In VitroCell.Dcv.Biol.28A:782-786,November-December1992 © 1992TissueCultureAssociation 0883-8364/'92 $01.50+0.00
POLYMORPHONUCLEAR LEUKOCYTES-MEDIATED LYSIS OF A 4 3 1 CELLS INDUCED BY IgG1 MOUSE ANTI-EPIDERMAL GROWTH FACTOR RECEPTOR MONOCLONAL ANTIBODIES TOMOYUKI KAWAMOTO, KOUJI K1SHIMOTO, KOJIRO TAKAHASHI, TOMOHIRO MATSUMURA, J. DENRY SATO, AND SHIGEHIKO TANIGUCHI
Department of Biochemistry (T. K., K. T., S. T.) and Department of Maxillofacial Surgery H (K. K., T. M.), Okayama UniversityDental School, Shikata 2-5-1, Okayama City 700, Japan; and IV. Alton Jones Cell Science Center, Inc. (J. D. S.), 10 Old Barn Road, Lake Placid, New York 12946 (Accepted 27 July 1992)
SUMMARY
The abilityof different Fc receptors (Fc~ R) on lgG-mediated eytotoxicityfor humanepidermoidcarcinomaA431 cells bearing large number of epidermal growth factor receptors (EGFRs) was examined by using two isotypes (IgG1 and IgG2a) of murine monoclonal antibodies (MoAbs) against EGFR in the presence of human monocytes and granulocytes. Two MoAbs (225 and LA1) of the IgG1 isotype exhibited effective cytolytic activity for A431 cells with human polymorphonuclear leukocytes (PMN) rather than with human mononuclear cells (MNC). In contrast, two MoAbs (528 and 579) of the IgG2a-isotype lysed the cells less effectively with PMN than with MNC. Anti-Fc'y R II (CDw32) MoAb 2 E l inhibited the IgGl-mediated cytotoxicity by PMN, and anti-Fc'y R III (CD16) MoAb 80H3 did not inhibit the IgG2a-mediated cytotoxicity by MNC. Under these conditions, antibody-dependent cell-mediated cytotoxicity by mouse MoAb IgG1 isotypes resulted from antibody binding to the Fc'y R II (CDw32) of PMN.
Key words: EGF receptor; monoclonal antibody; polymorphonuclear leukocyte; mononuclear leukocyte; cytotoxicity.
INTRODUCTION
MATERIALSANDMETHODS
As the proliferation of cancer cell lines is regulated by hormones and growth factors in vitro, and it has been suggested that a similar regulatory mechanism is operating in vivo (3), there have been several attempts to use monoclonal antibodies (MoAbs) against hormone or growth factor receptors for analysis of the regulatory mechanism and for immunotherapy of malignant tumors (1,21,23,25,30). Trowbridge and Domingo (30) reported that treatment with anti-transferrin receptor MoAb inhibited tumor formation by a human melanoma cell line in athymic mice. Masui et al. (21) reported that anti-epidermal growth factor receptor (EGFR) MoAbs inhibited tumor formation of A431 cell xenografts in athymic mice and that MoAb 528-IgG2a induced A431 cell lysis with activated mouse macrophages or rabbit complement in vitro (22). Rodeck et al. (25) also showed that anti-EGFR MoAb 425-IgG2a directed A431 cell lysis by human monocytes, lymphocytes and mouse macrophages. It has been proposed that the IgG2a isotype MoAbs were most effective in monocyte- or macrophage-mediated ADCC (antibody-dependent cell-mediated cytotoxicity) among MoAbs against tumor-associated antigens (including receptors) (6,10,13,18). However, little is known about polymorphonuclear leukocyte (PMN)-mediated ADCC, though PMN has Fc receptors for IgG (Fc'y R) and is able to promote cytotoxicity of target tumor cells (16,17). In this report, we propose a mechanism of PMN-mediated ADCC employing human PMN and mouse anti-EGFR MoAb isotype IgG1.
Target cells. Human epidermoid carcinoma A431 ceils, which express an unusually large number of EGFR (1-3 X 106/cell), were used as target tumor cells in the present study. The A431 cells were maintained in a mixture (DME/F12-5% FBS) of Dulbecco's modified Eagle's and Ham's F-12 medium (Nissui Seiyaku Co. Ltd., Japan) containing 15 mM H EPES and heat-inactivated 5% fetal bovine serum (FBS: GIBCO Laboratories, USA) at 37 ° C in a 5% CO2 and 95% air atmosphere. The A431 cells were incubated with 0.1% trypsin and 1 mM EDTA at 37 ° C for 5 rain, harvested, washed with DME/F12-5% FBS medium, and then used for cytolysis assay. Effector cells. Human PMN and mononuclear cells (MNC) used as effector cells were separated from heparinized healthy human peripheral blood with Mono-Poly Resolving Medium (M-PRM: Flow Laboratories Ltd., Australia) as described previously (4). Four ml of human whole blood were layered onto 3 ml of M-PRM in a sterilized 17 × 120 mm test tube (Falcon 2095: Becton Dickinson Overseas Inc., USA). The tube was centrifuged at 300 Xg for 30 min at room temperature, and the top band of MNC fraction and second band of PMN fraction were transferred into individual tubes. The MNC and PMN, histologically distinguishable each other by Diff Quick Dyeing (Kokusai-Shiyaku, Kobe), were thereby quantitated and the contamination thus estimated was below 5% in each sample. After washing with DME/F12 medium, the MNC and PMN cells were resuspended to the required concentration in DME/F12-5% FBS medium and used as the effector cells. Viability of the effector cells in this method was estimated to be over 99% by means of the trypan blue dye exclusion test. The top leukocyte band contained over 94% MNC (lymphocytes and monocytes), whereas the second band contained over 96% PMN. Monoclonal antibodies. Four mouse MoAbs (528- and 579-MoAbs as IgG2a isotype and 225- and LA1-MoAbs as IgG1 isotype) raised against EGFR of A431 ceils were used for ADCC assay. The development and characterization of these MoAbs have been reported (12,26,27). LAI has
782
783
POLYMORPHONUCLEAR LEUKOCYTE-MEDIATED CELL LYSIS 0r,II
; "oF" .... (A)
-]
)
"
, ,
,
*
lO
0 0.01 0.1 1 10
25
50
~
)
•
I
40
,
i
m
(B)
i,o 100
0 0.1 1 10 [ LAI-lgG1
[ 225-1gG1 ] ( nM )
25 50 1( nM )
Fie. 1. Cytolytic activity for A431 cells dependent on mouse anti-EGFR MoAb IgG1 isotype (225 and LA1). Experimental procedures were described in Materials and Methods. Results are expressedas mean -+ SD of triplicate data. PMN or MNC (8 X 105 cells/well) as cffector cells and [aH]uridine-labeled A431 (5 × 103 cells/welt) as target cells were incubated with increasing concentrations of (A) 225-IGG1 or (B) LAI-lgG1 for 24 b. Effector cells used were PMN (Q), MNC (A) and without cells (V1). Ratio (E/T) on cell number of the effector to the target was 160.
