Journal of Immunological Methods, 147 (1992) 13-19 © 1992 Elsevier Science Publishers B.V. All rights reserved 0022-1759/92/$05.00
13
JIM 06182
An improved europium release assay for complement-mediated cytolysis Jian Cui and Jean-Oaude Bystryn Department of Dermat%gy, New York University School of Medicine, New York, NY, USA (Received 7 June 1991, revised received 21 August 1991, accepted 4 September 1991)
An improved assay for complement-mediated cytolysis is described. The target cells are labeled with europium complexed to diethylenetriaminopentaacetate (Eu-DTPA). Cytolysis caused by antibody plus complement leads to the release of the Eu-DTPA complex which is then formed into a highly fluorescent chelate by the addition of 2-naphthoyltrifluoroacetone (2-NTA). The amount of europium chelate formed - a measurement of cell death - is then quantified with a time-resolved fluorometer. The results of the assay are reproducible. Complement-mediated cytolysis when measured by europium release was five times more sensitive than when measured by conventional Ster release and three times than when measured by trypan blue exclusion. Because europium does not decay, target cells can be labelled in batches and stored frozen until use, which speeds and simplifies the assay. Thus, europium release assay is a simple and quantitative method to measure complement-mediated cytolysis which is sensitive and more rapid than conventional assays. Key words: Europium; Complement; Cytotoxicity; Antibody; Cytolysis; Melanoma
Introduction The traditional method to measure complement-mediated cytolysis is by dye exclusion, a procedure which is read manually under a microscope and so is lengthy and the results subjective. A more quantitative method to measure immune cytolysis based on the release of slchromium was introduced by Goodman (1961) and applied to measure T cell and complement mediated cytolysis (Scomik et al., 1978; Russell et al., 1980; Moen et al., 1983; Schlaser et al., 1983). Though more accurate than dye exclusion, Sler release
Correspondence to: J.-C. Bystryn, NYU School of Medicine, 560 First Avenue, New York, NY 10016, USA
has the disadvantages that target cells must be labelled prior to each experiment, a lengthy procedure; and that radioactive waste is generated. In addition, chromium binds to cytoplasmic proteins by ionic or other forces and this may delay or decrease its release from damaged cells and thus reduce the sensitivity of the assay. These problems have stimulated attempts to develop non-radioactive, sensitive and rapid assays for cytolysis. Recently, Blomberg and co-workers (1986a,b) have described an assay to measure cell-mediated cytolysis in which the target cells are labelled with a europium-diethylenetriaminopentaacetate (Eu 3 +-DTPA) complex. After release of the complex from lysed cells, Eu3+ is chelated with 2-naphthoyltrifluoroacetone (2NTA) to form a highly fluorescent chelate which
14
can be detected with a fluorometer. Some biological tissues can release fluorescence, but the decay time of this biological fluorescence is only 10-20 ns. By contrast, the fluorescence decay time of Eu chelate is in the range of 100-1000 ns and can be quantified and distinguished from the biological fluorescence by time-resolved fluorometry (Soini et aI., 1983; Jackson et aI., 1986). When applied to measure cell-mediated cytolysis, this assay is more sensitive and faster than 51Cr release assay (Hemmila et aI., 1984; Blomberg et aI., 1986a,b; Grenberg et aI., 1988; Volgmann et aI., 1989; Maley et aI., 1990). This study was conducted to adopt the europium release method to measure antibody dependent complement-mediated cytolysis. Materials and methods Target cells Human melanoma cells SK-mel-28 (provided by Dr. J. Fogh, Sloan-Kettering Institute, Rye, NY) were used as the target. The cells were cultured in DME medium (Gibco, Grand Island, NY) with 10% fetal calf serum (Gibco). Antibody and complement R24 monoclonal antibody (a generous gift of Dr. Allan Houghton, Memorial Sloan-Kettering Cancer Center, New York) to the GD3 surface antigen of SK-mel-28 was used as the antibody source. Human complement sera (Sigma, St. Louis, MO) was used as the complement source. Labelling of target cells with europium Cells were washed 2 X with 50 ml buffer I (pH 7.4) consisting of 50 mM Hepes, 93 mM NaCl, 5 mM KCI, and 2 mM MgCl z and then suspended at approximately 5 X 10 6 cells/ml in labelling buffer (buffer I supplemented with 200 IL M EuCI 3 (Fluka, Ronkonkoma, NY), 1.0 mM diethylenetriaminopentaacetate (DTPAXSigma, St. Louis) and 500 ILg/ml dextran sulfate (MW 500,000, Pharmacia Fine Chemicals, Piscataway, NJ). After incubation at room temperature for 15 min with occasional shaking, the labelling reaction was stopped by the addition of 4 ILI/ml of 1 M CaCl z solution. After 5 min the cells were washed 2 X
