[CANCER RESEARCH 53, 362-367. January 15, 19931

A Mechanism for Neutrophil-mediated Lysis of Human Neuroblastoma Cells I

Edward Barker and Ralph A. Reisfeld 2

Department of lmmunology, The Scripps Research Institute, 10666 North Torrey Pines Road, IMMI3, La Jolla, California 92037

A B S T R A C T

Neutrophils mediate the lysis of human neuroblastoma cells coated with human/mouse chimeric anti-GD2 ganglioside antibody ch14.18. This study examined the mechanism(s) by which this occurs. Neutrophil degranula- tion was found to be a required step for lysis, since release of granular enzymes from neutrophils correlated with the lysis of antibody-coated neuroh!astoma cells. In addition, agents which block degranulation spe- cifically inhibited this process. Antibody-dependent lysis of neuroblastoma cells was enhanced by exposing neutrophils to granulocyte-macrophage colony stimulatory factor. An increased release of lytic granular molecules was found to be responsible for this lymphokine-mediated phenomenon. Among the molecules released from neutrophil granules that were shown to be involved in neuroblastoma cell lysis were defensins, Mr 30011-4000 neutrophil granular proteins which are known to bind and permeabilize tumor cells. In addition, cathepsin-G, a neutrophil granular protease, was demonstrated for the first time to mediate the lysis of human neuroblas- toma cells. The enzymatic activity of cathepsin-G was found to be required for the lysis of these tumor cells, since phenylmethyisulfonyl fluoride blocks the lytic ability of this protein.

I N T R O D U C T I O N

It is well established that neutrophils are involved in the destruction

of microbial pathogens. Several reports indicate these cells to be

capable of mediating the lysis of leukemic as well as solid tumor cells

(1-7). The mechanisms utilized by neutrophils to mediate tumor cell

lysis include the generation of reactive oxygen intermediates (2, 8)

and/or the release of lytic molecules prepackaged in their granules

(9, 10). The lysis of tumor cells by neutrophils also was shown to be enhanced by various cytokines (6, 7, 11-14).

Recently we (7) as well as others (5, 6) reported that mouse mon-

oclonal antibodies directed against disialoganglioside Gt)2 surface

antigens can mediate the lysis of human neuroblastoma cells by neu-

trophils. We also observed that human/mouse chimeric anti-GD2 an-

tibody ch14.18 mediates the lysis of human neuroblastoma cells by

neutrophils more effectively than its mouse counterpart, 14.G2a,

which recognizes the same antigen (7). In addition, we (7) and others (5, 6) have demonstrated that GM-CSF 3 can enhance antibody-medi-

ated lysis of human neuroblastoma cells by neutrophils.

Although the mechanism(s) by which neutrophils lyse neuroblas-

toma cells and GM-CSF enhances this process is still unknown, it has

been reported that reactive oxygen intermediates are not involved in the lysis of neuroblastoma cells. This conclusion is based on the

observation that scavengers of oxygen metabolites do not affect the

ability of neutrophils to lyse antibody-coated neuroblastoma cells (5, 15). In addition, neutrophils from individuals with chronic gran-

ulomatous disease which are deficient in NADPH oxidase and con-

Received 8/3/92; accepted 11/5/92. The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

l Supported by NIH Grant IR35CA42508. This is Scripps Research Institute manu- script number 7173-IMM.

2 To whom requests for reprints should be addressed, at Department of Immunology, IMMI3, The Scripps Research Institute, 10666 North Torrey Pines Road, La Jolla, CA 92037.

3 The abbreviations used are: GM-CSF, granulocyte-macrophage colony-stimulating factor; ADCC, antibody-dependent cellular cytotoxicity; DIDS, 4,4'-diisothiocyanostil- bene-2,2'-disulfonic acid; DMX, 1,3-dimethylxanthine; ISO, l-(3',4'-dihydroxyphenyl)- 2-isopropyl aminoethanol; HBSS, Hanks' balanced salt solution; PMSF, phenylmethyl- sulfonyl fluoride.

sequently cannot produce superoxide anions (16) are still able to

mediate the lysis of anti-GD2 antibody-coated neuroblastoma cells

(5, 15).