also been characterized recently as IgG1 isotype (Sato, J. D. et al., unpublished result). These four MoAbs bind to the extracellular peptide domain of EGFR and compete with EGF for binding to EGFRs. The Fab fragments of 225- and 528-MoAbs were prepared with lmmunopure Fab Preparation Kit (Pierce, USA) and were shown to be free of intact IgG and Fc fragments
100 --/~ . . . .
n
,
by analysis of 10% SDS-PAGE under nonreducing conditions. MoAb ARB229 (lgG1 isotype), developed against a haptene (tyrosyl arsenate) (Hybritech, Inc., USA), was used as control. Mouse MoAb 2 E l developed against human Fc7 R II (CDw32) (7,11,19,29) and MoAb 80H3 specific to the CD16 Fc'y RIII (20) were obtained from Immunotech (France).
,
100 -I~ ,
(A) o~
80
"-i
60
,
, #
,
,
(B)
A
80
40
0
0
2O
O:
0 0.01 0.1 1 10
25 50
[ 528-1gG2a ] ( nM )
100
0 0.1 1 10
25 5O
[ 579-1gG2a ] ( nM )
Fie. 2. Cytolytic activity for A431 cells dependent on mouse anti-EGFR MoAb IgG2a isotype (528 and 579). (A) and (B) for 528and 579-lgG2a isotypes, respectively. Experimental procedures were the same as in Fig. 1. Effector cells used were PMN (O), MNC (L~) and without cells (E3). Ratio of E / T was 160.
784
KAWAMOTO ET AL.
50["
1
I
!
I
|
A
o~
40
.>_
30
20
0
10
0 10 20 40
5
80
120
160
E/T ratio Ftc. 3. Influence of effector cell number of 225-lgGl-dependent cytolyric activity. [sH]Uridine-labeled A431 (5 × 103 celts/well) was incubated with increasing numbers of effector cells in the presence or absence of 17 nM 225-1gG1 for 24 h. PMN with 225-IGG1 (e), PMN without 225-IGG1 (©), MNC with 225-IGG1 (A) and MNC without 225-IGG1 (A). Results are expressed as mean + SD of triplicate data.
against A431 target cells in the presence of MNC as mentioned by Masui et al. (22), but showed remarkable cytolytic activity in the presence of PMN (Fig. 1 A). LAI-IgG1 also showed a similar cytolytic activity with PMN as shown also in Fig. 1 B. 528- and 579IgG2a isotypes, on the other hand, had little cytolytic activity with PMN (Fig. 2), even though they showed significant cytolytic activity in the presence of MNC (Fig. 2) as reported previously (6,13,18,22). Clear ADCC activity of 225-IGG1 was evident at antibody concentrations above 10 nM (Fig. 1). At 17 nM of 225IgG1, the cytolytic activity increased in accordance with the increased ratio of effector cell (PMN)/target cell (A431) (Fig. 3). A ratio of PMN/A431 in excess of 160 was not practical for the ADCC assay, owing to the difficulty of handling higher numbers of PMN in the assay. Fab fragments of both 225-IGG1 and 528-lgG2a induced no significant A431 cell lysis with PMN or MNC. Mouse MoAb ARB229 lgG1, which was incapable of binding to EGFR augmented ADCC activities with neither PMN nor MNC (data not shown). These results demonstrated that 225- and LAI-IgG1 mediated A431 cell lysis by PMN through binding to the EGFR of the target cells, whereas 528- and 579-IgG2a mediated A431 cell lysis by MNC more effectively than by PMN.
Effect of anti-Fc'y R MoAbs on 225-IgG l-dependent cytolytic activity by PMN. Human PMN constitutively expresses two types of Fc receptors for lgG (Fc'y R): Fc'y R II (CDw32) (15) and CD16 Fc'y R III (8). Mouse MoAb 2 E l capable of binding specifically to Fc'y R II (CDw32) (7,11,19,29) inhibited the 225-IGG1 mediated cytolytic activity by PMN, but MoAb 80H3 specific to CD16 Fc~y R Ill (20) did not inhibit the ADCC activity by PMN (Fig. 4). Another anti-EGFR MoAb LAI-IgGI gave similar results as 225-IGG1 (data not shown). These results indicated that IgGl-de-
A
40
Cytolysis assay. Tumor cell cytolysis can be observed with the [SH]uridine release assay (34). A431 cells (5)< 10s) were placed into each of 96-well flat-bottomed microplates (Falcon 3072) containing 0.2 ml of DME/F12-5% FBS medium. After preincubation for 6 to 8 h, 0.2 ~tCiof [SH]uridine (2-10 Ci per mmol: NEN Research Products, USA) was added to each well. The A431 cells were incubated overnight at 37 ° C, washed three times with DME/F12-5% FBS medium, then both effector cells and anti-EGFR MoAbs were added in a total volume of 0.2 ml DME/F12-5% FBS medium per well. After incubation for 24 h, sample solutions from the microplates were centrifuged at 200 ×g for 5 min. Using 0.1 ml aliquot of each supernatant, the experimental release (ER) of radioactivity from target cells was measured with a liquid scintillation counter. The cytolytic activity [CA(%)] was calculated by the following equation:
20
CA(%) = 100 X {ER(cpm) - SR(cpm)}/{MR(cpm) - SR(cpm)} where SR = spontaneous release of radioactivity from the target cells cultured without the effector cells and MoAbs, and MR = maximum release of radioactivity from target cells after treatment with 1% Triton X-100. The obtained CA(%) values are expressed as the average of triplicate data with the standard deviation. ADCC inhibition assay by anti-FcT R MoAb. [sH]Uridine-labeled A431 cells (5 X 103/well) were incubated with the effector cells (8 X 10S/well) and varied concentrations of anti-Fc7 R MoAb (2El or 80H3) at 37 o C for 30 min. 17 nM of anti-EGFR MoAbs was then added, incubated for 24 h, and the CA(%) value of each well was estimated as described above. RESULTS
Anti-EGFR MoAb lgG1 isotype-dependent cell-mediated cytotoxicity. 225-IGG1 isotype showed no significant cytolytic activity
O 0
~ 0
1
10
100
[ Anti-FcyR MoAb ] ( nM ) Fic. 4. Effects of anti-FcT R II MoAb 2El and anti-Fc'y R Ill MoAb 80H3 on 225-1gGl-dependent cytolytic activity mediated by PMN. [SH]Uridine-labeled A431 (5 X 103 cells/well) and PMN (8 X 105 cells/ well) cells were incubated with increasing concentrations of anti-Fc'y R MoAbs for 30 min. 225-IGG1 (17 riM) was then added and incubation was continued for 24 h. Results are expressed as mean + SD of triplicate data with MoAb 2El (e) and MoAb 80H3 (C)).