with 50 ml buffer III (buffer I supplemented with 2 mM CaCI z and 10 mM glucose), then 2 X with 50 ml DME culture medium. Labelled cells were resuspended in DME culture medium containing 10% DMSO at 10 6 cells/ml, then 1 ml aliquots dispensed into 2 ml freezer vials (LOP, North Haledon, NJ) and frozen at - SOOC until use, usually within 4 months. For use, the required number of vials were thawed in a 37'C water bath. The cells were washed 2 X with 50 mI HBSS, 2 X with 50 ml DME culture medium, counted by trypan blue exclusion, and resuspended in DME culture medium at 105 live cells/mt. Labelling of target cells with "chromium Approximately 10 7 cells were incubated with 100 ILCi of 51Cr (Du Pont, Wilmington, DE) in 0.5 ml of DME culture medium for 2 h at 37'"C with frequent shaking, washed 4 X with 50 ml DME medium, and adjusted to a concentration of 10 5 live cells/ml medium. Complement-mediated cytolysis 50 ILl of europium or 5lCr-labelled cells (lOs cell/mJ) and 50 ILl of diluted R24 antibody were pipetted into the wells of round-bottomed 96 well microtiter tissue culture plates and shaken in an icebath for 45 min. Then 100 ILl of 1/5 diluted human complement serum was added to each well and incubated at 37°C. After 2 h, the plates were centrifuged at 100 X g for 5 min, and 20 ILl of supernatant from each well transferred to the wells of flat-bottomed microtiter strips (FlOW Laboratories, McLean, VA) containing 200 #,1 of enhancement solution consisting of 0.1% (v/v) Triton X-100, 15ILM 2-naphthoyltrifluoroacetone (2-NTA) and 50 ILM Tri-n-octylphosphine oxide (TOPO) (Wallac, Turku, Finland). After mixing at room temperature for 5 min, fluorescence was measured in a time-resolved fluorometer (Arcus 1230, LKB-Wallac, Turku, Finland). The super_ natant of 51Cr-labelled cells was collected by har_ vesting frames (Skatron, Sterling, VA) and the radioactivity counted in a gamma-counter (l282 CompuGamma, LKB-Wallac). All assays were performed in triplicate. Controls included cells incubated only with culture
15
medium, complement, or antibody. Percent cytolysis was calculated from:
100r---------------~
ao~---------~~~~-~
................................
% specific lysis
.o~-~~----~--~----~
experiment release - spontaneous release
... ...
- - - - : - - - - - - - - - - x 100
40~~~~~--------~o.---~
maximum release - spontaneous releases
In both europium and SICr release assays, spontaneous release was the amount of label released by target cells incubated with complement only. Maximum release for europium was the amount released from target cells lysed with 0.1 % Triton X-1OO (Sigma, St. Louis, MO) and for SICr that released by target cells lysed with 10% Linbro (Flow laboratories, McLean, VA). To measure cytolysis by dye exclusion, the cells following incubation with complement were spun down at 100 x g for 5 min and the supernatant removed. The cells were resuspended in 30 J.Ll of 0.4% trypan blue (Gibco) and the ratio of live/dead cells counted under a microscope.
Results Measurement of complement mediated cytolysis by europium release
The effect of incubating europium labelled target cells (5000 /well) for increasing periods of time with R24 antimelanoma antibody (0.20 J.Lg/well) and complement is illustrated in Fig. l. There was rapid release of europium during the first 15 min of incubation followed by a slower but continued release. 60% of total europium was released in the first 15 min of. incubation with antibody and complement and 87% after 4 h. By contrast, spontaneous release of europium from cells incubated only with complement was much slower - resulting in 23% release in 15 min and 38% in 4 h. Specific release increased steadily with time reaching a maximum of 78% at 4 h. The ratio of specifically to spontaneously released europium was greatest at 2 h of incubation, and this incubation time was used for all subsequent assays. To study the effect of target cell number on cell lysis, experiments were conducted in which a fIXed dose of R24 antimelanoma antibody (0.20
......._.._.-------- ----_._ -_....--
10~~~~-----------~
O~-~--~---~--~~-~-~ o ,. at .0 110 240 TIME OF Eu RELEASE (MINI
Fig. 1. Assay of complement mediated cytolysis by europium release. Replicate aliquots of europium-labelled SK-mel-28 melanoma cells (5000 celis/weI\) were incubated with R24 antimelanoma monoclonal antibody (0.2 ~g/weH) and complement (total release) or with complement only (spontaneous release) for increasing periods of time. Specific release was calculated by subtracting spontaneous release from total release. Results are mean of three experiments.
J.Lg/well) and complement was incubated with increasing numbers of cells for 2 h. The results are shown in Fig. 2. The proportion of europium spontaneously released was stable over a cell concentration of 20,Ooo-2500/well but increased rapidly when the number or' target cells was reduced to 1250/well. Specific cytolysis increased gradually with decreasing numbers of target cells reaching a peak value of 100% at a target cell number of 1250/well. The ratio of specific to 100
ao
!
i
-
.0 40 ......... __ ••••• - ......... -----.- .. - - 4 ......
~~
10 0 10,000
10,000
1,000
1,100
1,110
CELL NUMBER/w,11
Fig. 2. Effect of number of target cell number on europium release assay. Increasing number of europium-labelled Skmel-28 melanoma cells (12S0-20,OOO cells/well) were incubated with R24 anti melanoma antibody (0.2 ~g/well) and complement or without complement (spontaneous release). Specific release was calculated as descripted. Results are mean of three experiments.
16 TABLE I REPRODUCIBILITY OF EUROPIUM RELEASE ASSAY Specific cytolysis (%)
Experiment
Incubation time (min) 15
30
60
120
240
3
35 35 34
50 58 46
60 55 57
62 66 60
85 73 76
Average±SD
35 ± 0.6
51±6
57±2.5
63±3
78±6
1
2
spontaneous release was greatest at 5000 cells/well and was used in all subsequent assays.
antibody concentrations and 13% at one concentration (Table 11).