We report here on a nonoxidative mechanism by which antibody-

coated human neuroblastoma cells are lysed by neutrophils. We indi-

cate the importance of neutrophil degranulation in the lysis of human

neuroblastoma cells and demonstrate that GM~CSF increases anti-

body-dependent lysis of these cells by enhancing degranulation. We

also identified cathepsin-G and defensins as molecules within neutro-

phil granules that are responsible for mediating the lysis of human

neuroblastoma cells.

M A T E R I A L S A N D M E T H O D S

Reagents. DIDS, DMX, and ISO were purchased from Sigma Chemical Co. (St. Louis, MO) and prepared for use in the lytic assay as described by Dallegri et al. (4). PMSF was also obtained from Sigma Chemical Co. and prepared as described elsewhere (17). Elastase, cathepsin-G, lysozyme, and myeloperoxidase, purified from human neutrophils, were purchased from ICN Biochemical Co. (Irvine, CA). The purity of cathepsin-G was >98% by SDS- PAGE. The purified defensin molecule HNP-1 was kindly provided by Dr. Robert I. Lehrer (University of California, Los Angeles, CA). Recombinant human GM-CSF was a gift from Behring Werke (Marburg, Germany).

Cell Lines and Monocional Antibody. The human neuroblastoma cell line NMB-7 was a gift from Dr. S. K. Liao (McMaster University, Hamilton, Ontario, Canada) and has been maintained in our laboratory for more than 10 years. The human neuroblastoma cell lines LAN-1 and LAN-5 were kindly provided by Dr. R. Seeger (University of Southern California, Los Angeles, CA). These cell lines were grown as described previously (7). Human/mouse chimeric anti-GD2 antibody ch 14.18 was constructed by Dr. Stephen D. Gillies (Abbott Biotech, Boston, MA) as described elsewhere (18).

Isolation of Neutrophils. The source of human neutrophils was venous blood obtained from healthy donors. The blood was collected into 7-ml vacu- tainer tubes (Becton Dickinson, Mountain View, CA) containing the potassium salt of EDTA as an anticoagulant (19). Neutrophils were isolated according to the method of Wright (20). The number of neutrophils recovered varied from 1 X 106 to 2 x 106 cells/ml of whole blood. Percentage viability always exceeded 98% as determined by exclusion of trypan blue dye (0.4%; Sigma Chemical Co.).

Isolation of Neutrophil Granular Proteins. Neutrophil granular proteins were isolated according to the methods described by Okrent et aL (21) and Ganz et al. (22), which were applied with some modifications. Human neu- trophils (0.5-1 X 101~ were routinely obtained from a single healthy donor by leukopheresis. Contaminating erythrocytes were removed by dextran sedimen- tation and hypotonic lysis (20). Neutrophils were suspended at a final concen- tration of 108 cells/ml in Ca2+-free HBSS (Irvine Scientific Co., Santa Ana, CA) containing 2.5 rnM MgC12 (Mallinckrodt, Inc., Paris, KY), pH 7.4. The cell suspension was placed in a nitrogen bomb and pressurized to 750 psi for 20 min at 4~ The suspension was released dropwise into Ca 2§ and Mg2+-free HBSS (Irvine Scientific Co.) containing 5 rnM EDTA (Sigma Chemical Co) at pH 7.4, while stirring. This was followed by centrifugation (Technospin R; Dupont Co., Burbank, CA) at 250 x g for 10 min to remove nuclear and cellular debris. More than 90% of neutrophils were lysed by this procedure as determined by exclusion of trypan blue dye (0.4%; Sigma Chemical Co.). The supernatant was centrifuged (L3-50 ultracentrifuge; Beckman Instruments, Fullerton, CA) at 27,000 x g for 30 min. The pellet was resuspended in either 0.4 M sodium acetate (JT Baker, Inc., Phillipsburg, NJ), pH 4.0, or 35 rnM acetic acid (Mallinckrodt, Inc.) at a final concentration of 2 • 108 cell equivalents/ml. Proteins from the neutrophil granules were extracted over a 12-h period at 4~ with gentle stirring. The residue was removed by centrifugation at 27,000 x g for 30 min, and the supernatants were collected and stored at -70~ Extraction of granular proteins from the pellet was repeated at least twice. The extracts obtained by acetic acid treatment of neutrophil granules were pooled and