POLYMORPHONUCLEAR LEUKOCYTE-MEDIATED CELL LYSIS TABLE 1 ADCC EFFECTS OF ISOTYPE IgG AND THE INHIBITION BY ANTI-Fcs' R MoAbs IN THE PRESENCE OF MNC OR PMN~ ADCC(%)
IgG1 isotype: ARB229 225 LA1 IgG2a isotype: 528 579 Anti-Fc'y RII MAb 2El: Anti-Fc'y RIII MAb 80H3:
225 528 225 528
MNC
PMN
0 6 13 52 72
0 43 45 9.5 2
-3.5 (-9) -2.5 -11
-23 -6.5 -11 -3.5
a Ratio of E/T: 160. The concentrations of IgGs and anti-FcR Ab were both 100 nM.
pendent cytolytic activity of PMN was mediated by binding to the Fc"/R II (CDw32) of PMN. DISCUSSION Since Steplewski et at. (28) proposed that macrophages are effector cells for human tumor destruction directed by MoAbs against tumt)r cells, mouse MoAb of the IgG2a isotype was found to be most effective in macrophage- or lymphocyte-mediated ADCC (6,10,13,18,22,25). Although 225-IGG1 completely inhibited the growth of A431 cell xenografts in athymic mice (21), the role of mouse MoAb IgG1 isotype in ADCC remained unclear because 225-1gG1 did not induce A431 cell lysis by activated mouse macrophages (22). As summarized in Table 1, we showed that mouse anti-EGFR MoAbs of the IgG1 isotype (225 and LA1) could lyse A431 cells in the presence of human PMN as effectively as IgG2a antibodies (528 and 579) do in the presence of human MNC. The inhibition of IgG1 isotype-dependent cytolysis by MoAb 2 E l specific for CDw32 Fc"/R II of human PMN indicated that 225- and LAI-IgG1 isotype MoAbs directed cytolytic activity by binding to the CDw32 Fc7 R II of PMN. This result is consistent with the findings that murine IgG1/2b antibodies bind to Fc3' R I1 (2,31), and that Fc'y R II of PMN could be the cytotoxic trigger molecule for tumor cells (9,14). In the present study, we demonstrated a cytolytic activity mediated by human PMN in an in vitro system. ADCC, when examined in vivo, may prove more complex. Lopez et al. (•6) reported that mouse granulocytes were cytotoxic to target cells coated with mouse IgG1, IgG2a or IgG2b antibodies, and that MoAb 4.2 G2 reactive with mouse Fc'y R blocked the binding of these antibody-coated cells to the granulocytes. This contrasts with our finding that mouse lgG1 mediated A431 cell lysis by human PMN more effectively than in the case of IgG2a. In our study, however, human PMN instead of mouse PMN was used and Fc'y R I capable of binding to mouse IgG2a was shown to be undetectable on fresh human PMN. Recently, it was found that interferon-"/induced Fc'y R I on human PMN (24). In previous studies, we ascertained that 528-IgG2a also effectively mediated the lysis of A431 cells by human PMN in the presence of interferon-'y (data not shown), Human monocytes have two types of FcR for IgG, Fc'y R I and Fc'y R II (32). Boot et al. (5)
785
reported that human monocytes could be triggered to ADCC by routine MoAb IgG1 which was predominantly mediated through Fc3' R II of the monocytes. In our study, however, the 225- or LAI-IgG1 isotype was less effective than the case of 528-IgG2a isotype to the ADCC by MNC (Fig. 1). It remains to be clarified whether the Fc'y R II s of human PMN and MNC are essentially identical. Modification of the conformation of these Fc'y Rs could contribute to differences between PMN and MNC; Van de Winkel et al. (33) reported that the interaction of low responder monocytes with mouse IgG 1 was dramatically increased by protease treatment. Comprehensive and well rounded studies of ADCC are urgently required. REFERENCES 1. Abound-Pirak, E.; Hurwitz, E.; Pirak, M. E., et al. Effcacy of antibodies to epidermal growth factor receptor against KB carcinoma in vitro and in nude mice. J. Natl. Cancer Inst. 80:1605-1611; 1988. 2. Anderson, C. L.; Ryan, D. H.; Loony, R. J., et al. Structural polymorphism of the human monocyte 40 kilodahon Fc receptor for IgG. J. Immuuol. 138:2254-2256; 1987. 3. Barnes, D.; Sato, G. Methods for growth ofcuhured cells in serum-free medium. Anal. Biochem. 102:255-270; 1980. 4. Bignald, L. P.; Ferrante, A. Mechanismof separation of polymorphonuclear leukocytes from whole blood by the one-step Hypaque-Ficoll method. J. Immunol. Methods 96:29-33; 1987. 5. Boot, J. H. A.; Geerts, M. E. J.; Aarden, L. A. Functional polymorphisms of Fc receptors in human monocyte-mediated cytotoxicity towards erythrocytes induced by murine isotype switch variants. J. Immunol. 142:1217-1223; 1989. 6. Denkers, E. Y.; Badger, C. C.; Ledbetter, J. A., et al. Influence of antibody isotype on passive serotherapy of lymphoma. J. Immunol. 135:2183-2185; 1985. 7. Farace, F.; Mitajavila, M. T.; Betaieb, A., et al. New hematopoietic differentiation antigens detected by anti-K562 monoclonal antibodies. Cancer Res. 48:5758-5765; 1983. 8. Fleit, H. B.; Wright, S. D.; Unkeless, J. C. Human neutrophil Fc'y receptor distribution and structure. Proc. Nat. Acad. Sci. USA 79:3275-3279; 1982. 9. Graziano, R. F.; Fanger, M. W. Fc'y RI and Fc'y RII on monocytes and granulocytes are cytotoxic trigger molecules for tumor cells. J. Immunol. 139:3536-3541; 1987. 10. Heike, Y.; Hamada, H.; Inamura, N., et al. Monoclonal anti-P-glyeoprotein antibody-dependent killing of multidrug-resistant tumor cells by human mononuelear cells. Jap. J. Cancer Res. 81:11551161; 1990. 