Reproducibility of europium release assay To measure the reproducibility of the assay, replicate experiments were performed on three different days during a period of 12 weeks in which 5000 melanoma cells/well were incubated with a fixed dose of R24 antibody (0.20 JLg/well) and complement for different periods of time. The results are shown in Table I. As previously noted, specific cytolysis increased steadily with time in all three experiments. The standard deviation of specific release at each time point was between 0.6 and 6%. In a subsequent study, the reproducibility of the assay was examined by incubating increasing doses of antibody (0.01-0.20 JLg/well) with a fixed number of target cells (5000 /well) for a fIXed period of time (2 h). In three different experiments, the standard deviation of specific release was 3-5% at four of the
Comparison of europium release, 51Chromium release and dye exclusion assays Time sequence studies were conducted concurrently with europium and St Cr-Iabelled cells. Spontaneous release of europium and StCr was similar, while specific release of europium is more rapid than that of StCr (Fig. 3). Following 30 min of incubation, 50% of europium was specifically released vs only 10% of StCr. The amount of europium released was also greater than StCr. Maximum specific release of europium after 4 b of incubation was 78% vs. 43% for StCr. 50% of the maximum amount of europium that could be specifically released by antibody plus complement 100
10 10
~
TABLE II
RELATION BElWEEN ANTIBODY CONCENTRATION AND CYTOLYSIS Experiment
10
Specific cytolysis (%) Antibody • (#,g/welJ)
o
0.01
0.02
0.05
0.10
0.20
28 33 53
50 59
63 68 73
77
3
31 37 30
Average±SD
32±4
38±13
5S±5
68±5
74±3
1
2
S5
• R24 antimelanoma monoclonal antibody.
40
71 75
/
o
/
/
Eu
...................
"'...-
.
.................... 'I
140
Fig. 3. Kinetics of complement mediated cytolysis measured by europium or SICr release. Replicate batches of SK·mel-28
melanoma cells (5000 cells/well) were labelled as indicated and cytolysis in the presence of R24 antimelanoma antibody (0.2 #,g/well) plus complement measured. All assays were performed concurrently on the same days.
17 100
•is>...
()
~
III
es
'#.
80
10
40
"'0
to oL---~----~--~----~--~--~ 0.00 0.01 O.O! 0.01 0.10 0.10
ANTIBODY DOSE (ut' •••) Fig. 4. Comparison of assays of complement mediated cytolysis by europium or Ster release or trypan blue exclusion. Replicate batches of SK-mel-28 melanoma cells (5000 cells/well) were labelled as indicated and cytolysis in the presence of increasing concentrations of R24 antimelanoma antibody plus complement measured. All assays were performed concurrently on the same days.
was released within 20 min; whereas the comparable figure for release of Ster was over 80 min. To compare the sensitivity of europium release to sl chromium release and dye exclusion assays, a pool of target cells was divided into three aliquots and labelled with europium, Ster or left unlabelled for trypan blue exclusion studies. All three batches were then tested on the same day for complement-mediated cytolysiS with increasing doses of R24 antibody. The results are illustrated in Fig. 4. Percent specific cytolysis at 2 h was greatest when measured by europium release (71 ± 10% SO at 0.20 JLg/well antibody concentration) vs. 60 ± 1% SO for trypan blue exclusion and 36 ± 6% SO for Ster release. The minimum level of antibody (defined as the antibody dose required to induce > 20% specific cytolysis) detectable by europium release assay was 0.01 JLg/well, which was three-fold lower than that detectable by trypan blue exclusion (0.03 JLg/well) and five-fold lower than that detectable by Ster release assay (0.05 JLg/well).
Discussion
The results of this study show that europium release provides a sensitive, rapid, and repro-
ducible method to measure antibody dependent, complement mediated cytolysis. The method has several significant advantages over current methods of measuring cytolysis - it is faster, more sensitive, and safer. Complement mediated killing of target cells is most commonly assayed by measuring the release of Ster from damaged cells. The assay is simple and reproducible; but is lengthy as target cells must be labelled with Ster prior to each experiment, lacks sensitivity as Ster binds to cytoplasmic proteins and may not be released until the cells are completly lysed (Volgmann et ai., 1989), generates radioactive waste which is increasingly difficult to dispose, and poses a health hazard to individuals working with it. Another common method to detect cell damage is by dye exclusion, which measures the ability of viable cells to exclude certain dyes from penetrating into the cytoplasm. A variety of dyes are used, the most common being trypan blue. This method is simple and very sensitive; but the results must be read visually under a microscope a procedure which is lengthy and subjective. The use of europium diethylenetriaminopentaacetate (EuOPTA) as a marker of cell damage circumvents many of these problems. The procedure of europium release assays is similar to that of a conventional Ster release assay, with the major difference being in the use of a non-radioactive coumpound to label the cells and of timeresolved fluorometry (Soini et ai., 1983; Hemmila et ai., 1984) rather than a gamma counter to measure the released label. Europium is a lanthanide metal, which when chelated with a B-diketone turns into a highly fluorescent complex. Europium is inexpensive, inert (Hemmila et al., 1984), and non-toxic. It is assayed by time resolved fluoremetry which is very sensitive and rapid requiring only 1 s per tube (Blomberg et at, 1986a). A drawback is the relatively high price of time resolved fluorometers. Using this assay, we found that following incubation of europium labelled melanoma cells with antimelanamoma antibody and complement europium was rapidly and selectively released. Sixty percent of total cellular europium was released in 15 min and almost 90% in 4 h. Spontaneous release of europium by cells incubated similarly
