- Email: [email protected]
Standardization of a micro-cytotoxicity assay for human natural killer cell lytic activity
JOURNALOF IMMUNOLOGICAL METHODS ELSEVIER
Journal of Immunological Methods 172 (1994) 173-178
Standardization of a micro-cytotoxicity assay for human natural killer cell lytic activity E. M a r i a n i .,a,b M . C . G . M o n a c o a, S. S g o b b i a, J.F. d e Z w a r t a, A . R . M a r i a n i c A. Facchini b a Laboratorio di Immunologia e Genetica, Istituto di Ricerca Codivilla Putti I.O.R., Bologna, Italy, b Istituto di Clinica Medica e Gastroenterologia, Universit?z di Bologna, Bologna, Italy, c Istituto di Anatomia Umana Normale, Universitdt di Bologna, Bologna, Italy
Received 2 August 1993; revised received 4 January 1994; accepted 14 February 1994
Abstract Cytotoxicity assays are widely used to evaluate the functional activity o f NK and T cells against tumour target cells and the release of radioactive sodium chromate from labelled target cells is still the most commonly used marker of target lysis in culture supernatants. We describe here the standardization of a micro-cytotoXicity test in which the number of cytolytic effector and tumour target cells have been decreased by a factor of 10. The release obtained by 500 tumour target cells was compared with the release obtained by 5000 target cells in the standard cytotoxicity assay for target : effector cell ratios from 1 : 1 to 1 : 100. Both 3' and 13 emissions of the 5~Cr isotope were evaluated to determine the assay release. The results obtained by the micro-cytotoxicity assay (500 target cells) were comparable to those of the standard assay (5000 target cells) and 51Cr release evaluation using the 3, cour~ter was the most sensitive method of determining lytic activity using 500 tumour target cells. 13 counter evaluation using solid phase scintillation was found to be a reproducible alternative method, even if the lyric curves cannot be compared with those obtained using the traditional method. Key words: Cytotoxicity assay; Lytic activity; Tumor target cell; Oytolytic effector cell; Natural killer cell
1. Introduction Cytotoxicity assays are widely used to e v a l u a t e the lytic activity of N K a n d T cells against t u m o u r
* Corresponding author. At: Laboratorio di Immunologia e Genetica, Istituto di Ricerca Codivilla Putti I.O.R., Via di Barbiano, 1/10, 40136 Bologna, Italy. Tel.: +39-51-6366803; Fax: +39-51-6366807.
target cells. Although new methods for the determination of lytic activity have been described over the years (Celada and Rotman 1967; Douglas et al., 1971; H o r a n and Kappler, 1977; Bonavida et al., 1983; Vargas-Cortes et al., 1983; Blomberg et al., 1986; Slezak and Horan, 1989; Papa et al., 1988; Vitale et al., 1991), they have not been widely adopted and radioactive sodium chromate (Na~lCrO4) release from labelled cells is still the most commonly used marker of target
0022-1759/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0022-1759(94)00061 -Z
174
E. Mariani et al. /Journal of lmmunological Methods 172 (1994) 173-178
cell lysis in culture supernatants (Brunner et al., 1968). In fact, the amount of 51Cr released by labelled target cells in a given time can be taken as a measure of the number of target cells lysed by cytolytic effector cells. Assaying the lytic activity of peripheral blood lymphocytes may present some limitations such as the high number of effector cells required in studies analysing poorly represented subpopulations (NK subsets, 3,/3 T cells, sorted or cloned lymphocytes), or when it is possible to collect only a minimal amount of blood, as in children or elderly people. In these situations, the possibility of using a low number of effector cells would greatly facilitate the performance of cytotoxicity assays. 51Chromium is a radioisotope which produces a 3' type disintegration with an energy of 0.323 MeV of radiation (Levi, 1963). The isotope is characterized by a 0.01 MeV mean energy of /3 radiation: that is about twice the mean energy of the /3 radiation of tritium (0.006 MeV) (Levi, 1963), the isotope most widely used to evaluate cell proliferative responses. This study aimed to standardize a cytotoxicity assay in which the number of cytolytic effector and tumour target cells could be decreased to minimal levels compatibile with the detection limits of the instrumentation. Given the physical characteristics of 51Cr, and the availability of a new solid scintillator (Ready Cap, Beckman) we also evaluated the possibility of improving the use of /3 emissions to determine isotope release in cytotoxicity assays.