11. Hogg, N.; Horton, M. A. Myeloidantigens: new and previously defined clusters. In: McMichael, A. J., et al., eds. Leucocyte typing. IIl. White cell differentiation antigens. Oxford, England: Oxford Univ. Press; 1987:576-602. 12. Kawamoto, T.; Sato, J. D.; Le, A., et al. Growth stimulation of A431 cells by epidermal growth factor: identification of high-affinity receptors for epidermal growth factor by an anti-receptor monoclonal antibody. Proc. Natl. Acad. Sei. USA 80:1337-1341; 1983. 13. Kipps, T. J.; Paraham, P.; Punt, J., et al. Importance of immunoglobulin isotype in human antibody-dependent, cell-mediated eytotoxicity directed by murine monoclonal antibodies. J. Exp. Med. 161:1-17; 1985. 14. Lanier, L. L.; Ruitenberg, J. J.; Phihps, J. H. Funetional and bioehemical analysis of CD16 antigen on natural killer cells and granulocytes. J. ImmunoL 141:3478-3485; 1988. 15. Loony, R. J.; Ryan, D. H.; Takahashi, K., et at. Identification of a second class of lgG Fc receptors on human neutrophils: a 40 kilodalton molecule also found on eosinophils. 1. Exp. Med. 163:826836; 1986. 16. Lopez, A. F.; Strath, M.; Sanderson, C. J. Mouse immunoglobulin isotypes mediating cytotoxicity of target cells by eosinophils and neutrophils. Immunology 48:503-509; 1983.
786
KAWAMOTO ET AL.
17. Lopez, A. F.; Battye, F. L.; Vadas, M. A. Fc receptors on mouse neutrophils and eosinophils: antigenic characterization, isotype specificity and relative cell number density measured by flow cytometry. Immunology 55:125-133; 1985. 18. Lubeek, M. D.; Steplewski, Z.; Saglia, F., et al. The interaction of murine IgG subclass proteins with human monocyte Fc receptor. J. Immunol. 135:1299-1304; 1985. 19. Macintyre, E. A.; Roberts, P. J.; Abdul-Gaffar, R., et al. Mechanism of human monocyte activation via the 40-kDa Fc receptor for IgG. J. lmmunol. 141:4333-4343; 1988. 20. Mannoni, P.; Janowska-Wieczorek, A.; Turner, A. R., et al. Monoclohal antibodies against human granulocytes and myeloid differentiation antigens. Human lmmunol. 5:309-323; 1982. 21. Masui, H.; Kawamoto, T.; Sato, J. D., et al. Growth inhibition of human tumor cells in athymic mice by anti-epidermal growth factor receptor monoclonal antibodies. Cancer Res. 44:1002-1007; 1984. 22. Masui, H.; Moroyama, T.; Mendelsohn, J. Mechanism of anti-tumor activity in mice for anti-epidermal growth factor receptor monoclonal antibodies with different isotype. Cancer Res. 46:5592-5598; 1986. 23. Masui, H.; Kamrath, H.; Apell, G., et al. Cytotoxicity against human tumor cells mediated by the conjugate of anti-epidermal growth factor receptor monoclonal antibody to recombination ficin A chain. Cancer Res. 49:3482-3488; 1989. 24. Perussia, B.; Dayton, E. T.; Lagarus, R., et al. Immune interferon induces the receptor for monomeric IgGl on human monocytic and morpeloid cells. J. Exp. Med. 158:1092-1113; 1983. 25. Rodeck, U.; Herlyn, M.; Herlyn, D., et al. Tumor growth modulation by a monoclonal antibody to the epidermal growth factor receptor: immunologically mediated and effector cell-independent effects. Cancer Res. 47:3692-3696; 1987.
26. Sato, J. D.; Kawamoto, T.; Le, A., et ak Biological effects in vitro of monoclonal antibodies to human epidermal growth factor receptors. Mol. Biol. Med. 1:511-529; 1983. 27. Sato, J. D.; Le, A. D.; Kawamoto, T. Derivation and assay of biological effects of monoclonal antibodies to epidermal growth factor receptors. Methods Enzymol. 146:63-81; 1987. 28. Steplewski, Z.; Herlyn, D.; Maul, G., et al. Hypothesis: macrophages as effector cells for human tumor destruction mediated by monoclonal antibody. Hybridoma 2:1-5; 1983. 29. Tenero, P. A. T.; Van der Shoot, E.; Visser, F. J., et al. Three different types of Fc'y receptors on human leucocytes defined by workshop antibodies: Fc~y Rl0~ on neutrophils, Fc'y w~ of K/NK lymphocytes and Fc'y R 11. In: McMichael, A. J., et al., eds. Leucocyte typing. 11I. White cell differentiation antigens. Oxford, England: Oxford Univ. Press; 1987:702-706. 30. Trowbridge, I. S.; Domingo, D. L. Anti-transferrin receptor monoclonal antibody and toxin-antibody conjugates affect growth of human tumor cells. Nature 294:171-173; 1981. 31. Van de Winkel, J. G. J.; Van Duijnhoven, H. L. P.; Van Ommen, R., et al. Selective modulation of two human monocyte Fc receptors for IgG by immobilized immune complexes. J. lmmunoi. 140:35153521; 1988. 32. Van de Winkel, J. G. J.; Boonen, G. J.; Janssen, P. L., et al. Activity of Fc receptors, Fc gamma R I and Fc gamma R II, in human monocyte cytotoxicity to sensitized erythrocytes. Scand. J. Immuno|. 29:23-31; I989. 33. Van de Winkel, J. G. J.; Van Ommen, R.; Huizinga, J. W. J., et al. Proteolysis induces increased binding affinity of the monocyte type II FcR for human IgG. J. Immunol. 143:571-578; 1989. 34. Watabc, S.; Sendo, F.; Kimura, S., et al. Activation of cytotoxic polymorphonuclcar leukocytes by in vivo administration of a streptococcal preparation, OK-432. J. Natl. Cancer Inst. 72:1365-1370; 1984.
