranulocyte membrane glycoprotein (Mol) J Clin Invest 1983, 72:171-179 16. Todd RF III, Arnaout MA, Rosin RE, Crowley CA, Peters WA, Babior BM: Subcellular localization of the large subunit of Mol, a surface glycoprotein associated with neutrophil adhesion. J Clin Invest 1984, 74:1280-1290 17. Borregaard N, Heiple JM, Simons ER, Clark RA: Subcellular localization of the b cytochrome component of the human neutrophil microbicidal oxidase: Translocation during activation. J Cell Biol 1983, 97:52-61 18. Arnaout MA, Spits H, Terhorst C, Pitt J, Todd RF III: Deficiency of a leukocyte surface glycoprotein (LFA-1) in two patients with Mo-I deficiency. J Clin Invest 1984, 74:1291 19. Berger M, O'Shea J, Cross AS, Folks TM, Chused TM, Brown EJ, Frank MM: Human neutrophils increase expression of C3bi as well as C3b receptors upon activation. J Clin Invest 1984, 74:1566-1571 20. LeClair K, VanAghovten A, Terhorst C: A rapid method for isolation of antigenically detectable human cell surface antigens associated with 02 microglobulin using monoclonal antibody. J Immunol Meth 1981, 41:137-139 21. Todd RF III, Meuer SC, Romain PL, Schlossman SF: A monoclonal antibody that blocks class II histocompatability-related immune interactions. Hum Immunol 1984, 10:23-40 22. Boyum A: Isolation of mononuclear cells and granulocytes from human blood: Isolation of mononuclear cells by centrifugation and of granulocytes by combining centrifugation and sedimentation at 1 G. Scand J Clin Lab Invest 1968, 21:77-89 23. Zurier RB, Hoffstein S, Weissmann G: Cytochalasin B: Effect on lysosomal enzyme release from human leukocytes. Proc Natl Acad Sci USA 1973, 70:844-848 24. Brittinger GR, Hirschhorn R, Douglas SD, Weissmann G: Studies on lysosomes: XI. Characterization of hydrolase-rich fraction from human lyphocytes. J Cell Biol 1968, 37:394-411 25. Worthington Enzyme Manual. Freehold, NJ, Worthington Biochemical Co., 1972, pp 100-101 26. Smith GP, Peters TJ: The release of granule components from human polymorphonuclear leukocytes in response to both phagocytic and chemical stimuli. Biochim Biophys Acta 1982, 719:304-308 27. Wacker WEC, Ulmer DD, Vallee BL: Metalloenzymes and myocardial infarction: II. Malic and lactic dehydrogenase activities and zinc ion concentration in serum. N Eng J Med 1956, 255:449-456 28. Muirhead KA, Schmitt TC, Muirhead AR: Determination of linear fluorescence intensity from flow cytometric data accumulated with logarithmic amplifiers. Cytometry 1983, 3:251-256