2. Material and methods 2.1. Tumour target cell labelling
The K562 myeloid cell line was utilized as a tumour target for cytotoxicity assays. The cells were kept in culture in RPMI 1640 + 10% FCS, 4 mM glutamine, 100 U / m l penicillin and 10 /xg/ml streptomycin and used during the logarithmic growth phase. 51Chromium (Na~lCrO4) solution in normal saline was used to label target cells. 5~Cr was
obtained with a specific activity of 400-1200 Ci/g, (14.8-44.4 TBq) (NEN, Germany) and was used no later than 15 days after the reference day. Briefly 2 x 106 K562 cells were washed once with fresh medium and the supernatant completely removed and incubated with 100 /zCi of radioactive sodium chromate in differing volumes, according to the decay table. The cells were incubated at 37°C in 5% CO 2 for 1 h with occasional shaking at 10-15 min intervals. The target tumour cells were then washed three times at 4°C in cold RPMI 1640 + 10% FCS and the last washing performed just before seeding the cells at the optimal concentration. The viability of the cell preparation was determined by the eosin Y dye exclusion test and was greater than 95% after 51Cr incubation. 2.2. Mononuclear cell preparation
Blood was collected into heparin containing tubes (Liquemin, Roche, Switzerland) by venepuncture of the antecubital vein. Mononuclear ceils (MNC) were isolated by conventional density gradient and were washed twice in RPMI 1640 medium + 10% FCS (heat inactivated) (Gibco, USA), then used as cytolytic effector cells in the following assays. 2.3. N K cell cytotoxicity assays
NK cell lytic activity was tested against K562 tumour target cells using a 4 h 51Cr assay. The cytotoxicity assays were performed in V bottom 96 well microtitre plates (Nunc, Denmark) with a final volume of 150 /~l as previously described (Mariani et al., 1990). Each target:effector cell ratio (T : E) was seeded in triplicate using at least six replicate wells in order to evaluate spontaneous and maximum release. (1) Standard cytotoxicity assay. Various numbers of mononuclear effector cells from 5 × 105 to 5 )< 103 in 100 ~1 RPMI 1640 + 10% FCS and 5 x 103/50 /xl 51Cr labelled tumour target cells were seeded with a T : E ranging from 1:100 to 1:1. (2) Micro-cytotoxicity assay. A ten-fold lower number of cytolytic effector cells from 5 × 10 4 to
E. Marianiet al. /Journal of Immunological Methods 172 (1994) 173-178 5 X 102 in 100 ~1 RPMI 1640 + 10% FCS and a ten-fold lower number of 51Cr labelled target cells (5 x 102/50 ~I) were seeded with a maintained T : E ranging from 1 : 100 to 1 : 1 as in the standard assay. In both assays the plates were then centrifuged at 4°C for 7 min and incubated at 37°C in 5% CO 2 for 4 h.
175
target cells lysed by 1% Triton X-100 detergent and was close to 100%. In preliminary experiments the minimum optimal counting time was evaluated. This time was unified to 2 rain because the intratest coefficient of variation was lower than 3% both for/3 and 3' counters (data not shown).
2.5. Statistical analysis 2.4. Evaluation of 51Cr release After incubation and 7 min centrifugation at 4°C, 75/zl of culture supernatant were harvested and 51Cr release was determined for both the standard assay and the micro-assay using for each collected sample the following methods to evaluate emission. (1) 3' counter. Culture supernatants were collected into tubes and directly counted in a 3' counter (Beckman, Mod. 5500 B). Data were acquired in counts per minute (cpm).
(2) fl counter-solid phase scintillation (s.p.s.). Culture supernatants were collected on caps with a new solid scintillator, Ready Cap with Xtalscint solvent-free scintillation medium for nonvolatile micro volume samples (Ready Cap, Beckman) as replacement for liquid scintillation solutions (Wunderly, 1989). Caps were dried at 37°C, put in maxi scintillation vials without the addition of a liquid scintillation cocktail and counted for /3 emission in a/3 counter (Beckman Mod. LS 5000 CE). Data were recorded as a count rate (cpm). [3 counter-liquid phase scintillation (l.p.s.) was used as a control for solid phase scintillation. Culture supernatants collected in mini scintillation vials were supplemented with 3 ml of a liquid scintillation cocktail for aqueous and nonaqueous samples (Instagel, Packard) and counted for /3 emission in a fl counter (Beckman, Mod. LS 5000 CE). Data were acquired in units of disintegrations per minute (dpm). Specific 51Cr release was calculated as follows: (test r e l e a s e - spontaneous release)/(maximum release - spontaneous release) x 100 (Ortaldo et al., 1977; Facchini et al., 1987) where spontaneous release represented the 51Cr release from target tumour cells with medium alone and the maximum release was the 51Cr release from the
A variable number of cytotoxicity assays (at least eight) were compared for the different experimental conditions: standard and micro cytotoxicity assay for each method of 51Cr release evaluation. The results obtained were processed using the SPSS program, expressed as mean percentages + S D and compared using Student's t test for paired data.
3. Results
3.1. N K cell cytotoxic assays (1) Standard cytotoxicity assay. Using 5000 K562 as target cells, the curves of NK lytic activity showed different slopes depending on the method of detection used, The results obtained in the solid phase scintillation release assays (Ready Cap) were 5 - 1 5 % lower than those obtained using the y counter (Fig. 1). Similar release values were obtained only for target:effector cell ratios from 1 : 1 to 1 : 25 while the release values for the remaining target:effector cell ratios (1:50 and 1 : 100) were significantly different from those obtained in the y counter ( p < 0.05 at least). The release values obtained with Ready Cap were higher than those obtained with liquid phase scintillation ( p values ranging from < 0.05 to < 0.005) except for target : effector cell ratios of 1 : 1 and 1 : 3. We observed that the 51Cr release data obtained with liquid phase scintillation at 1:100 T : E were similar to the release obtained in both solid phase scintillation and y counter assays at 1:50 T : E . Since the same sample was evaluated by each method, the one logarithm shift in the
E. Mariani et al. /Journal of lrnmunological Methods 172 (1994) 173-178
176 100
5 0 0 0 target cells •
80
• B counter s.p.s. •
0 0
7 counter B counter I.p.s.