POLYMORPHONUCLEAR LEUKOCYTES-MEDIATED LYSIS OF A 4 3 1 CELLS INDUCED BY IgG1 MOUSE ANTI-EPIDERMAL GROWTH FACTOR RECEPTOR MONOCLONAL ANTIBODIES TOMOYUKI KAWAMOTO, KOUJI K1SHIMOTO, KOJIRO TAKAHASHI, TOMOHIRO MATSUMURA, J. DENRY SATO, AND SHIGEHIKO TANIGUCHI
Department of Biochemistry (T. K., K. T., S. T.) and Department of Maxillofacial Surgery H (K. K., T. M.), Okayama UniversityDental School, Shikata 2-5-1, Okayama City 700, Japan; and IV. Alton Jones Cell Science Center, Inc. (J. D. S.), 10 Old Barn Road, Lake Placid, New York 12946 (Accepted 27 July 1992)
SUMMARY
The abilityof different Fc receptors (Fc~ R) on lgG-mediated eytotoxicityfor humanepidermoidcarcinomaA431 cells bearing large number of epidermal growth factor receptors (EGFRs) was examined by using two isotypes (IgG1 and IgG2a) of murine monoclonal antibodies (MoAbs) against EGFR in the presence of human monocytes and granulocytes. Two MoAbs (225 and LA1) of the IgG1 isotype exhibited effective cytolytic activity for A431 cells with human polymorphonuclear leukocytes (PMN) rather than with human mononuclear cells (MNC). In contrast, two MoAbs (528 and 579) of the IgG2a-isotype lysed the cells less effectively with PMN than with MNC. Anti-Fc'y R II (CDw32) MoAb 2 E l inhibited the IgGl-mediated cytotoxicity by PMN, and anti-Fc'y R III (CD16) MoAb 80H3 did not inhibit the IgG2a-mediated cytotoxicity by MNC. Under these conditions, antibody-dependent cell-mediated cytotoxicity by mouse MoAb IgG1 isotypes resulted from antibody binding to the Fc'y R II (CDw32) of PMN.
Key words: EGF receptor; monoclonal antibody; polymorphonuclear leukocyte; mononuclear leukocyte; cytotoxicity.
INTRODUCTION
MATERIALSANDMETHODS
As the proliferation of cancer cell lines is regulated by hormones and growth factors in vitro, and it has been suggested that a similar regulatory mechanism is operating in vivo (3), there have been several attempts to use monoclonal antibodies (MoAbs) against hormone or growth factor receptors for analysis of the regulatory mechanism and for immunotherapy of malignant tumors (1,21,23,25,30). Trowbridge and Domingo (30) reported that treatment with anti-transferrin receptor MoAb inhibited tumor formation by a human melanoma cell line in athymic mice. Masui et al. (21) reported that anti-epidermal growth factor receptor (EGFR) MoAbs inhibited tumor formation of A431 cell xenografts in athymic mice and that MoAb 528-IgG2a induced A431 cell lysis with activated mouse macrophages or rabbit complement in vitro (22). Rodeck et al. (25) also showed that anti-EGFR MoAb 425-IgG2a directed A431 cell lysis by human monocytes, lymphocytes and mouse macrophages. It has been proposed that the IgG2a isotype MoAbs were most effective in monocyte- or macrophage-mediated ADCC (antibody-dependent cell-mediated cytotoxicity) among MoAbs against tumor-associated antigens (including receptors) (6,10,13,18). However, little is known about polymorphonuclear leukocyte (PMN)-mediated ADCC, though PMN has Fc receptors for IgG (Fc'y R) and is able to promote cytotoxicity of target tumor cells (16,17). In this report, we propose a mechanism of PMN-mediated ADCC employing human PMN and mouse anti-EGFR MoAb isotype IgG1.
Target cells. Human epidermoid carcinoma A431 ceils, which express an unusually large number of EGFR (1-3 X 106/cell), were used as target tumor cells in the present study. The A431 cells were maintained in a mixture (DME/F12-5% FBS) of Dulbecco's modified Eagle's and Ham's F-12 medium (Nissui Seiyaku Co. Ltd., Japan) containing 15 mM H EPES and heat-inactivated 5% fetal bovine serum (FBS: GIBCO Laboratories, USA) at 37 ° C in a 5% CO2 and 95% air atmosphere. The A431 cells were incubated with 0.1% trypsin and 1 mM EDTA at 37 ° C for 5 rain, harvested, washed with DME/F12-5% FBS medium, and then used for cytolysis assay. Effector cells. Human PMN and mononuclear cells (MNC) used as effector cells were separated from heparinized healthy human peripheral blood with Mono-Poly Resolving Medium (M-PRM: Flow Laboratories Ltd., Australia) as described previously (4). Four ml of human whole blood were layered onto 3 ml of M-PRM in a sterilized 17 × 120 mm test tube (Falcon 2095: Becton Dickinson Overseas Inc., USA). The tube was centrifuged at 300 Xg for 30 min at room temperature, and the top band of MNC fraction and second band of PMN fraction were transferred into individual tubes. The MNC and PMN, histologically distinguishable each other by Diff Quick Dyeing (Kokusai-Shiyaku, Kobe), were thereby quantitated and the contamination thus estimated was below 5% in each sample. After washing with DME/F12 medium, the MNC and PMN cells were resuspended to the required concentration in DME/F12-5% FBS medium and used as the effector cells. Viability of the effector cells in this method was estimated to be over 99% by means of the trypan blue dye exclusion test. The top leukocyte band contained over 94% MNC (lymphocytes and monocytes), whereas the second band contained over 96% PMN. Monoclonal antibodies. Four mouse MoAbs (528- and 579-MoAbs as IgG2a isotype and 225- and LA1-MoAbs as IgG1 isotype) raised against EGFR of A431 ceils were used for ADCC assay. The development and characterization of these MoAbs have been reported (12,26,27). LAI has
782
783
POLYMORPHONUCLEAR LEUKOCYTE-MEDIATED CELL LYSIS 0r,II
; "oF" .... (A)
-]
)
"
, ,
,
*
lO
0 0.01 0.1 1 10
25
50
~
)
•
I
40
,
i
m
(B)
i,o 100
0 0.1 1 10 [ LAI-lgG1
[ 225-1gG1 ] ( nM )
25 50 1( nM )
Fie. 1. Cytolytic activity for A431 cells dependent on mouse anti-EGFR MoAb IgG1 isotype (225 and LA1). Experimental procedures were described in Materials and Methods. Results are expressedas mean -+ SD of triplicate data. PMN or MNC (8 X 105 cells/well) as cffector cells and [aH]uridine-labeled A431 (5 × 103 cells/welt) as target cells were incubated with increasing concentrations of (A) 225-IGG1 or (B) LAI-lgG1 for 24 b. Effector cells used were PMN (Q), MNC (A) and without cells (V1). Ratio (E/T) on cell number of the effector to the target was 160.
