Discrepancy between direct and antibody-dependent cytotoxic activities of human LAK cells

Immunology Letters, 41 (1994) 13-17 0165 - 2478 / 94 / $ 7.00 © 1994 Elsevier Science B.V. All rights reserved IMLET 02116

Discrepancy between direct and antibody-dependent cytotoxic activities of human LAK cells Michael P. Potapnev a'*, Tatyana S. Garbuzenco a, Natalia V. Goncharova a, Vladimir D. Zobnin a, Oleg V. Shadrin b and Svetlana N. Bykovskaya b ~Institute of Hematology and Blood Transfusion, Minsk 223059, Belarus; bCancer Research Centre, Moscow 115478, Russia (Received 20 January 1994; accepted 16 February 1994)

1.

Summary

Human lymphokine-activated killer (LAK) cells display cytotoxic activity against natural killer (NK)-resistant tumor cells in an antibody-independent and -dependent manner. We compared LAK cell-mediated antibody-independent cytotoxicity (LAK activity) and antibody-dependent cellular cytotoxicity (ADCC) against untreated and antibodycoated Raji cells, respectively. Human lymphocytes showed drastically increased LAK activity after stimulation with interleukin-2 (IL-2) for 3 or 7 days when compared to non-activated cells. The level of ADCC was reduced for 3-day-generated LAK cells and augmented for 7-day-generated LAK cells as compared to non-activated cultured lymphocytes. Phenotypical analysis revealed IL-2-induced up-regulation of the proportion of C D I I b + (but not CD16 +) lymphocyte subpopulation in 7-day-generated LAK cells. The data imply that human LAK cells exhibit antibody-dependent and -independent cytotoxic activities via distinct effector pathways at different stages of generation. These stages may be associated with changes in adhesion molecule (CD 1 l b/CD 18) expression on the surface of IL-2-activated lymphocytes.

been shown to exert strong anti-tumor activity in vivo and in vitro [1,2]. Generation of LAK cells in culture of human peripheral blood lymphocytes activated with interleukin 2 (IL-2) for short (2-3 days) or long (7-10 days) term may induce accordingly natural killer (NK)-like or T-cell-like LAK cells as estimated on the basis of phenotypical marker differences [3,4]. Nevertheless, direct cytotoxic activity of LAK cells has been shown not to depend upon their NK/T-cell-related surface markers [5]. LAK cells also exert indirect, antibody-dependent cellular cytotoxicity (ADCC) after induction with IL-2 in vivo and in vitro [6]. It is known that both resting and activated lymphocytes display ADCC and that FcTRbearing NK-like cells are the principal effector lymphocytes [7,8]. We have undertaken a comparative study of the direct cytotoxic (LAK) and indirect (ADCC) activities of short- and long-term generated LAK cells. We found that after IL-2 activation for 3 days or 7 days human lymphocytes exhibit a different profile of LAK and ADCC activities and peculiarities in CD16 and CD1 lb/CD18 surface molecules expression may be associated with these differences.

3. Materials and Methods 2. Introduction

Lymphokine-activated killer (LAK) cells have Key words: Lymphokine-activated killer cell; Phenotyping; Direct cytotoxicity; Antibody-dependent cellular cytotoxicity; (Human) *Corresponding author: Michael P.Potapnev, Institute of Hematology and Blood Transfusion, Dolginovsky tr. 160, 223059, Minsk, Belarus. SSDI 0 1 6 5 - 2 4 7 8 ( 9 4 ) 0 0 0 3 2 - M

3.1. Cells

Mononuclear cells were isolated from fresh heparinized donor blood by a standard Ficoll/Hypaque centrifugation procedure. Purified lymphocytes were obtained from mononuclear cells treated with carbonyl iron (4.5-5.2 /tm, Sigma) to delete phagocytic cells (monocytes). Immunofluorescent analysis of purified lymphocytes indicated < 0.1% contamina-