18
with complement only was 20% in 15 min and 38% in 4 h. The spontaneous release of europium was similar to that of 51Cr. Specific release of europium (calculated by subtracting the amount released by cells incubated with complement only from that released by cells incubated with antibody plus complement) was detectable after 15 min of incubation and increased gradually with time to reach a maximum value after 4 h. The ratio of specifically to spontaneously released europium was maximal after 2 h of incubation. By contrast it took 30 min to first detect specific release of StCr. Others have similarly shown that there is a lag period of approximately 25 min before SICr is rapidly released by damaged cells (Blomberg et aI., 1896a). 50% of the maximum amount of europium that could be specifically released by antibody + complement was released within 20 min; whereas the comparable figure for release of SICr in parallel experiments was over 80 min. As a result of the rapid release of europium from damaged cells, the time required to perform an assay was reduced by half in comparison to 51Cr release assay. Another major reduction in the time required to perform europium release assays is achieved because europium is not radioactive and does not decay. As a result, target cells can be labelled with europium in batches and stored at - 70"C until needed, obviating the need to label target cells prior to each assay. As a consequence europium assays can easily be completed in half a day as opposed to a whole day required to perform a standard SICr release assay. The release of europium was reproducible and quantitative. The standard deviation of replicate experiments performed under similar conditions on different days using the same pool of prelabelled target cells was usually in the range of 0.6-6%. Reproducibility is further enhanced by the ability to pre-label target cells in large batches and storing them until needed; which provides a more standard source of target cells. There was an essentially linear relationship between the amount of europium specifically released and the amount of antibody present in the assay over a 2 log antibody concentration range; indicating that the result of the assay provides a quantitative measure of antibody level.
Europium release provided a sensitive assay of cells damage. In parallel experiments, the minimum level of antibody detected by europiuID release assay was five times less than that detected by SICr release assay and three times less than that detected by trypan blue exclusion. The greater sensitivity is probably the result of several factors. The detection of fluorescence by time resolved fluorometry is extremely sensitive (Jackson et aI., 1986). SICr is more reactive than europium and binds to cytoplasmic components of cells. As a result, cells need to be more extensively damaged before StCr can be released (Voigmann et al., 1989), the rate of release of StCr is slower as discussed above, and the total amount which is released is less than that of europium. In our experiments, maximum specific release of europium was almost twice that of StCr. In summary, europium release is a more sensitive, simpler and faster method of measuring cell death induced by complement-mediated cytolysis than conventional StCr release or trypan blue exclusion methods.
Acknowledgements This work was supported in part by USPHS Research Grants AR 27663, AR 39749, CA53468-01Al, and FD-R-000632.
References Blomberg, K., Grenbera, C., Hemmila, I. and Lovaren, T. (1986a) Europium-labelled target cells in an assay of natu_ ral killer cell activity. I. A novel non-radioactivity method based on time resolved fluorescence. J. Immunol. Methods 86,225. Blomberg, K., Grenberg, C., Hemmila, I. and Lovaren, T. (1986b) Europium-labelled target cells in an assay of natural killer cell activity. II. Significance and specificity of the method. J. Immunol. Methods 92, 117. Goodman, H.S. (1961) A general method for the quantitatioQ of immune cytolysis. Nature 190, 269. Grenberg, C., Blombera, K., Hemmila, I. and Lovaren, T. (1988) Determination of cytotoxic T lymphocyte activity by time-resolved fluorometry using europium-labelled concanavalin A-stimulated cells u target. J. Immunol. Methods 114, 191.
19 Hemmila, I., Dakubu, S., Mukkala, V.M., Siitari, H. and Lovgren, T. (1984) Europium as a label in time-resolved immunofluorometric assays. Anal. Biochem. 137, 335. Jackson, T.M. and Ekins, R.P. (1986) Theoretical limitations on immunoassay sensitivity: current practice and potential advantages of fluorescence Eu-chelates as non-radioisotopic tracers. J. Imrnunol. Methods 87, 13. Maley, D.T. and Simon, P. (1990) Cytotoxicity assays using cryopreserved target cells pre-labelled with the fluorescent marker europium. J. Immunol. Methods 134, 61-70. Moen, J.E. and Warnaar, S.O. (1983) Quantitation of cell surface antigens by antibody-complement-mediated cytotoxicity. Methods Enzymol. 93, 253. Russell, J.H., Masakowski, V.R. and Dobos, C.B. (1980) Mechanisms of immune lysis. I. Physiological distinction
between target cell death mediated by cytotoxic T lymphocytes and antibody plus complement. J. Immunol. 124, 1100.
Schlaser, S.I. and Adams, A.C. (1983) Use of dyes and radio isotopic markers in cytotoxicity tests. Methods Enzymol. 93,233. Scomik, J.C. and Klein, P.A. (1978) Antibody-dependent lysis of tumor cells in vitro. II. Elimination of chromium-51 as a measurement of cytolysis. J. Natl. Cancer Inst. 61, 1149. Soini, E. and Kojola, H. (1983) Time-resolved fluorometer for lanthanide chelates - a new generation of non-isotopic immunoassays. Clin. Chern. 29, 65. Volgmann, T., Klein-Struckmeier, A. and Mohr, H. (1989) A fluorescence-based assay for quantitation of lymphokineactivated killer cell activity. J. Immunol. Methods 119,45.