also been characterized recently as IgG1 isotype (Sato, J. D. et al., unpublished result). These four MoAbs bind to the extracellular peptide domain of EGFR and compete with EGF for binding to EGFRs. The Fab fragments of 225- and 528-MoAbs were prepared with lmmunopure Fab Preparation Kit (Pierce, USA) and were shown to be free of intact IgG and Fc fragments
100 --/~ . . . .
n
,
by analysis of 10% SDS-PAGE under nonreducing conditions. MoAb ARB229 (lgG1 isotype), developed against a haptene (tyrosyl arsenate) (Hybritech, Inc., USA), was used as control. Mouse MoAb 2 E l developed against human Fc7 R II (CDw32) (7,11,19,29) and MoAb 80H3 specific to the CD16 Fc'y RIII (20) were obtained from Immunotech (France).
,
100 -I~ ,
(A) o~
80
"-i
60
,
, #
,
,
(B)
A
80
40
0
0
2O
O:
0 0.01 0.1 1 10
25 50
[ 528-1gG2a ] ( nM )
100
0 0.1 1 10
25 5O
[ 579-1gG2a ] ( nM )
Fie. 2. Cytolytic activity for A431 cells dependent on mouse anti-EGFR MoAb IgG2a isotype (528 and 579). (A) and (B) for 528and 579-lgG2a isotypes, respectively. Experimental procedures were the same as in Fig. 1. Effector cells used were PMN (O), MNC (L~) and without cells (E3). Ratio of E / T was 160.
784
KAWAMOTO ET AL.
50["
1
I
!
I
|
A
o~
40
.>_
30
20
0
10
0 10 20 40
5
80
120
160
E/T ratio Ftc. 3. Influence of effector cell number of 225-lgGl-dependent cytolyric activity. [sH]Uridine-labeled A431 (5 × 103 celts/well) was incubated with increasing numbers of effector cells in the presence or absence of 17 nM 225-1gG1 for 24 h. PMN with 225-IGG1 (e), PMN without 225-IGG1 (©), MNC with 225-IGG1 (A) and MNC without 225-IGG1 (A). Results are expressed as mean + SD of triplicate data.
against A431 target cells in the presence of MNC as mentioned by Masui et al. (22), but showed remarkable cytolytic activity in the presence of PMN (Fig. 1 A). LAI-IgG1 also showed a similar cytolytic activity with PMN as shown also in Fig. 1 B. 528- and 579IgG2a isotypes, on the other hand, had little cytolytic activity with PMN (Fig. 2), even though they showed significant cytolytic activity in the presence of MNC (Fig. 2) as reported previously (6,13,18,22). Clear ADCC activity of 225-IGG1 was evident at antibody concentrations above 10 nM (Fig. 1). At 17 nM of 225IgG1, the cytolytic activity increased in accordance with the increased ratio of effector cell (PMN)/target cell (A431) (Fig. 3). A ratio of PMN/A431 in excess of 160 was not practical for the ADCC assay, owing to the difficulty of handling higher numbers of PMN in the assay. Fab fragments of both 225-IGG1 and 528-lgG2a induced no significant A431 cell lysis with PMN or MNC. Mouse MoAb ARB229 lgG1, which was incapable of binding to EGFR augmented ADCC activities with neither PMN nor MNC (data not shown). These results demonstrated that 225- and LAI-IgG1 mediated A431 cell lysis by PMN through binding to the EGFR of the target cells, whereas 528- and 579-IgG2a mediated A431 cell lysis by MNC more effectively than by PMN.
Effect of anti-Fc'y R MoAbs on 225-IgG l-dependent cytolytic activity by PMN. Human PMN constitutively expresses two types of Fc receptors for lgG (Fc'y R): Fc'y R II (CDw32) (15) and CD16 Fc'y R III (8). Mouse MoAb 2 E l capable of binding specifically to Fc'y R II (CDw32) (7,11,19,29) inhibited the 225-IGG1 mediated cytolytic activity by PMN, but MoAb 80H3 specific to CD16 Fc~y R Ill (20) did not inhibit the ADCC activity by PMN (Fig. 4). Another anti-EGFR MoAb LAI-IgGI gave similar results as 225-IGG1 (data not shown). These results indicated that IgGl-de-
A
40
Cytolysis assay. Tumor cell cytolysis can be observed with the [SH]uridine release assay (34). A431 cells (5)< 10s) were placed into each of 96-well flat-bottomed microplates (Falcon 3072) containing 0.2 ml of DME/F12-5% FBS medium. After preincubation for 6 to 8 h, 0.2 ~tCiof [SH]uridine (2-10 Ci per mmol: NEN Research Products, USA) was added to each well. The A431 cells were incubated overnight at 37 ° C, washed three times with DME/F12-5% FBS medium, then both effector cells and anti-EGFR MoAbs were added in a total volume of 0.2 ml DME/F12-5% FBS medium per well. After incubation for 24 h, sample solutions from the microplates were centrifuged at 200 ×g for 5 min. Using 0.1 ml aliquot of each supernatant, the experimental release (ER) of radioactivity from target cells was measured with a liquid scintillation counter. The cytolytic activity [CA(%)] was calculated by the following equation:
20
CA(%) = 100 X {ER(cpm) - SR(cpm)}/{MR(cpm) - SR(cpm)} where SR = spontaneous release of radioactivity from the target cells cultured without the effector cells and MoAbs, and MR = maximum release of radioactivity from target cells after treatment with 1% Triton X-100. The obtained CA(%) values are expressed as the average of triplicate data with the standard deviation. ADCC inhibition assay by anti-FcT R MoAb. [sH]Uridine-labeled A431 cells (5 X 103/well) were incubated with the effector cells (8 X 10S/well) and varied concentrations of anti-Fc7 R MoAb (2El or 80H3) at 37 o C for 30 min. 17 nM of anti-EGFR MoAbs was then added, incubated for 24 h, and the CA(%) value of each well was estimated as described above. RESULTS
Anti-EGFR MoAb lgG1 isotype-dependent cell-mediated cytotoxicity. 225-IGG1 isotype showed no significant cytolytic activity
O 0
~ 0
1
10
100
[ Anti-FcyR MoAb ] ( nM ) Fic. 4. Effects of anti-FcT R II MoAb 2El and anti-Fc'y R Ill MoAb 80H3 on 225-1gGl-dependent cytolytic activity mediated by PMN. [SH]Uridine-labeled A431 (5 X 103 cells/well) and PMN (8 X 105 cells/ well) cells were incubated with increasing concentrations of anti-Fc'y R MoAbs for 30 min. 225-IGG1 (17 riM) was then added and incubation was continued for 24 h. Results are expressed as mean + SD of triplicate data with MoAb 2El (e) and MoAb 80H3 (C)).