60
~r"
~3
3.2. Evaluation of 51Cr release: comparison between 5000 and 500 tumour target cells
40
20
1:100
1:50
1:25
1:12
1:6
1:3
1:1
Target : Effector cell ratios Fig. 1. Standard cytotoxicity assay. Cytotoxicity tests were performed in V bottom 96 well microtitre plates containing various numbers of effector cells from 5 × 105 to 5 × 103 and 5000 51Cr labelled K562 target cells (target : effector cell ratios from 1 : 100 to 1 : 1). Each point represents the means + SD of eight paired observations obtained by evaluation with (e) 3' counter, ( • )/3 counter solid phase scintillation (s.p.s.), ( • )/3 counter-liquid phase scintillation (l.p.s.). The differences between 3' counter and /3 counter s.p.s, were significant for 1:100, 1:50 T : E ratios ( p < 0.03) and not significant for 1:25, 1 : 12, 1:6, 1:3, 1 : 1 T : E ratios. The differences between s.p.s, and 1.p.s. were significant for 1:100 ( p < 0.01), 1:50 ( p < 0 . 0 0 5 ) , 1:25 and 1:12 ( p < 0 . 0 5 ) T : E ratios but not significant for the remaining ratios.
(1) y counter assays. The evaluation of 51Cr release with a 3' counter showed that the curve profiles obtained using 5000 or 500 K562 target cells overlapped (Fig. 3) for all the target:effector cell ratios tested. Maximum release obtained from 500 K562 target cells ranged from 1500 to 3600 cpm, with a spontaneous release that never exceeded 16% of the maximum release; background counts were about 100 cpm. The coefficient of variation of the cpm obtained by different tests for all T : E considered ranged from a minimum of 5.4 +_ 0.69% to a maximum of 8.8 + 0.90%. When the percentages of 5~Cr released by 500 target cells were plotted against those released by 5000 target cells for all target:effector cell ratios examined, a highly significant correlation was observed (r = 0.99; p < 0.001). (2) [3 counter-solid phase scintillation. The data obtained with solid phase scintillation were in 100 500 target cells o 7 counter A B counter s.p.s.
80
o B counter I.p.s.
0
release values may be due to the higher efficiency of solid phase scintillation compared to liquid phase scintillation. (2) Micro-cytotoxicity assay. The NK cell cytotoxicity assay against 500 K562 target cells further e m p h a s i z e d the d i f f e r e n c e s b e t w e e n the chromium release curves. Under these conditions the release values obtained with low T : E using solid phase [3 scintillation also became lower than in the y counter assay and the high T : E further increased their differences (from about 20% to 50%; from p < 0 . 0 0 5 to p < 0 . 0 0 1 for all the target: effector cell ratios considered) (Fig. 2). In the micro-cytotoxicity test there was also a shift in the release values, corresponding to one or two T : E ratios lower than those obtained with the counter assay. The values obtained with solid phase scintillation were similar to those obtained with liquid phase scintillation.
0 o
60
E:
(3
40
20
1:100
1:50
1:25
1:12
1:6
1:3
1:1
Target : Effector cell ratios Fig. 2. Micro-cytotoxicity assay. Cytotoxicity tests were performed in V bottom 96 well microtitre plates containing various numbers of effector cells form 5 × 104 to 5 × 102 and 500 51Cr labelled K562 target cells (target: effector cell ratios from 1:100 to 1:1). Each point represents the means + SD of 16 paired observations obtained by evaluation with (©) 3' counter, (zx)/3 counter-solid phase scintillation (s.p.s.), ([])/3 counter-liquid phase scintillation (1.p.s.). The differences between 3' counter and /3 counter s.p.s, were significant for all the T : E cell ratios from 1:100 to 1:1 (1:100 to 1:12 T : E p < 0 . 0 0 1 ; 1:6 T : E p < 0 . 0 0 2 ; 1:3 and 1:1 T : E p < 0 . 0 0 5 ) but not significant between s.p.s, and l.p.s.
E. Mariani et aL /Journal of Immunological Methods 172 (1994) 173-178 100 7 counter • 5 0 0 0 target cells o 5 0 0 target cells
80
~
6o
(3
4o
177
phase scintillation assays, 500 target cell release was also significantly lower ( p < 0.05 at least) than 5000 target release (for T : E from 1 : 6 to 1:100) and correlated with 5000 target cell release (r = 0.98; p < 0.001). The coefficients of variation were similar to those obtained with liquid scintillation (from 5 . 9 + 1.36% to 11.3_+ 2.07%).
20
4. D i s c u s s i o n 0 1:100
1:50
1:25
1:12
1:6
1:3
1:1
Target : Effector cell ratios
Fig. 3. ~, counter: evaluation of 5tCr release. Each point represents the mean + SD of 44 paired observations obtained by evaluation with 3' counter of the experimental conditions of standard (e) and micro (©)-cytotoxicity assays. No difference was observed between the assays for all the target: effector cell ratios tested.
general lower than those obtained with the 3' counter but were slightly higher than the values obtained with liquid scintillation, in particular when 5000 K562 were used (Fig. 4). In solid
100 g counter s.p.s. • 5 0 0 0 target cells A 5 0 0 target cells
80
60 IB
¢r"
(3
P
40
20
1:100
1:50
1:25
1:12
1:6
1:3
1:1
Target : Effector cell ratios
Fig. 4./3 counter-solid phase scintillation (s.p.s.): evaluation of 51Cr release. Each point represents the mean+SD of 16 paired observations obtained by the evaluation with/3 counter and s.p.s, of the experimental conditions of standard ( • ) and micro (A)-cytotoxicity assays. The differences between 5000 and 500 targets were significant for T:E of 1:100 (p < 0.01); for 1:50, 1:25 (p <0.05); for 1:12 (p < 0.02); for 1:6 ( p < 0.002) but not significant for T: E 1: 3 and 1 : l.