362

Research. on July 4, 2019. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

NEUTROPH1L-MEDIATED I.YSIS OF NEUROBLASTOMA CELLS

concentrated by lyophilization (Virtis, Gardiner, NJ). Extracts obtained by acetate treatment of neutrophil granules were concentrated by ultrafiltration (YC-05 filter; Amicon Corporation, Beverly, MA). The two different extraction methods were chosen since defensins are most effectively extracted by acetic acid, a procedure which was reported to destroy the activity of the enzymes in the various fractions (23). Each extract was concentrated to a final volume of 5.0 ml and layered on top of a Sephadex G-100 (Pharmacia, Piscataway, NJ) gel filtration column (200 cm • 1.5 cm; BioRad, Richmond, CA). Elution of proteins was achieved with 0.4 M sodium acetate, pH 4.0, containing 0.1 M CaCI2 (Sigma Chemical Co.) or 17.5 mM acetic acid, respectively. Fractions (10 ml) were collected and measured at 280-nm wavelength with a UV-160 spectrophotometer (Cole Scientific, Chicago, IL). Essentially the same A~so profiles were observed as those described by others (21, 22). The respective Kd values at which each fraction eluted were calculated as:

v~-v0 Kd- V,- Vo

supematant was combined with a reaction mixture (0.075 ml) consisting of 4.5 mM peptide and 0.33 mM Ellman's Reagent in phosphate-buffered saline (lrvine Scientific Co.), pH 7.4, and 10% dimethyl sulfoxide (Sigma Chemical Co.). The plates were incubated at room temperature for 3-5 h and then read at 405 nm in an EL-310 enzyme-linked immunosorbent assay microplate reader (Biotek Instruments, Burlington, VT).

Percentage of degranulation was calculated as:

Absorbance of experimental release - mean absorbance of spontaneous release

Mean absorbance of maximal release - mean absorbance of spontaneous release

X 100

Although the percentage of specific degranulation achieved by neutrophils from various individuals varied from experiment to experiment, the overall pattern of results remained the same.

where Ve is the elution volume, V0 is the void volume (208.5 ml), and Vt is the total column volume (567 ml). The Kd value for the fractions combined from the column eluted with sodium acetate buffer were 0.22--0.28 (fraction A); 0.40-0.41 (fraction B); 0.62-0.63 (fraction C). A Ko value of 0.80-0.82 was calculated for the fraction obtained from the column eluted with acetic acid that contained defensins. Fractions corresponding to these respective Ka values that represented the major peaks in the A28o profiles of the gel filtration columns were pooled and concentrated by either ultrafiltration or lyophilization and then dialyzed against phosphate-buffered saline (Irvine Scientific Co.), pH 7.4, in spectapore 3 tubing with a molecular weight cut-off of 2000 (Spectrum, Houston, TX).

Protein Determination. Protein concentration was measured with the BCA protein assay system (Pierce Chemical Co., Rockford, IL) using chicken egg lysozyme (ICN Biochemical Co.) as the standard. Each fraction as well as purified neutrophil proteins were subjected to electrophoresis on 12.5% poly- acrylamide gels in 6.25 M urea at a final pH of 4.0.