tion of CD14 + cells. Raji (Epstein Barr virus-positive human B lymphoma) cells were used as NK-resistant targets in the cytotoxicity tests. Cells were cultured in RPMI-1640 medium (Serva) supplemented with 10% heat-inactivated human AB serum or fetal calf serum (FCS), penicillin (100 U/ ml), streptomycin (100 pg/ml) and 2 mM L-glutamine. 3.2. Generation of LAK cells Lymphocytes (2-3 x 106 cells/ml) were cultured in RPMI-1640 medium with 10% AB serum for 3 or 7 days with or without 1000 IU/ml of recombinant human IL-2 (rlL-2, Cetus, sp. act. 3 x 106 Cetus U/mg) as described [9]. When lymphocytes were grown for 7 days the culture medium was changed at the 4th day for fresh complete RPMI-1640 with or without rlL-2, respectively. The rlL-2 activity in IU was determined in CTLL-2 proliferation assay using the 1st International Standard for IL-2 (86/504). After termination of the culture-untreated and IL2-activated (LAK cells) lymphocytes were washed twice and used for evaluation of cytotoxic activity in vitro and surface marker expression. The viability of the washed cultured cells was determined by the conventional trypan dye exclusion test and was not less than 80%. 3.3. Cytotoxieity tests Both direct cytotoxic (LAK) activity and ADCC were performed in microwell round-bottomed plates (ICN Flow) in complete RPMI-1640 medium containing 5% FCS. To estimate L A K activity Raji target cells were labelled overnight with 100 #Ci of Naz[51Cr]O4 as described [5]. Labelled target cells were washed 4 times and incubated with effector lymphocytes (untreated or L A K cells) for 6 h in a total volume of 250 #1. An effector-to-target cell (E/T) ratio of 40:1 or 20:1 was usually used. At the end of incubation plates were centrifuged and 100 /~1 of supernatant was aspirated and counted in y- (Behring) or fl-counters (Beckman). All assays were performed in triplicate. Percentage specific lysis was calculated as follows: Specific lysis ( % ) = 51Cr release f r o m s a m p l e - m e a n s p o n t a n e o u s release × 100, m e a n m a x i m a l release - m e a n s p o n t a n e o u s release

14

51Cr release being measured in cpm. Mean maximal release was that of 51Cr-labelled target cells (104) incubated with 2% SDS. Spontaneous release was 51Cr released from the target cells incubated with complete medium only. Spontaneous release did not exceed 15% of the maximal release. To test for ADCC 51Cr-labelled Raji cells were pre-incubated with hyperimmune heat-inactivated rabbit anti-Raji cell serum (final dilution, 1:500) for 30 min at 37°C. Normal rabbit serum was added into the control wells. Then the effector lymphocytes were added at an E/T ratio of 40:1 or 20:1. After incubation at 37°C for 16 h the microplates were centrifugated and 100 /tl of the supernatant was aspirated and counted in 7- (Behring) or /~-counters (Beckman). All experiments were performed in triplicate. The percentage of specific lysis was calculated as the difference between the amount of 51Cr released by target cells in the presence of effector cells + antiRaji serum and in the presence of effector cells + normal rabbit serum [8]. Total and spontaneous 5aCr release in ADCC assay were estimated as described earlier for L A K activity. All assays were performed in triplicate. The standard deviation was < 15% of the mean value. 3.4. Cell surface marker analysis Several function-associated L A K cell markers were examined in the indirect immunofluorescence [10]. L N K 16 (CD 16) monoclonal antibodies (mAbs) specific for FcTR111 was provided by Dr. A.V. Filatov (Institute of Immunology, Moscow). ICO-GM1 (CD1 lb) and ICO-90 (CD3) mAbs were provided by Dr. A.Ju. Baryshnikov (Cancer Research Center, Moscow). FITC-conjugated F(ab')2 fragments of rabbit IgG anti-mouse IgG (secondary Abs) were obtained from Medbioservice (Moscow). FITC-conjugated Leu M3 (CD14) mAbs (Becton Dickinson) were used to indicate the monocyte contamination of lymphocyte population. Freshly isolated or cultured lymphocytes (including L A K cells) were washed twice in phosphate-buffered saline (PBS) containing bovine serum albumin and sodium azide. Cell viability usually was 80-95%. The cell pellet was mixed with appropriately diluted mAbs, resuspended and incubated for 30 min at 4°C. The cells were washed twice and incubated with the FITC-conjugated secondary Abs. At the end of incubation cells were washed and fixed in 1% paraformaldehyde (Serva). Fixed cells were analyzed by flow cytometry on FACScan (Becton Dickinson).