We have described a modification of the standard cytotoxicity assay for N K cell lytic activity (Brunner et al., 1968), using a ten-fold reduction in the n u m b e r of cytolytic effector cells and consequently of 5~Cr labelled K562 target cells. The test was improved by the use of an isotope with high specific activity and by carefully removing supernatant to avoid competition with cold sodium ions present in the culture medium. The optimal conditions which gave reproducibile release of the isotope and comparable data using 500 and 5000 target cells utilized the 3' emission from 51Cr in the culture supernatants. Although release curves with different slopes were obtained using /3 counter evaluation and 51Cr release showed different values (mainly for high target : effector cell ratios and for low ( ~ 500) target cell numbers), the solid phase scintillation assay proved to be generally reliable when 5000 target ceils were used. Yron and Shohat (1986) described a miniaturization of the standard 5~Cr release assay in which the main modification consisted of increasing the chromium uptake by Yac-1 target cells, by labelling these ceils with high specific activity 51Cr (350-600 m C i / m g ) in the presence of isotonic sucrose. Isotonic sucrose permits greater 5!Cr uptake into the cells and was added to increase the concentration of chromium ions n e a r the cell surface because of lack of competition by cold sodium ions. However, in this study the percentages of spontaneous chromium release and the standard deviations of the cytotoxic curves for low target effector cell ratios were very high. During our standardization experiments we also used chromium with a high specific activity (400-
178
E. Mariani et al. /Journal of lmmunological Methods 172 (1994) 173-178
1200 Ci/g), but the introduction of sucrose did not improve the maximum uptake of the isotope (data not shown). In conclusion, the results obtained in the micro-cytotoxicity assay using ten-fold fewer numbers of cytolytic effector and K562 target cells were comparable to those obtained using the standard assay. In addition we have demonstrated that StCr release evaluation in a y counter is the most sensitive method for determining lytic activity in a micro-cytotoxicity assay using 500 tumour target cells and we have utilized this method in the evaluation of lytic activity of poorly represented NK subsets present in the peripheral blood of the elderly. /3 counter evaluation with Ready Cap solid scintillator is a reproducible method, even though the lytic curves cannot be compared with those obtained by the traditional assay. However, it permits a combination of methodological approaches in which different parameters can be evaluated on the same cell population using different isotopes with /3 emission. Finally, solid scintillators have the advantage that they are easier to dispose of than liquid scintillators.
Acknowledgements This work was partially supported by grants from the Ministero Ricerca Scientifica e Tecnologica 60% and the Istituti Ortopedici Rizzoli, Progetti di Ricerca Corrente. Anne Collins revised the English text.
References Blomberg, K., Granberg, C., Hemmil/i, I. and L6vgren, T. (1986) Europium-labelled target cells in an assay of natural killer cell activity. II. A novel non-radioactive method base on time-resolved fluorescence. Significance and specificity of the method. J. Immunol. Methods 92, 117.
Bonavida, B., Bradley, T.P. and Grimm, E.A. (1983) The single cell assay in cell mediated cytotoxicity. Immunol. Today 4, 196. Brunner, K.T., Mauel, J., Cerottini, J.C. and Chapuis, B. (1968) Quantitative assay of the lytic interaction of immune lymphoid cells on 51Cr Cr-labeled allogeneic cells in vitro; inhibition by isoantibody and drugs. Immunology 14, 181. Celada, F. and Rotman, B. (1967) A fluorochromatic test for immunocytotoxicity against tumour cell and leukocytes in agarose plates. Proc. Natl. Acad. Sci. USA 57, 630. Douglas, K., Perkins, H.A., Cochrum, K.C. and Kountz, S.L. (1971) Comparison of the HL-A phenotypes of lymphocytes and kidney cells determined by the fluorochromasia cytotoxicity assay. Clin. Invest. 50, 274. Facchini, A., Mariani, E., Mariani, A.R., Papa, S., Vitale, M. and Manzoli, F.A. (1987) Increased number of circulating Leu 11+ (CD16) large granular lymphocytes and decreased NK activity during human ageing. Clin. Exp. Immunol. 68, 340. Horan, P.K. and Kappler, J.W. (1977) Automated fluorescent analysis for cytotoxicity assays. J. Immunol. Methods 18, 309. Levi, H. (1963) Documenta Geigy - Tables Scientifiques, 16th edn. J.R. Geigy D6partement Pharmaceutique, Basel, p. 277. Mariani, E., Roda, P., Mariani, A.R., Vitale, M., Degrassi, A., Papa, S. and Facchini, A. (1990) Age-associated changes in CD8+ and CD16+ cell reactivity: clonal analysis. Clin. Exp. Immunol. 81,479. Ortaldo, J.R., Bonnard, G.D. and Herbermann, R.B. (1977) Cytotoxic reactivity of human lymphocytes cultured in vitro. J. Immunol. 119, 1351. Papa, S., Vitale, M., Mariani, A.R., Roda, P., Facchini, A. and Manzoli, F.A. (1988) Natural killer function in flow cytometry. I. Evaluation of NK lytic activity on K562 cell line. J. Immunol. Methods 107, 73. Vargas-Cortes, M., Hellstrom, U. and Perlmann, P. (1983) Surface markers of human natural killer cells as analyzed in a modified single cell cytotoxicity assay on poly-L-lysine coated coverslips. J. Immunol. Methods 62, 88. Vitale, M., Zamai, L., Neri, L.M., Manzoli, L., Facchini, A. and Papa, S. (1991) Natural killer function in flow cytometry: identification of human lymphoid subsets able to bind to the NK sensitive target K562. Cytometry 12, 717. Wunderly, S.W. (1989) Solid scintillation counting: a new technique for measuring radiolabeled compounds. Appl. Radiat. Isot. 40, 569. Yron, I. and Shohat, L. (1986) Miniaturization of the standard 51Cr release assay for long term follow-up of NK activity of individual mice. J. Immunol. Methods 93, 193.