Lytie Assays. Assays of ADCC were performed as described previously (7). In cases where inhibitors were used, 0.05 ml of neutrophils (ranging from 0.5 X 10 6 tO 2 • 10 6 cells) were incubated with 0.05 ml of each inhibitor in U-bottomed 96-well plates (Coming Glass Works, Coming, NY) for 1 h at 37~ prior to adding antibody and target cells. Fetal bovine serum (Gibco, Grand Island, NY) was omitted from lytic assays which evaluated the ability of either proteins extracted from neutrophil granules or purified neutrophil proteins to lyse neuroblastoma cells. Although the percentage of specific lysis achieved by neutrophils and their granular proteins varied somewhat from experiment to experiment, the overall pattern of results remained the same.

Degranulation Assay. Degranulation was measured by the release of elastase in an assay described by Boros et al. (24), which was applied with slight modifications. Briefly, neutrophils (2 x 10 6 cells) were suspended in HBSS without phenol red (Irvine Scientific Co.) containing 0.1 mg/ml bovine serum albumin (Sigma Chemical Co.) and added to 24-well tissue culture plates (Linbro; Flow Laboratories, McLean, VA). Next, 1 • 105 NMB-7 neuroblastoma cells were added to each well. Either 0.025 ml of HBSS-bovine serum albumin or ch14.18 were placed in the appropriate wells. The final concentration of the antibody in each well ranged from 0.01 to 1 pg/ml. In cases where GM-CSF was used, 0.025 ml of the cytokine was added to the appropriate wells to a final concentration of 100 ng/ml. Each well was adjusted to 1 ml total volume with HBSS-bovine serum albumin. Control wells con- sisted of those containing neutrophils without antibody, tumor cells, or GM- CSE Maximal release of elastase was achieved by adding 0.1% Nonidet P-40 (Sigma Chemical Co.) in 0.125 ml phosphate-buffered saline (Irvine Scientific Co.) to neutrophils instead of antibody and tumor cells. Each treatment group was tested in triplicate. All plates were incubated at 37~ for 3 h.

Enzyme release was quantified by taking 0.1-ml aliquots of supematant from the 24-well plates and placing them into V-bottomed 96-well plates (Flow Laboratories). These plates were then centrifuged for 5 min at 250 x g (TJ-6 Centrifuge; Beckman Instruments) to remove any cells that were present in the supernatants. Aliquots of 0.025 ml were removed from each well and placed into fiat-bottomed 96-well microtiter plates (Flow Laboratories). Elastase activity of neutrophils was assayed by cleavage of N-t-BOC-ala-pro- nva thiobenzyl ester (Sigma Chemical Co.) in the presence of Ellman's reagent [5,5'-dithio-bis-(2-nitrobenzoic acid)] (Pierce Chemical Co.). Each aliquot of

RESULTS

Ability of Antibody-coated Neuroblastoma Cells to Mediate Neutrophil Degranulation. We have previously reported that human

neutrophils lyse monoclonal antibody ch 14.18-coated neuroblastoma

cells in a 4-h 5~Cr release assay (7). We now determined the extent of

neutrophil degranulation in this lytic process and demonstrated its

importance (Fig. 1). Here a clear correlation is seen between the

magnitude of degranulation (Fig. la) and the extent of tumor cell lysis

(Fig. lb). In the absence of antibody, the NMB-7 neuroblastoma cells

were unable to induce degranulation.

Effect of lnhibitors of Neutrophil Degranulation on Neutrophil- mediated Lysis of Antibody-coated Neuroblastoma Cells. Since

a correlation was observed between degranulation and ADCC in-

duced by antibody-coated human neuroblastoma cells, we determined

whether agents which inhibit degranulation would also effect ADCC.