4. Results and Discussion

As the broad range of lymphoid and non-lymphoid blood cells display cytotoxic activity (especially ADCC) we used only purified lymphocytes in our experiments. Purification procedure included carbonyl iron treatment of peripheral blood mononuclear cells to eliminate monocytes and residual polymorphonuclear cells and platelets. The preliminary investigation showed that there were no monocytes in the lymphocyte population as assessed by surface phenotyping and cytochemical tests. Polymorphonuclear cells and platelets were absent when estimated by light microscopy. Most purified lymphocytes (72.2 + 3.7%) expressed surface CD3 antigen (Table 1). The amount of NK cells (CD16 ÷ cells) was 30 + 5.5% in the initial lymphocyte population (Table 1, Day 0). We used CD16 ÷ marker of NK cells whereas it is known that a small subset of CD3 ÷ CD16 + lymphocytes exist [2,11]. C D l l b marker represents the family of adhesion molecules (f12 integfins) on NK cells and subset of CD3 + CD8 ÷ T lymphocytes [10,12-14]. There were 42 + 3.8% of CD11 b ÷ cells in the initial lymphocyte population. During cultivation of the lymphocytes for 3 or 7 days without IL-2 we observed a reduction of CD1 lb and CD16 surface marker expression as was described previously for NK cells [9,13]. The percentage of T lymphocytes (CD3 ÷ cells) was more constant throughout the culture period. A slight increase of CD3 ÷ cells (P>0.1) in unstimulated lymphocytes was shown by day 7 of culture especially when compared with the 3-day culturing period. Incubation with IL-2 for 3 days facilitated the tendency for decreasing the proportion of activated lymphocytes expressing CD1 lb or CD16 surface markers. It was described that non-NK cell populations such as CD8 ÷ C D l l b + cytotoxic subset of T lymphocytes exhibited the same tendency after stimulation with TABLE 1 SURFACE MARKER

EXPRESSION OF CULTURED

IL-2 [12]. After activation of purified NK cells with IL-2, Robertson et al. [13] and Melder et al. [14] observed an unchangeable level of cells that express CD1 lb marker but drastically increased the number of cells expressing CD1 la, CD2, CD54, CD58, CD18 surface adhesion molecules. It is also known that after activation both NK and T cells down-modulate their surface CD16 and adhesion molecules [9,10,12,15]. In our experiments prolonged activation of lymphocytes with IL-2 for 7 days resulted in the further decrease of cell populations that express CD16 or CD1 lb markers when compared with the 3-day period. After 7 days of culture the level of CD16 + cells was identical in populations of untreated and IL-2activated lymphocytes. However, we were surprised to find that the level of C D l l b ÷ cells was 2-fold higher in 7-day-generated LAK cells as compared to parallel untreated lymphocytes (Table 1). We did not identify the lymphocyte subset that expresses CDI lb surface marker but it is known that at day 7 most proliferating LAK cells are T cells (with CD8 ÷ phenotype) although NK-like LAK cells are also present [4,13]. The recently obtained data that the CR3 (CDllb/CD18) molecule can be associated with FcTRIII (CD16) in neutrophil membranes [16] does not exclude the possibility of different regulation pathways of their expression on leucocyte surface. Functional characterization of 3- and 7-day-generated LAK cells included evaluation of their cytotoxicity against NK-resistant Raji cells (LAK activity) and antibody-coated Raji cells (ADCC activity). As shown in Table 2, the lymphocytes cultured for 3 days in the absence of rlL-2 failed to exhibit direct cytotoxicity against Raji cells in most experiments. The level of ADCC activity mediated by resting lymphocytes was high in all experiments. ADCC activity was observed at both 40:1 and 20:1 E/T ratios (data not shown). CD16 ÷ cells are known to be the princi-

IL-2-ACTIVATED HUMAN

LYMPHOCYTES

The d a t a represent the m e a n o f 4 separate experiments. Marker

Positive cells ( % ) Day 0

CD3 CD16 CDllb

72.2 _+ 3.7 30 _+ 5.5 42 + 3.8

3-day-cultured Ly*

7-day-cultured Ly

Untreated

IL-2-act.