13
JIM 06182
An improved europium release assay for complement-mediated cytolysis Jian Cui and Jean-Oaude Bystryn Department of Dermat%gy, New York University School of Medicine, New York, NY, USA (Received 7 June 1991, revised received 21 August 1991, accepted 4 September 1991)
An improved assay for complement-mediated cytolysis is described. The target cells are labeled with europium complexed to diethylenetriaminopentaacetate (Eu-DTPA). Cytolysis caused by antibody plus complement leads to the release of the Eu-DTPA complex which is then formed into a highly fluorescent chelate by the addition of 2-naphthoyltrifluoroacetone (2-NTA). The amount of europium chelate formed - a measurement of cell death - is then quantified with a time-resolved fluorometer. The results of the assay are reproducible. Complement-mediated cytolysis when measured by europium release was five times more sensitive than when measured by conventional Ster release and three times than when measured by trypan blue exclusion. Because europium does not decay, target cells can be labelled in batches and stored frozen until use, which speeds and simplifies the assay. Thus, europium release assay is a simple and quantitative method to measure complement-mediated cytolysis which is sensitive and more rapid than conventional assays. Key words: Europium; Complement; Cytotoxicity; Antibody; Cytolysis; Melanoma
Introduction The traditional method to measure complement-mediated cytolysis is by dye exclusion, a procedure which is read manually under a microscope and so is lengthy and the results subjective. A more quantitative method to measure immune cytolysis based on the release of slchromium was introduced by Goodman (1961) and applied to measure T cell and complement mediated cytolysis (Scomik et al., 1978; Russell et al., 1980; Moen et al., 1983; Schlaser et al., 1983). Though more accurate than dye exclusion, Sler release
Correspondence to: J.-C. Bystryn, NYU School of Medicine, 560 First Avenue, New York, NY 10016, USA
has the disadvantages that target cells must be labelled prior to each experiment, a lengthy procedure; and that radioactive waste is generated. In addition, chromium binds to cytoplasmic proteins by ionic or other forces and this may delay or decrease its release from damaged cells and thus reduce the sensitivity of the assay. These problems have stimulated attempts to develop non-radioactive, sensitive and rapid assays for cytolysis. Recently, Blomberg and co-workers (1986a,b) have described an assay to measure cell-mediated cytolysis in which the target cells are labelled with a europium-diethylenetriaminopentaacetate (Eu 3 +-DTPA) complex. After release of the complex from lysed cells, Eu3+ is chelated with 2-naphthoyltrifluoroacetone (2NTA) to form a highly fluorescent chelate which
14
can be detected with a fluorometer. Some biological tissues can release fluorescence, but the decay time of this biological fluorescence is only 10-20 ns. By contrast, the fluorescence decay time of Eu chelate is in the range of 100-1000 ns and can be quantified and distinguished from the biological fluorescence by time-resolved fluorometry (Soini et aI., 1983; Jackson et aI., 1986). When applied to measure cell-mediated cytolysis, this assay is more sensitive and faster than 51Cr release assay (Hemmila et aI., 1984; Blomberg et aI., 1986a,b; Grenberg et aI., 1988; Volgmann et aI., 1989; Maley et aI., 1990). This study was conducted to adopt the europium release method to measure antibody dependent complement-mediated cytolysis. Materials and methods Target cells Human melanoma cells SK-mel-28 (provided by Dr. J. Fogh, Sloan-Kettering Institute, Rye, NY) were used as the target. The cells were cultured in DME medium (Gibco, Grand Island, NY) with 10% fetal calf serum (Gibco). Antibody and complement R24 monoclonal antibody (a generous gift of Dr. Allan Houghton, Memorial Sloan-Kettering Cancer Center, New York) to the GD3 surface antigen of SK-mel-28 was used as the antibody source. Human complement sera (Sigma, St. Louis, MO) was used as the complement source. Labelling of target cells with europium Cells were washed 2 X with 50 ml buffer I (pH 7.4) consisting of 50 mM Hepes, 93 mM NaCl, 5 mM KCI, and 2 mM MgCl z and then suspended at approximately 5 X 10 6 cells/ml in labelling buffer (buffer I supplemented with 200 IL M EuCI 3 (Fluka, Ronkonkoma, NY), 1.0 mM diethylenetriaminopentaacetate (DTPAXSigma, St. Louis) and 500 ILg/ml dextran sulfate (MW 500,000, Pharmacia Fine Chemicals, Piscataway, NJ). After incubation at room temperature for 15 min with occasional shaking, the labelling reaction was stopped by the addition of 4 ILI/ml of 1 M CaCl z solution. After 5 min the cells were washed 2 X