Journal of Immunological Methods 172 (1994) 173-178
Standardization of a micro-cytotoxicity assay for human natural killer cell lytic activity E. M a r i a n i .,a,b M . C . G . M o n a c o a, S. S g o b b i a, J.F. d e Z w a r t a, A . R . M a r i a n i c A. Facchini b a Laboratorio di Immunologia e Genetica, Istituto di Ricerca Codivilla Putti I.O.R., Bologna, Italy, b Istituto di Clinica Medica e Gastroenterologia, Universit?z di Bologna, Bologna, Italy, c Istituto di Anatomia Umana Normale, Universitdt di Bologna, Bologna, Italy
Received 2 August 1993; revised received 4 January 1994; accepted 14 February 1994
Abstract Cytotoxicity assays are widely used to evaluate the functional activity o f NK and T cells against tumour target cells and the release of radioactive sodium chromate from labelled target cells is still the most commonly used marker of target lysis in culture supernatants. We describe here the standardization of a micro-cytotoXicity test in which the number of cytolytic effector and tumour target cells have been decreased by a factor of 10. The release obtained by 500 tumour target cells was compared with the release obtained by 5000 target cells in the standard cytotoxicity assay for target : effector cell ratios from 1 : 1 to 1 : 100. Both 3' and 13 emissions of the 5~Cr isotope were evaluated to determine the assay release. The results obtained by the micro-cytotoxicity assay (500 target cells) were comparable to those of the standard assay (5000 target cells) and 51Cr release evaluation using the 3, cour~ter was the most sensitive method of determining lytic activity using 500 tumour target cells. 13 counter evaluation using solid phase scintillation was found to be a reproducible alternative method, even if the lyric curves cannot be compared with those obtained using the traditional method. Key words: Cytotoxicity assay; Lytic activity; Tumor target cell; Oytolytic effector cell; Natural killer cell
1. Introduction Cytotoxicity assays are widely used to e v a l u a t e the lytic activity of N K a n d T cells against t u m o u r
* Corresponding author. At: Laboratorio di Immunologia e Genetica, Istituto di Ricerca Codivilla Putti I.O.R., Via di Barbiano, 1/10, 40136 Bologna, Italy. Tel.: +39-51-6366803; Fax: +39-51-6366807.
target cells. Although new methods for the determination of lytic activity have been described over the years (Celada and Rotman 1967; Douglas et al., 1971; H o r a n and Kappler, 1977; Bonavida et al., 1983; Vargas-Cortes et al., 1983; Blomberg et al., 1986; Slezak and Horan, 1989; Papa et al., 1988; Vitale et al., 1991), they have not been widely adopted and radioactive sodium chromate (Na~lCrO4) release from labelled cells is still the most commonly used marker of target
0022-1759/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0022-1759(94)00061 -Z
174
E. Mariani et al. /Journal of lmmunological Methods 172 (1994) 173-178
cell lysis in culture supernatants (Brunner et al., 1968). In fact, the amount of 51Cr released by labelled target cells in a given time can be taken as a measure of the number of target cells lysed by cytolytic effector cells. Assaying the lytic activity of peripheral blood lymphocytes may present some limitations such as the high number of effector cells required in studies analysing poorly represented subpopulations (NK subsets, 3,/3 T cells, sorted or cloned lymphocytes), or when it is possible to collect only a minimal amount of blood, as in children or elderly people. In these situations, the possibility of using a low number of effector cells would greatly facilitate the performance of cytotoxicity assays. 51Chromium is a radioisotope which produces a 3' type disintegration with an energy of 0.323 MeV of radiation (Levi, 1963). The isotope is characterized by a 0.01 MeV mean energy of /3 radiation: that is about twice the mean energy of the /3 radiation of tritium (0.006 MeV) (Levi, 1963), the isotope most widely used to evaluate cell proliferative responses. This study aimed to standardize a cytotoxicity assay in which the number of cytolytic effector and tumour target cells could be decreased to minimal levels compatibile with the detection limits of the instrumentation. Given the physical characteristics of 51Cr, and the availability of a new solid scintillator (Ready Cap, Beckman) we also evaluated the possibility of improving the use of /3 emissions to determine isotope release in cytotoxicity assays.