Untreated

IL-2-act.

67.3 ___ 3.8 24.2 ___ 2.1 29.5 ___ 4.9

69.1 ___ 4.2 17.9 ___ 3.1 22.9 ___ 4.8

79.3 + 3.5 13.7 _ 5.0 7.4 ___ 1.3

72.1 _+ 4.8 13.4 ___ 4.5 15.4 + 4.5

*Ly, lymphocytes.

15

TABLE 2 COMPARATIVE LEVEL OF LAK AND ADCC ACTIVITY MEDIATED BY UNSTIMULATED OR IL-2-ACTIVATED HUMAN LYMPHOCYTES, CULTURED FOR 3 DAYS The ability of human iymphocytes cultured for 3 days in the presence or absence of recombinant human IL-2 (1000 IU/ml) to kill Raji target cells (LAK activity) or antibody-coated Raji target cells (ADCC activity) was assayed by 51Cr release as described in Materials and Methods. Mean of triplicate cultures is shown. Standard deviation did not exceed 15%. Exp. no.

E/T ratio

Tumor target cell lysis (%) LAK activity of effector cells

ADCC activity of effector cells

unstim.

IL-2-act.

unstim.

IL-2-act.

1

40:1 20:1

2.2 -2.9

28.4 11.7

57.5 101.8

40.1 18.9

2

40:1 20:1

5.3 1.2

33.2 26.9

36.3 12.8

7.5 --25.9

3

40:1 20:1

23.4 24.2

43.7 37.4

29.9 30.4

3.5 7.2

pal ADCC effector lymphocyte population [2,8], although cooperation of several lymphocyte subsets is desirable for achieving greater ADCC activity [17]. Three-day-generated LAK cells (unlike unstimulated lymphocytes) displayed an increased direct cytotoxic activity against Raji target cells in all experiments. It was of interest that ADCC activity of these LAK cells was inhibited when compared to the unstimulated lymphocytes. This discrepancy was associated with simultaneous decline in proportion of lymphocytes expressing surface FcyRIII that initiates ADCC and adhesion molecules that mediate E/T cell binding as described in our phenotypical investigations and by other authors [10,13]. Our data support the finding of Abrams and Brahmi [7] that antibody-independent cytolytic activity of NK cells and effectors of ADCC are not cross-linking in absorption experiments. It is also known that separated lymphocyte subsets display different pathways of cytotoxicity [19,20]. Thus we assumed that the mechanisms of LAK and ADCC activities mediated by LAK cells are different. Additionally it is known that antibody Fcy binding may cause a negative regulation of cytolytic machinery in ADCC [18]. The LAK/ADCC activity correlation of LAK cells is changing after 7 days of IL-2 activation. As shown in Table 3, 7-day-generated LAK cells exhibit the increased level of both LAK and ADCC activities as compared to unstimulated lymphocytes. It depends primarily on the drastically decreased ADCC activity exhibited by 7-day-cultured unstimulated lymphoid cells. In most experiments LAK activity of 716

day-generated LAK cells was slightly lower than that of 3-day-generated LAK cells. ADCC activity of 7day-generated LAK cells was always higher than that of 3-day-generated LAK cells. We assume that up-regulation of ADCC activity of 7-day-generated LAK cells is associated with the concomitant increase in the expression of surface adhesion molecules (not particular CD1 l b) during the long-term generation of LAK cells [21,22]. It is also of interest that the level of cytotoxic cytokines (lymphotoxin, tumor necrosis factor
TABLE 3 COMPARATIVE LEVEL OF LAK AND ADCC ACTIVITY MEDIATED BY UNSTIMULATED OR IL-2-ACTIVATED HUMAN LYMPHOCYTES, CULTURED FOR 7 DAYS See Table 2 for footnotes. Exp. no.

E/T ratio

Tumor target cell lysis (%) LAK activity of effector cells

ADCC activity of effector cells

unstim.

IL-2-act.

unstim.

IL-2-act.