with 50 ml buffer III (buffer I supplemented with 2 mM CaCI z and 10 mM glucose), then 2 X with 50 ml DME culture medium. Labelled cells were resuspended in DME culture medium containing 10% DMSO at 10 6 cells/ml, then 1 ml aliquots dispensed into 2 ml freezer vials (LOP, North Haledon, NJ) and frozen at - SOOC until use, usually within 4 months. For use, the required number of vials were thawed in a 37'C water bath. The cells were washed 2 X with 50 mI HBSS, 2 X with 50 ml DME culture medium, counted by trypan blue exclusion, and resuspended in DME culture medium at 105 live cells/mt. Labelling of target cells with "chromium Approximately 10 7 cells were incubated with 100 ILCi of 51Cr (Du Pont, Wilmington, DE) in 0.5 ml of DME culture medium for 2 h at 37'"C with frequent shaking, washed 4 X with 50 ml DME medium, and adjusted to a concentration of 10 5 live cells/ml medium. Complement-mediated cytolysis 50 ILl of europium or 5lCr-labelled cells (lOs cell/mJ) and 50 ILl of diluted R24 antibody were pipetted into the wells of round-bottomed 96 well microtiter tissue culture plates and shaken in an icebath for 45 min. Then 100 ILl of 1/5 diluted human complement serum was added to each well and incubated at 37°C. After 2 h, the plates were centrifuged at 100 X g for 5 min, and 20 ILl of supernatant from each well transferred to the wells of flat-bottomed microtiter strips (FlOW Laboratories, McLean, VA) containing 200 #,1 of enhancement solution consisting of 0.1% (v/v) Triton X-100, 15ILM 2-naphthoyltrifluoroacetone (2-NTA) and 50 ILM Tri-n-octylphosphine oxide (TOPO) (Wallac, Turku, Finland). After mixing at room temperature for 5 min, fluorescence was measured in a time-resolved fluorometer (Arcus 1230, LKB-Wallac, Turku, Finland). The super_ natant of 51Cr-labelled cells was collected by har_ vesting frames (Skatron, Sterling, VA) and the radioactivity counted in a gamma-counter (l282 CompuGamma, LKB-Wallac). All assays were performed in triplicate. Controls included cells incubated only with culture
15
medium, complement, or antibody. Percent cytolysis was calculated from:
100r---------------~
ao~---------~~~~-~
................................
% specific lysis
.o~-~~----~--~----~
experiment release - spontaneous release
... ...
- - - - : - - - - - - - - - - x 100
40~~~~~--------~o.---~
maximum release - spontaneous releases
In both europium and SICr release assays, spontaneous release was the amount of label released by target cells incubated with complement only. Maximum release for europium was the amount released from target cells lysed with 0.1 % Triton X-1OO (Sigma, St. Louis, MO) and for SICr that released by target cells lysed with 10% Linbro (Flow laboratories, McLean, VA). To measure cytolysis by dye exclusion, the cells following incubation with complement were spun down at 100 x g for 5 min and the supernatant removed. The cells were resuspended in 30 J.Ll of 0.4% trypan blue (Gibco) and the ratio of live/dead cells counted under a microscope.
Results Measurement of complement mediated cytolysis by europium release
The effect of incubating europium labelled target cells (5000 /well) for increasing periods of time with R24 antimelanoma antibody (0.20 J.Lg/well) and complement is illustrated in Fig. l. There was rapid release of europium during the first 15 min of incubation followed by a slower but continued release. 60% of total europium was released in the first 15 min of. incubation with antibody and complement and 87% after 4 h. By contrast, spontaneous release of europium from cells incubated only with complement was much slower - resulting in 23% release in 15 min and 38% in 4 h. Specific release increased steadily with time reaching a maximum of 78% at 4 h. The ratio of specifically to spontaneously released europium was greatest at 2 h of incubation, and this incubation time was used for all subsequent assays. To study the effect of target cell number on cell lysis, experiments were conducted in which a fIXed dose of R24 antimelanoma antibody (0.20
......._.._.-------- ----_._ -_....--
10~~~~-----------~
O~-~--~---~--~~-~-~ o ,. at .0 110 240 TIME OF Eu RELEASE (MINI
Fig. 1. Assay of complement mediated cytolysis by europium release. Replicate aliquots of europium-labelled SK-mel-28 melanoma cells (5000 celis/weI\) were incubated with R24 antimelanoma monoclonal antibody (0.2 ~g/weH) and complement (total release) or with complement only (spontaneous release) for increasing periods of time. Specific release was calculated by subtracting spontaneous release from total release. Results are mean of three experiments.
J.Lg/well) and complement was incubated with increasing numbers of cells for 2 h. The results are shown in Fig. 2. The proportion of europium spontaneously released was stable over a cell concentration of 20,Ooo-2500/well but increased rapidly when the number or' target cells was reduced to 1250/well. Specific cytolysis increased gradually with decreasing numbers of target cells reaching a peak value of 100% at a target cell number of 1250/well. The ratio of specific to 100
ao
!
i
-
.0 40 ......... __ ••••• - ......... -----.- .. - - 4 ......
~~
10 0 10,000
10,000
1,000
1,100
1,110
CELL NUMBER/w,11
Fig. 2. Effect of number of target cell number on europium release assay. Increasing number of europium-labelled Skmel-28 melanoma cells (12S0-20,OOO cells/well) were incubated with R24 anti melanoma antibody (0.2 ~g/well) and complement or without complement (spontaneous release). Specific release was calculated as descripted. Results are mean of three experiments.
16 TABLE I REPRODUCIBILITY OF EUROPIUM RELEASE ASSAY Specific cytolysis (%)
Experiment
Incubation time (min) 15
30
60
120
240
3
35 35 34
50 58 46
60 55 57
62 66 60
85 73 76
Average±SD
35 ± 0.6
51±6
57±2.5
63±3
78±6
1
2
spontaneous release was greatest at 5000 cells/well and was used in all subsequent assays.
antibody concentrations and 13% at one concentration (Table 11).