2. Material and methods 2.1. Tumour target cell labelling
The K562 myeloid cell line was utilized as a tumour target for cytotoxicity assays. The cells were kept in culture in RPMI 1640 + 10% FCS, 4 mM glutamine, 100 U / m l penicillin and 10 /xg/ml streptomycin and used during the logarithmic growth phase. 51Chromium (Na~lCrO4) solution in normal saline was used to label target cells. 5~Cr was
obtained with a specific activity of 400-1200 Ci/g, (14.8-44.4 TBq) (NEN, Germany) and was used no later than 15 days after the reference day. Briefly 2 x 106 K562 cells were washed once with fresh medium and the supernatant completely removed and incubated with 100 /zCi of radioactive sodium chromate in differing volumes, according to the decay table. The cells were incubated at 37°C in 5% CO 2 for 1 h with occasional shaking at 10-15 min intervals. The target tumour cells were then washed three times at 4°C in cold RPMI 1640 + 10% FCS and the last washing performed just before seeding the cells at the optimal concentration. The viability of the cell preparation was determined by the eosin Y dye exclusion test and was greater than 95% after 51Cr incubation. 2.2. Mononuclear cell preparation
Blood was collected into heparin containing tubes (Liquemin, Roche, Switzerland) by venepuncture of the antecubital vein. Mononuclear ceils (MNC) were isolated by conventional density gradient and were washed twice in RPMI 1640 medium + 10% FCS (heat inactivated) (Gibco, USA), then used as cytolytic effector cells in the following assays. 2.3. N K cell cytotoxicity assays
NK cell lytic activity was tested against K562 tumour target cells using a 4 h 51Cr assay. The cytotoxicity assays were performed in V bottom 96 well microtitre plates (Nunc, Denmark) with a final volume of 150 /~l as previously described (Mariani et al., 1990). Each target:effector cell ratio (T : E) was seeded in triplicate using at least six replicate wells in order to evaluate spontaneous and maximum release. (1) Standard cytotoxicity assay. Various numbers of mononuclear effector cells from 5 × 105 to 5 )< 103 in 100 ~1 RPMI 1640 + 10% FCS and 5 x 103/50 /xl 51Cr labelled tumour target cells were seeded with a T : E ranging from 1:100 to 1:1. (2) Micro-cytotoxicity assay. A ten-fold lower number of cytolytic effector cells from 5 × 10 4 to
E. Marianiet al. /Journal of Immunological Methods 172 (1994) 173-178 5 X 102 in 100 ~1 RPMI 1640 + 10% FCS and a ten-fold lower number of 51Cr labelled target cells (5 x 102/50 ~I) were seeded with a maintained T : E ranging from 1 : 100 to 1 : 1 as in the standard assay. In both assays the plates were then centrifuged at 4°C for 7 min and incubated at 37°C in 5% CO 2 for 4 h.
175
target cells lysed by 1% Triton X-100 detergent and was close to 100%. In preliminary experiments the minimum optimal counting time was evaluated. This time was unified to 2 rain because the intratest coefficient of variation was lower than 3% both for/3 and 3' counters (data not shown).
2.5. Statistical analysis 2.4. Evaluation of 51Cr release After incubation and 7 min centrifugation at 4°C, 75/zl of culture supernatant were harvested and 51Cr release was determined for both the standard assay and the micro-assay using for each collected sample the following methods to evaluate emission. (1) 3' counter. Culture supernatants were collected into tubes and directly counted in a 3' counter (Beckman, Mod. 5500 B). Data were acquired in counts per minute (cpm).
(2) fl counter-solid phase scintillation (s.p.s.). Culture supernatants were collected on caps with a new solid scintillator, Ready Cap with Xtalscint solvent-free scintillation medium for nonvolatile micro volume samples (Ready Cap, Beckman) as replacement for liquid scintillation solutions (Wunderly, 1989). Caps were dried at 37°C, put in maxi scintillation vials without the addition of a liquid scintillation cocktail and counted for /3 emission in a/3 counter (Beckman Mod. LS 5000 CE). Data were recorded as a count rate (cpm). [3 counter-liquid phase scintillation (l.p.s.) was used as a control for solid phase scintillation. Culture supernatants collected in mini scintillation vials were supplemented with 3 ml of a liquid scintillation cocktail for aqueous and nonaqueous samples (Instagel, Packard) and counted for /3 emission in a fl counter (Beckman, Mod. LS 5000 CE). Data were acquired in units of disintegrations per minute (dpm). Specific 51Cr release was calculated as follows: (test r e l e a s e - spontaneous release)/(maximum release - spontaneous release) x 100 (Ortaldo et al., 1977; Facchini et al., 1987) where spontaneous release represented the 51Cr release from target tumour cells with medium alone and the maximum release was the 51Cr release from the
A variable number of cytotoxicity assays (at least eight) were compared for the different experimental conditions: standard and micro cytotoxicity assay for each method of 51Cr release evaluation. The results obtained were processed using the SPSS program, expressed as mean percentages + S D and compared using Student's t test for paired data.
3. Results
3.1. N K cell cytotoxic assays (1) Standard cytotoxicity assay. Using 5000 K562 as target cells, the curves of NK lytic activity showed different slopes depending on the method of detection used, The results obtained in the solid phase scintillation release assays (Ready Cap) were 5 - 1 5 % lower than those obtained using the y counter (Fig. 1). Similar release values were obtained only for target:effector cell ratios from 1 : 1 to 1 : 25 while the release values for the remaining target:effector cell ratios (1:50 and 1 : 100) were significantly different from those obtained in the y counter ( p < 0.05 at least). The release values obtained with Ready Cap were higher than those obtained with liquid phase scintillation ( p values ranging from < 0.05 to < 0.005) except for target : effector cell ratios of 1 : 1 and 1 : 3. We observed that the 51Cr release data obtained with liquid phase scintillation at 1:100 T : E were similar to the release obtained in both solid phase scintillation and y counter assays at 1:50 T : E . Since the same sample was evaluated by each method, the one logarithm shift in the
E. Mariani et al. /Journal of lrnmunological Methods 172 (1994) 173-178
176 100
5 0 0 0 target cells •
80
• B counter s.p.s. •
0 0
7 counter B counter I.p.s.
60
~r"
~3
3.2. Evaluation of 51Cr release: comparison between 5000 and 500 tumour target cells
40
20
1:100
1:50
1:25
1:12
1:6
1:3
1:1
Target : Effector cell ratios Fig. 1. Standard cytotoxicity assay. Cytotoxicity tests were performed in V bottom 96 well microtitre plates containing various numbers of effector cells from 5 × 105 to 5 × 103 and 5000 51Cr labelled K562 target cells (target : effector cell ratios from 1 : 100 to 1 : 1). Each point represents the means + SD of eight paired observations obtained by evaluation with (e) 3' counter, ( • )/3 counter solid phase scintillation (s.p.s.), ( • )/3 counter-liquid phase scintillation (l.p.s.). The differences between 3' counter and /3 counter s.p.s, were significant for 1:100, 1:50 T : E ratios ( p < 0.03) and not significant for 1:25, 1 : 12, 1:6, 1:3, 1 : 1 T : E ratios. The differences between s.p.s, and 1.p.s. were significant for 1:100 ( p < 0.01), 1:50 ( p < 0 . 0 0 5 ) , 1:25 and 1:12 ( p < 0 . 0 5 ) T : E ratios but not significant for the remaining ratios.