1

40:1 20:1

0.7 -0.6

18.8 25.5

13.2 5.5

36.6 35.8

2

40:1 20:1

2.4 23.4

16.9 25.1

7.3 7.3

10.7 o. 1

3

40:1 20:1

18.3 13.0

26.7 26.1

30.0 30.1

41.7 32.6

cells in a n a n t i b o d y - d e p e n d e n t o r - i n d e p e n d e n t m a n ner. A D C C m a y p l a y a n essential r o l e in t u m o r cell lysis b y l o n g - t e r m g e n e r a t e d - L A K cells.

References [1] Rosenberg, S.A. (1988) Immunol. Today 9, 58. [2] Lindermann, A., Herrmann, F., Oster, W. and Mertelsmann, R. (1989) Blut 59, 375. [3] Ortaldo, J.R., Masson, A. and Overton, R. (1986) J. Exp. Med. 164, 1193. [4] Roussel, E., Gerrard, J.M. and Greenberg, A.H. (1990) Clin. Exp. Immunol. 82, 416. [5] Chong, A.S.-F., Aleksijevic, A., Scuderi, P. and Grimes, W.J. (1989) Cancer Immunol. Immunopathol. 29, 270. [6] Eisenthal, A. and Rosenberg, S.A. (1989) Cancer Res. 49, 6953. [7] Abrams, S.I. and Brahmi, Z. (1988) J. Immunol. 140, 2090. [8] Shen, L., Graziano, R.F. and Fanger, M.W. (1989) Mol. Immunol. 26, 959. [9] Roberts, K., Lotze, M.T. and Rosenberg, S.A. (1987) Cancer Res. 47, 4366. [10] Lebow, L.T. and Bonavida, B. (1990) Proc. Natl. Acad. Sci. USA 87, 6063. [11] Loughran T.P. (1993) Blood 82, 1. [12] Dianzani, U., Zarcone, D., Pistoia, V., Grossi, C.E., Pileri, A., Massaia, M. and Ferrarini M. (1989) Eur. J. Immunol. 19, 1037. [13] Robertson, M., Caligiuri, M.A., Manley, T.J., Levine, H. and Ritz, J. (1990) J. Immunol. 145, 3194.

[14] Melder, R.J., Walker, E., Herberman, R.B. antl Whiteside, T.L. (1991) Cell. Immunol. 132, 177. [15] Jaaskelainen, J., Maenpaa, A., Patarroyo, M., Gahmberg, C.G., Somersalo, K., Tarkkanen, J., Kallio, M. and Timonen, T. (1992) J. Immunol. 149, 260. [16] Seghal, G., Zhang, K., Todd, R.F., Boxer, L.A. and Petty, H.R. (1993) J. Immunol. 150, 4571. [17] Richter, M. and Kazaniwsky, N. (1987) Cell. Immunol. 108, 297. [18] Fanger, M.W., Segal, M.D. and Romet-Lemonne J.-L. (1991) Immunol. Today 12, 51. [19] Smyth, M.J., Norihisa, Y. and Ortaldo J.R. (1992) J. Immunol. 148, 55. [20] Umehara, H., Minami, Y., Domae, N. and Bloom, E.T. (1993) Lymphokine Cytokine Res. 12, 347. [21] Mikamo, S. (1992) Hum. Cell 5, 256. [22] Triozzi, P.L., Eicher, D.M., Smooth, J. and Rinchart, J.J. (1992) Exp. Hematol. 20, 1072. [23] Limb, G.A., Meager, A., Wooley, J., Wadhwa, M., Biggerstaff, J., Brown, K.A. and Wolstencroft (1989) Immunology 68, 514. [24] Abe, Y., van Eden, M., Gatanaga, M., Wang, F.J., Brightbill, H.D., Granger, G.A. and Gatanaga, T. (1993) Lymphokine Cytokine Res. 12, 279. [25] Abe, Y., Masazumi, M., Horiuchi, A., Kimura, S. and Hitsumoto, Y. (1991) Jpn J. Cancer Res. 82, 23. [26] Duke, R.C., Sellins, K.S. and Cohen, J.J. (1988) J. Immunol. 141, 2191. [27] Young, L.H.Y. (1990) Annu. Rev. Med. 41, 45. [28] Graziano, R.F., Erbe, D.V. and Fanger, M.W. (1989) J. Immunol. 143, 3894. [29] Smyth, M.J. and Ortaldo, J.R. (1993) J. Immunol. 151,740.

17