POLYMORPHONUCLEAR LEUKOCYTE-MEDIATED CELL LYSIS TABLE 1 ADCC EFFECTS OF ISOTYPE IgG AND THE INHIBITION BY ANTI-Fcs' R MoAbs IN THE PRESENCE OF MNC OR PMN~ ADCC(%)
IgG1 isotype: ARB229 225 LA1 IgG2a isotype: 528 579 Anti-Fc'y RII MAb 2El: Anti-Fc'y RIII MAb 80H3:
225 528 225 528
MNC
PMN
0 6 13 52 72
0 43 45 9.5 2
-3.5 (-9) -2.5 -11
-23 -6.5 -11 -3.5
a Ratio of E/T: 160. The concentrations of IgGs and anti-FcR Ab were both 100 nM.
pendent cytolytic activity of PMN was mediated by binding to the Fc"/R II (CDw32) of PMN. DISCUSSION Since Steplewski et at. (28) proposed that macrophages are effector cells for human tumor destruction directed by MoAbs against tumt)r cells, mouse MoAb of the IgG2a isotype was found to be most effective in macrophage- or lymphocyte-mediated ADCC (6,10,13,18,22,25). Although 225-IGG1 completely inhibited the growth of A431 cell xenografts in athymic mice (21), the role of mouse MoAb IgG1 isotype in ADCC remained unclear because 225-1gG1 did not induce A431 cell lysis by activated mouse macrophages (22). As summarized in Table 1, we showed that mouse anti-EGFR MoAbs of the IgG1 isotype (225 and LA1) could lyse A431 cells in the presence of human PMN as effectively as IgG2a antibodies (528 and 579) do in the presence of human MNC. The inhibition of IgG1 isotype-dependent cytolysis by MoAb 2 E l specific for CDw32 Fc"/R II of human PMN indicated that 225- and LAI-IgG1 isotype MoAbs directed cytolytic activity by binding to the CDw32 Fc7 R II of PMN. This result is consistent with the findings that murine IgG1/2b antibodies bind to Fc3' R I1 (2,31), and that Fc'y R II of PMN could be the cytotoxic trigger molecule for tumor cells (9,14). In the present study, we demonstrated a cytolytic activity mediated by human PMN in an in vitro system. ADCC, when examined in vivo, may prove more complex. Lopez et al. (•6) reported that mouse granulocytes were cytotoxic to target cells coated with mouse IgG1, IgG2a or IgG2b antibodies, and that MoAb 4.2 G2 reactive with mouse Fc'y R blocked the binding of these antibody-coated cells to the granulocytes. This contrasts with our finding that mouse lgG1 mediated A431 cell lysis by human PMN more effectively than in the case of IgG2a. In our study, however, human PMN instead of mouse PMN was used and Fc'y R I capable of binding to mouse IgG2a was shown to be undetectable on fresh human PMN. Recently, it was found that interferon-"/induced Fc'y R I on human PMN (24). In previous studies, we ascertained that 528-IgG2a also effectively mediated the lysis of A431 cells by human PMN in the presence of interferon-'y (data not shown), Human monocytes have two types of FcR for IgG, Fc'y R I and Fc'y R II (32). Boot et al. (5)
785
reported that human monocytes could be triggered to ADCC by routine MoAb IgG1 which was predominantly mediated through Fc3' R II of the monocytes. In our study, however, the 225- or LAI-IgG1 isotype was less effective than the case of 528-IgG2a isotype to the ADCC by MNC (Fig. 1). It remains to be clarified whether the Fc'y R II s of human PMN and MNC are essentially identical. Modification of the conformation of these Fc'y Rs could contribute to differences between PMN and MNC; Van de Winkel et al. (33) reported that the interaction of low responder monocytes with mouse IgG 1 was dramatically increased by protease treatment. Comprehensive and well rounded studies of ADCC are urgently required. REFERENCES 1. Abound-Pirak, E.; Hurwitz, E.; Pirak, M. E., et al. Effcacy of antibodies to epidermal growth factor receptor against KB carcinoma in vitro and in nude mice. J. Natl. Cancer Inst. 80:1605-1611; 1988. 2. Anderson, C. L.; Ryan, D. H.; Loony, R. J., et al. Structural polymorphism of the human monocyte 40 kilodahon Fc receptor for IgG. J. Immuuol. 138:2254-2256; 1987. 3. Barnes, D.; Sato, G. Methods for growth ofcuhured cells in serum-free medium. Anal. Biochem. 102:255-270; 1980. 4. Bignald, L. P.; Ferrante, A. Mechanismof separation of polymorphonuclear leukocytes from whole blood by the one-step Hypaque-Ficoll method. J. Immunol. Methods 96:29-33; 1987. 5. Boot, J. H. A.; Geerts, M. E. J.; Aarden, L. A. Functional polymorphisms of Fc receptors in human monocyte-mediated cytotoxicity towards erythrocytes induced by murine isotype switch variants. J. Immunol. 142:1217-1223; 1989. 6. Denkers, E. Y.; Badger, C. C.; Ledbetter, J. A., et al. Influence of antibody isotype on passive serotherapy of lymphoma. J. Immunol. 135:2183-2185; 1985. 7. Farace, F.; Mitajavila, M. T.; Betaieb, A., et al. New hematopoietic differentiation antigens detected by anti-K562 monoclonal antibodies. Cancer Res. 48:5758-5765; 1983. 8. Fleit, H. B.; Wright, S. D.; Unkeless, J. C. Human neutrophil Fc'y receptor distribution and structure. Proc. Nat. Acad. Sci. USA 79:3275-3279; 1982. 9. Graziano, R. F.; Fanger, M. W. Fc'y RI and Fc'y RII on monocytes and granulocytes are cytotoxic trigger molecules for tumor cells. J. Immunol. 139:3536-3541; 1987. 10. Heike, Y.; Hamada, H.; Inamura, N., et al. Monoclonal anti-P-glyeoprotein antibody-dependent killing of multidrug-resistant tumor cells by human mononuelear cells. Jap. J. Cancer Res. 81:11551161; 1990. 