(1) y counter assays. The evaluation of 51Cr release with a 3' counter showed that the curve profiles obtained using 5000 or 500 K562 target cells overlapped (Fig. 3) for all the target:effector cell ratios tested. Maximum release obtained from 500 K562 target cells ranged from 1500 to 3600 cpm, with a spontaneous release that never exceeded 16% of the maximum release; background counts were about 100 cpm. The coefficient of variation of the cpm obtained by different tests for all T : E considered ranged from a minimum of 5.4 +_ 0.69% to a maximum of 8.8 + 0.90%. When the percentages of 5~Cr released by 500 target cells were plotted against those released by 5000 target cells for all target:effector cell ratios examined, a highly significant correlation was observed (r = 0.99; p < 0.001). (2) [3 counter-solid phase scintillation. The data obtained with solid phase scintillation were in 100 500 target cells o 7 counter A B counter s.p.s.
80
o B counter I.p.s.
0
release values may be due to the higher efficiency of solid phase scintillation compared to liquid phase scintillation. (2) Micro-cytotoxicity assay. The NK cell cytotoxicity assay against 500 K562 target cells further e m p h a s i z e d the d i f f e r e n c e s b e t w e e n the chromium release curves. Under these conditions the release values obtained with low T : E using solid phase [3 scintillation also became lower than in the y counter assay and the high T : E further increased their differences (from about 20% to 50%; from p < 0 . 0 0 5 to p < 0 . 0 0 1 for all the target: effector cell ratios considered) (Fig. 2). In the micro-cytotoxicity test there was also a shift in the release values, corresponding to one or two T : E ratios lower than those obtained with the counter assay. The values obtained with solid phase scintillation were similar to those obtained with liquid phase scintillation.
0 o
60
E:
(3
40
20
1:100
1:50
1:25
1:12
1:6
1:3
1:1
Target : Effector cell ratios Fig. 2. Micro-cytotoxicity assay. Cytotoxicity tests were performed in V bottom 96 well microtitre plates containing various numbers of effector cells form 5 × 104 to 5 × 102 and 500 51Cr labelled K562 target cells (target: effector cell ratios from 1:100 to 1:1). Each point represents the means + SD of 16 paired observations obtained by evaluation with (©) 3' counter, (zx)/3 counter-solid phase scintillation (s.p.s.), ([])/3 counter-liquid phase scintillation (1.p.s.). The differences between 3' counter and /3 counter s.p.s, were significant for all the T : E cell ratios from 1:100 to 1:1 (1:100 to 1:12 T : E p < 0 . 0 0 1 ; 1:6 T : E p < 0 . 0 0 2 ; 1:3 and 1:1 T : E p < 0 . 0 0 5 ) but not significant between s.p.s, and l.p.s.
E. Mariani et aL /Journal of Immunological Methods 172 (1994) 173-178 100 7 counter • 5 0 0 0 target cells o 5 0 0 target cells
80
~
6o
(3
4o
177
phase scintillation assays, 500 target cell release was also significantly lower ( p < 0.05 at least) than 5000 target release (for T : E from 1 : 6 to 1:100) and correlated with 5000 target cell release (r = 0.98; p < 0.001). The coefficients of variation were similar to those obtained with liquid scintillation (from 5 . 9 + 1.36% to 11.3_+ 2.07%).
20
4. D i s c u s s i o n 0 1:100
1:50
1:25
1:12
1:6
1:3
1:1
Target : Effector cell ratios
Fig. 3. ~, counter: evaluation of 5tCr release. Each point represents the mean + SD of 44 paired observations obtained by evaluation with 3' counter of the experimental conditions of standard (e) and micro (©)-cytotoxicity assays. No difference was observed between the assays for all the target: effector cell ratios tested.
general lower than those obtained with the 3' counter but were slightly higher than the values obtained with liquid scintillation, in particular when 5000 K562 were used (Fig. 4). In solid
100 g counter s.p.s. • 5 0 0 0 target cells A 5 0 0 target cells
80
60 IB
¢r"
(3
P
40
20
1:100
1:50
1:25
1:12
1:6
1:3
1:1
Target : Effector cell ratios
Fig. 4./3 counter-solid phase scintillation (s.p.s.): evaluation of 51Cr release. Each point represents the mean+SD of 16 paired observations obtained by the evaluation with/3 counter and s.p.s, of the experimental conditions of standard ( • ) and micro (A)-cytotoxicity assays. The differences between 5000 and 500 targets were significant for T:E of 1:100 (p < 0.01); for 1:50, 1:25 (p <0.05); for 1:12 (p < 0.02); for 1:6 ( p < 0.002) but not significant for T: E 1: 3 and 1 : l.