Reproducibility of europium release assay To measure the reproducibility of the assay, replicate experiments were performed on three different days during a period of 12 weeks in which 5000 melanoma cells/well were incubated with a fixed dose of R24 antibody (0.20 JLg/well) and complement for different periods of time. The results are shown in Table I. As previously noted, specific cytolysis increased steadily with time in all three experiments. The standard deviation of specific release at each time point was between 0.6 and 6%. In a subsequent study, the reproducibility of the assay was examined by incubating increasing doses of antibody (0.01-0.20 JLg/well) with a fixed number of target cells (5000 /well) for a fIXed period of time (2 h). In three different experiments, the standard deviation of specific release was 3-5% at four of the
Comparison of europium release, 51Chromium release and dye exclusion assays Time sequence studies were conducted concurrently with europium and St Cr-Iabelled cells. Spontaneous release of europium and StCr was similar, while specific release of europium is more rapid than that of StCr (Fig. 3). Following 30 min of incubation, 50% of europium was specifically released vs only 10% of StCr. The amount of europium released was also greater than StCr. Maximum specific release of europium after 4 b of incubation was 78% vs. 43% for StCr. 50% of the maximum amount of europium that could be specifically released by antibody plus complement 100
10 10
~
TABLE II
RELATION BElWEEN ANTIBODY CONCENTRATION AND CYTOLYSIS Experiment
10
Specific cytolysis (%) Antibody • (#,g/welJ)
o
0.01
0.02
0.05
0.10
0.20
28 33 53
50 59
63 68 73
77
3
31 37 30
Average±SD
32±4
38±13
5S±5
68±5
74±3
1
2
S5
• R24 antimelanoma monoclonal antibody.
40
71 75
/
o
/
/
Eu
...................
"'...-
.
.................... 'I
140
Fig. 3. Kinetics of complement mediated cytolysis measured by europium or SICr release. Replicate batches of SK·mel-28
melanoma cells (5000 cells/well) were labelled as indicated and cytolysis in the presence of R24 antimelanoma antibody (0.2 #,g/well) plus complement measured. All assays were performed concurrently on the same days.
17 100
•is>...
()
~
III
es
'#.
80
10
40
"'0
to oL---~----~--~----~--~--~ 0.00 0.01 O.O! 0.01 0.10 0.10
ANTIBODY DOSE (ut' •••) Fig. 4. Comparison of assays of complement mediated cytolysis by europium or Ster release or trypan blue exclusion. Replicate batches of SK-mel-28 melanoma cells (5000 cells/well) were labelled as indicated and cytolysis in the presence of increasing concentrations of R24 antimelanoma antibody plus complement measured. All assays were performed concurrently on the same days.
was released within 20 min; whereas the comparable figure for release of Ster was over 80 min. To compare the sensitivity of europium release to sl chromium release and dye exclusion assays, a pool of target cells was divided into three aliquots and labelled with europium, Ster or left unlabelled for trypan blue exclusion studies. All three batches were then tested on the same day for complement-mediated cytolysiS with increasing doses of R24 antibody. The results are illustrated in Fig. 4. Percent specific cytolysis at 2 h was greatest when measured by europium release (71 ± 10% SO at 0.20 JLg/well antibody concentration) vs. 60 ± 1% SO for trypan blue exclusion and 36 ± 6% SO for Ster release. The minimum level of antibody (defined as the antibody dose required to induce > 20% specific cytolysis) detectable by europium release assay was 0.01 JLg/well, which was three-fold lower than that detectable by trypan blue exclusion (0.03 JLg/well) and five-fold lower than that detectable by Ster release assay (0.05 JLg/well).
Discussion
The results of this study show that europium release provides a sensitive, rapid, and repro-
ducible method to measure antibody dependent, complement mediated cytolysis. The method has several significant advantages over current methods of measuring cytolysis - it is faster, more sensitive, and safer. Complement mediated killing of target cells is most commonly assayed by measuring the release of Ster from damaged cells. The assay is simple and reproducible; but is lengthy as target cells must be labelled with Ster prior to each experiment, lacks sensitivity as Ster binds to cytoplasmic proteins and may not be released until the cells are completly lysed (Volgmann et ai., 1989), generates radioactive waste which is increasingly difficult to dispose, and poses a health hazard to individuals working with it. Another common method to detect cell damage is by dye exclusion, which measures the ability of viable cells to exclude certain dyes from penetrating into the cytoplasm. A variety of dyes are used, the most common being trypan blue. This method is simple and very sensitive; but the results must be read visually under a microscope a procedure which is lengthy and subjective. The use of europium diethylenetriaminopentaacetate (EuOPTA) as a marker of cell damage circumvents many of these problems. The procedure of europium release assays is similar to that of a conventional Ster release assay, with the major difference being in the use of a non-radioactive coumpound to label the cells and of timeresolved fluorometry (Soini et ai., 1983; Hemmila et ai., 1984) rather than a gamma counter to measure the released label. Europium is a lanthanide metal, which when chelated with a B-diketone turns into a highly fluorescent complex. Europium is inexpensive, inert (Hemmila et al., 1984), and non-toxic. It is assayed by time resolved fluoremetry which is very sensitive and rapid requiring only 1 s per tube (Blomberg et at, 1986a). A drawback is the relatively high price of time resolved fluorometers. Using this assay, we found that following incubation of europium labelled melanoma cells with antimelanamoma antibody and complement europium was rapidly and selectively released. Sixty percent of total cellular europium was released in 15 min and almost 90% in 4 h. Spontaneous release of europium by cells incubated similarly