11. Hogg, N.; Horton, M. A. Myeloidantigens: new and previously defined clusters. In: McMichael, A. J., et al., eds. Leucocyte typing. IIl. White cell differentiation antigens. Oxford, England: Oxford Univ. Press; 1987:576-602. 12. Kawamoto, T.; Sato, J. D.; Le, A., et al. Growth stimulation of A431 cells by epidermal growth factor: identification of high-affinity receptors for epidermal growth factor by an anti-receptor monoclonal antibody. Proc. Natl. Acad. Sei. USA 80:1337-1341; 1983. 13. Kipps, T. J.; Paraham, P.; Punt, J., et al. Importance of immunoglobulin isotype in human antibody-dependent, cell-mediated eytotoxicity directed by murine monoclonal antibodies. J. Exp. Med. 161:1-17; 1985. 14. Lanier, L. L.; Ruitenberg, J. J.; Phihps, J. H. Funetional and bioehemical analysis of CD16 antigen on natural killer cells and granulocytes. J. ImmunoL 141:3478-3485; 1988. 15. Loony, R. J.; Ryan, D. H.; Takahashi, K., et at. Identification of a second class of lgG Fc receptors on human neutrophils: a 40 kilodalton molecule also found on eosinophils. 1. Exp. Med. 163:826836; 1986. 16. Lopez, A. F.; Strath, M.; Sanderson, C. J. Mouse immunoglobulin isotypes mediating cytotoxicity of target cells by eosinophils and neutrophils. Immunology 48:503-509; 1983.
786
KAWAMOTO ET AL.
17. Lopez, A. F.; Battye, F. L.; Vadas, M. A. Fc receptors on mouse neutrophils and eosinophils: antigenic characterization, isotype specificity and relative cell number density measured by flow cytometry. Immunology 55:125-133; 1985. 18. Lubeek, M. D.; Steplewski, Z.; Saglia, F., et al. The interaction of murine IgG subclass proteins with human monocyte Fc receptor. J. Immunol. 135:1299-1304; 1985. 19. Macintyre, E. A.; Roberts, P. J.; Abdul-Gaffar, R., et al. Mechanism of human monocyte activation via the 40-kDa Fc receptor for IgG. J. lmmunol. 141:4333-4343; 1988. 20. Mannoni, P.; Janowska-Wieczorek, A.; Turner, A. R., et al. Monoclohal antibodies against human granulocytes and myeloid differentiation antigens. Human lmmunol. 5:309-323; 1982. 21. Masui, H.; Kawamoto, T.; Sato, J. D., et al. Growth inhibition of human tumor cells in athymic mice by anti-epidermal growth factor receptor monoclonal antibodies. Cancer Res. 44:1002-1007; 1984. 22. Masui, H.; Moroyama, T.; Mendelsohn, J. Mechanism of anti-tumor activity in mice for anti-epidermal growth factor receptor monoclonal antibodies with different isotype. Cancer Res. 46:5592-5598; 1986. 23. Masui, H.; Kamrath, H.; Apell, G., et al. Cytotoxicity against human tumor cells mediated by the conjugate of anti-epidermal growth factor receptor monoclonal antibody to recombination ficin A chain. Cancer Res. 49:3482-3488; 1989. 24. Perussia, B.; Dayton, E. T.; Lagarus, R., et al. Immune interferon induces the receptor for monomeric IgGl on human monocytic and morpeloid cells. J. Exp. Med. 158:1092-1113; 1983. 25. Rodeck, U.; Herlyn, M.; Herlyn, D., et al. Tumor growth modulation by a monoclonal antibody to the epidermal growth factor receptor: immunologically mediated and effector cell-independent effects. Cancer Res. 47:3692-3696; 1987.
26. Sato, J. D.; Kawamoto, T.; Le, A., et ak Biological effects in vitro of monoclonal antibodies to human epidermal growth factor receptors. Mol. Biol. Med. 1:511-529; 1983. 27. Sato, J. D.; Le, A. D.; Kawamoto, T. Derivation and assay of biological effects of monoclonal antibodies to epidermal growth factor receptors. Methods Enzymol. 146:63-81; 1987. 28. Steplewski, Z.; Herlyn, D.; Maul, G., et al. Hypothesis: macrophages as effector cells for human tumor destruction mediated by monoclonal antibody. Hybridoma 2:1-5; 1983. 29. Tenero, P. A. T.; Van der Shoot, E.; Visser, F. J., et al. Three different types of Fc'y receptors on human leucocytes defined by workshop antibodies: Fc~y Rl0~ on neutrophils, Fc'y w~ of K/NK lymphocytes and Fc'y R 11. In: McMichael, A. J., et al., eds. Leucocyte typing. 11I. White cell differentiation antigens. Oxford, England: Oxford Univ. Press; 1987:702-706. 30. Trowbridge, I. S.; Domingo, D. L. Anti-transferrin receptor monoclonal antibody and toxin-antibody conjugates affect growth of human tumor cells. Nature 294:171-173; 1981. 31. Van de Winkel, J. G. J.; Van Duijnhoven, H. L. P.; Van Ommen, R., et al. Selective modulation of two human monocyte Fc receptors for IgG by immobilized immune complexes. J. lmmunoi. 140:35153521; 1988. 32. Van de Winkel, J. G. J.; Boonen, G. J.; Janssen, P. L., et al. Activity of Fc receptors, Fc gamma R I and Fc gamma R II, in human monocyte cytotoxicity to sensitized erythrocytes. Scand. J. Immuno|. 29:23-31; I989. 33. Van de Winkel, J. G. J.; Van Ommen, R.; Huizinga, J. W. J., et al. Proteolysis induces increased binding affinity of the monocyte type II FcR for human IgG. J. Immunol. 143:571-578; 1989. 34. Watabc, S.; Sendo, F.; Kimura, S., et al. Activation of cytotoxic polymorphonuclcar leukocytes by in vivo administration of a streptococcal preparation, OK-432. J. Natl. Cancer Inst. 72:1365-1370; 1984.