We have described a modification of the standard cytotoxicity assay for N K cell lytic activity (Brunner et al., 1968), using a ten-fold reduction in the n u m b e r of cytolytic effector cells and consequently of 5~Cr labelled K562 target cells. The test was improved by the use of an isotope with high specific activity and by carefully removing supernatant to avoid competition with cold sodium ions present in the culture medium. The optimal conditions which gave reproducibile release of the isotope and comparable data using 500 and 5000 target cells utilized the 3' emission from 51Cr in the culture supernatants. Although release curves with different slopes were obtained using /3 counter evaluation and 51Cr release showed different values (mainly for high target : effector cell ratios and for low ( ~ 500) target cell numbers), the solid phase scintillation assay proved to be generally reliable when 5000 target ceils were used. Yron and Shohat (1986) described a miniaturization of the standard 5~Cr release assay in which the main modification consisted of increasing the chromium uptake by Yac-1 target cells, by labelling these ceils with high specific activity 51Cr (350-600 m C i / m g ) in the presence of isotonic sucrose. Isotonic sucrose permits greater 5!Cr uptake into the cells and was added to increase the concentration of chromium ions n e a r the cell surface because of lack of competition by cold sodium ions. However, in this study the percentages of spontaneous chromium release and the standard deviations of the cytotoxic curves for low target effector cell ratios were very high. During our standardization experiments we also used chromium with a high specific activity (400-
178
E. Mariani et al. /Journal of lmmunological Methods 172 (1994) 173-178
1200 Ci/g), but the introduction of sucrose did not improve the maximum uptake of the isotope (data not shown). In conclusion, the results obtained in the micro-cytotoxicity assay using ten-fold fewer numbers of cytolytic effector and K562 target cells were comparable to those obtained using the standard assay. In addition we have demonstrated that StCr release evaluation in a y counter is the most sensitive method for determining lytic activity in a micro-cytotoxicity assay using 500 tumour target cells and we have utilized this method in the evaluation of lytic activity of poorly represented NK subsets present in the peripheral blood of the elderly. /3 counter evaluation with Ready Cap solid scintillator is a reproducible method, even though the lytic curves cannot be compared with those obtained by the traditional assay. However, it permits a combination of methodological approaches in which different parameters can be evaluated on the same cell population using different isotopes with /3 emission. Finally, solid scintillators have the advantage that they are easier to dispose of than liquid scintillators.
Acknowledgements This work was partially supported by grants from the Ministero Ricerca Scientifica e Tecnologica 60% and the Istituti Ortopedici Rizzoli, Progetti di Ricerca Corrente. Anne Collins revised the English text.
References Blomberg, K., Granberg, C., Hemmil/i, I. and L6vgren, T. (1986) Europium-labelled target cells in an assay of natural killer cell activity. II. A novel non-radioactive method base on time-resolved fluorescence. Significance and specificity of the method. J. Immunol. Methods 92, 117.
Bonavida, B., Bradley, T.P. and Grimm, E.A. (1983) The single cell assay in cell mediated cytotoxicity. Immunol. Today 4, 196. Brunner, K.T., Mauel, J., Cerottini, J.C. and Chapuis, B. (1968) Quantitative assay of the lytic interaction of immune lymphoid cells on 51Cr Cr-labeled allogeneic cells in vitro; inhibition by isoantibody and drugs. Immunology 14, 181. Celada, F. and Rotman, B. (1967) A fluorochromatic test for immunocytotoxicity against tumour cell and leukocytes in agarose plates. Proc. Natl. Acad. Sci. USA 57, 630. Douglas, K., Perkins, H.A., Cochrum, K.C. and Kountz, S.L. (1971) Comparison of the HL-A phenotypes of lymphocytes and kidney cells determined by the fluorochromasia cytotoxicity assay. Clin. Invest. 50, 274. Facchini, A., Mariani, E., Mariani, A.R., Papa, S., Vitale, M. and Manzoli, F.A. (1987) Increased number of circulating Leu 11+ (CD16) large granular lymphocytes and decreased NK activity during human ageing. Clin. Exp. Immunol. 68, 340. Horan, P.K. and Kappler, J.W. (1977) Automated fluorescent analysis for cytotoxicity assays. J. Immunol. Methods 18, 309. Levi, H. (1963) Documenta Geigy - Tables Scientifiques, 16th edn. J.R. Geigy D6partement Pharmaceutique, Basel, p. 277. Mariani, E., Roda, P., Mariani, A.R., Vitale, M., Degrassi, A., Papa, S. and Facchini, A. (1990) Age-associated changes in CD8+ and CD16+ cell reactivity: clonal analysis. Clin. Exp. Immunol. 81,479. Ortaldo, J.R., Bonnard, G.D. and Herbermann, R.B. (1977) Cytotoxic reactivity of human lymphocytes cultured in vitro. J. Immunol. 119, 1351. Papa, S., Vitale, M., Mariani, A.R., Roda, P., Facchini, A. and Manzoli, F.A. (1988) Natural killer function in flow cytometry. I. Evaluation of NK lytic activity on K562 cell line. J. Immunol. Methods 107, 73. Vargas-Cortes, M., Hellstrom, U. and Perlmann, P. (1983) Surface markers of human natural killer cells as analyzed in a modified single cell cytotoxicity assay on poly-L-lysine coated coverslips. J. Immunol. Methods 62, 88. Vitale, M., Zamai, L., Neri, L.M., Manzoli, L., Facchini, A. and Papa, S. (1991) Natural killer function in flow cytometry: identification of human lymphoid subsets able to bind to the NK sensitive target K562. Cytometry 12, 717. Wunderly, S.W. (1989) Solid scintillation counting: a new technique for measuring radiolabeled compounds. Appl. Radiat. Isot. 40, 569. Yron, I. and Shohat, L. (1986) Miniaturization of the standard 51Cr release assay for long term follow-up of NK activity of individual mice. J. Immunol. Methods 93, 193.