18
with complement only was 20% in 15 min and 38% in 4 h. The spontaneous release of europium was similar to that of 51Cr. Specific release of europium (calculated by subtracting the amount released by cells incubated with complement only from that released by cells incubated with antibody plus complement) was detectable after 15 min of incubation and increased gradually with time to reach a maximum value after 4 h. The ratio of specifically to spontaneously released europium was maximal after 2 h of incubation. By contrast it took 30 min to first detect specific release of StCr. Others have similarly shown that there is a lag period of approximately 25 min before SICr is rapidly released by damaged cells (Blomberg et aI., 1896a). 50% of the maximum amount of europium that could be specifically released by antibody + complement was released within 20 min; whereas the comparable figure for release of SICr in parallel experiments was over 80 min. As a result of the rapid release of europium from damaged cells, the time required to perform an assay was reduced by half in comparison to 51Cr release assay. Another major reduction in the time required to perform europium release assays is achieved because europium is not radioactive and does not decay. As a result, target cells can be labelled with europium in batches and stored at - 70"C until needed, obviating the need to label target cells prior to each assay. As a consequence europium assays can easily be completed in half a day as opposed to a whole day required to perform a standard SICr release assay. The release of europium was reproducible and quantitative. The standard deviation of replicate experiments performed under similar conditions on different days using the same pool of prelabelled target cells was usually in the range of 0.6-6%. Reproducibility is further enhanced by the ability to pre-label target cells in large batches and storing them until needed; which provides a more standard source of target cells. There was an essentially linear relationship between the amount of europium specifically released and the amount of antibody present in the assay over a 2 log antibody concentration range; indicating that the result of the assay provides a quantitative measure of antibody level.
Europium release provided a sensitive assay of cells damage. In parallel experiments, the minimum level of antibody detected by europiuID release assay was five times less than that detected by SICr release assay and three times less than that detected by trypan blue exclusion. The greater sensitivity is probably the result of several factors. The detection of fluorescence by time resolved fluorometry is extremely sensitive (Jackson et aI., 1986). SICr is more reactive than europium and binds to cytoplasmic components of cells. As a result, cells need to be more extensively damaged before StCr can be released (Voigmann et al., 1989), the rate of release of StCr is slower as discussed above, and the total amount which is released is less than that of europium. In our experiments, maximum specific release of europium was almost twice that of StCr. In summary, europium release is a more sensitive, simpler and faster method of measuring cell death induced by complement-mediated cytolysis than conventional StCr release or trypan blue exclusion methods.
Acknowledgements This work was supported in part by USPHS Research Grants AR 27663, AR 39749, CA53468-01Al, and FD-R-000632.
References Blomberg, K., Grenbera, C., Hemmila, I. and Lovaren, T. (1986a) Europium-labelled target cells in an assay of natu_ ral killer cell activity. I. A novel non-radioactivity method based on time resolved fluorescence. J. Immunol. Methods 86,225. Blomberg, K., Grenberg, C., Hemmila, I. and Lovaren, T. (1986b) Europium-labelled target cells in an assay of natural killer cell activity. II. Significance and specificity of the method. J. Immunol. Methods 92, 117. Goodman, H.S. (1961) A general method for the quantitatioQ of immune cytolysis. Nature 190, 269. Grenberg, C., Blombera, K., Hemmila, I. and Lovaren, T. (1988) Determination of cytotoxic T lymphocyte activity by time-resolved fluorometry using europium-labelled concanavalin A-stimulated cells u target. J. Immunol. Methods 114, 191.
19 Hemmila, I., Dakubu, S., Mukkala, V.M., Siitari, H. and Lovgren, T. (1984) Europium as a label in time-resolved immunofluorometric assays. Anal. Biochem. 137, 335. Jackson, T.M. and Ekins, R.P. (1986) Theoretical limitations on immunoassay sensitivity: current practice and potential advantages of fluorescence Eu-chelates as non-radioisotopic tracers. J. Imrnunol. Methods 87, 13. Maley, D.T. and Simon, P. (1990) Cytotoxicity assays using cryopreserved target cells pre-labelled with the fluorescent marker europium. J. Immunol. Methods 134, 61-70. Moen, J.E. and Warnaar, S.O. (1983) Quantitation of cell surface antigens by antibody-complement-mediated cytotoxicity. Methods Enzymol. 93, 253. Russell, J.H., Masakowski, V.R. and Dobos, C.B. (1980) Mechanisms of immune lysis. I. Physiological distinction
between target cell death mediated by cytotoxic T lymphocytes and antibody plus complement. J. Immunol. 124, 1100.
Schlaser, S.I. and Adams, A.C. (1983) Use of dyes and radio isotopic markers in cytotoxicity tests. Methods Enzymol. 93,233. Scomik, J.C. and Klein, P.A. (1978) Antibody-dependent lysis of tumor cells in vitro. II. Elimination of chromium-51 as a measurement of cytolysis. J. Natl. Cancer Inst. 61, 1149. Soini, E. and Kojola, H. (1983) Time-resolved fluorometer for lanthanide chelates - a new generation of non-isotopic immunoassays. Clin. Chern. 29, 65. Volgmann, T., Klein-Struckmeier, A. and Mohr, H. (1989) A fluorescence-based assay for quantitation of lymphokineactivated killer cell activity. J. Immunol. Methods 119,45.