For this purpose neutrophils were treated with 1 rnM of DIDS, ISO, or

DMX 1 h prior to the lytic assay. At this concentration DMX, ISO, and

DIDS inhibited the release of elastase by 90%, 85%, and 50%, re-

spectively (data not shown). These same inhibitors markedly de-

creased neutrophil-mediated lysis of chl4.18-coated NMB-7 cells

(Fig. 2). Thus degranulation is clearly of importance for the ability of

neutrophils to lyse antibody-coated neuroblastoma cells. Effect of GM-CSF on Degranulation of Neutrophils following

Their Exposure to Antibody-coated Neuroblastoma Cells. We

demonstrated previously that recombinant human GM-CSF enhanced

human neutrophil-mediated lysis of neuroblastoma cells coated with

monoclonal antibody ch14.18 (7). Since a correlation was observed

between the magnitude of degranulation of neutrophils and their abil-

ity to effect tumor cell lysis, it was determined whether GM-CSF

would actually enhance the level of neutrophil degranulation. The data

illustrated in Fig. 3 indicate that the same concentration of GM-CSF

(100 ng/ml) which enhanced neutrophil-mediated lysis of neuroblas- toma cells also increased the magnitude of neutrophil degranulation. The degree of lytic enhancement correlated with the concentration of antibody used. GM-CSF alone had no effect on neutrophil degranu- lation, although this cytokine increases neutrophil-mediated lysis of

antibody-coated neuroblastoma cells by enhancing the ability of these effector cells to degranulate.

Lytic Activity of Neutrophil Granular Proteins against Neuro- blastoma Cells. Since degranulation is an important step in the pro-

cess by which neutrophils lyse antibody-coated neuroblastoma cells,

an attempt was made to assess which granular protein(s) may be responsible for this lysis. Proteins from neutrophil granules were extracted and purified by techniques described in "Materials and Methods." This resulted in the isolation of 4 major protein frac- tions that were essentially the same as those reported previously by others (21, 22). The various fractions obtained from gel filtration

363

Research. on July 4, 2019. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

N E U T R O P H I L - M E D I A T E D LYSIS OF N E U R O B L A S T O M A CELLS

+l

m

C .= ol G) "O

.2

.m

O. u)

o~

0.01 0.1

Concentrat ion of ch 14.18 (~g/rnl)

5O d tti +l 40 .~_ >., - - 30 o ,=_ o

o. 20

lO

0 0 0.01 0.1 1

Concentrat ion of ch 14.18 (~g/ml)

Fig. 1. Correlation between neutrophil degranulation (a) and lysis of chl4.18-coated neuroblastoma cells mediated by these effector cells (b). Effector:target ratio for both assays was 100:1.

60

50

40

30

20

10

Untreated DIDS ISO DMX

Fig. 2. Inhibition of neutrophil-mediated lysis of ch 14.18-coated neuroblastoma cells by agents that block neutrophil degranulation. The concentration of monoclonal antibody ch 14.18 was 0.1/ag/ml. Numbers in parentheses, percentage inhibition of specific lysis of treated versus untreated groups.

chromatography of neutrophil granule extracts contained the follow- ing major proteins as demonstrated by acid-urea polyacrylamide gel electrophoresis (Fig. 4): myeloperoxidase (fraction A); cathepsin-G and elastase (fraction B); lysozyme (fraction C); and defensins (fraction D).

The proteins present in each of these major fractions as well as the unfractionated material were tested for their ability to lyse neuroblas- toma cells in a 4-h 5 ~Cr release assay. The lytic profile of the acetate- extracted neutrophil granular protein fractions (Fig. 5) indicates that cathepsin-G and elastase (fraction B) were able to lyse the NMB-7 neuroblastoma cells. Fractions A and C, containing mostly myeloper-

oxidase and lysozyme, did not lyse the NMB-7 neuroblastoma cells significantly. Purified myeloperoxidase and lysozyme from neutro- phils also failed to lyse the neuroblastoma cells (data not shown). The lytic activity of acetic acid-extracted neutrophil granular proteins re- sided essentially in the fraction containing the low-molecular-weight (Mr 3000--4000) defensins. This fraction was able to mediate the lysis of NMB-7, LAN-5, and LAN-1 human neuroblastoma cells (Fig. 6). Taken together, the data from these experiments indicate that neutro- phil granular protein fractions which contained either cathepsin-G/ elastase or defensins were able to lyse human neuroblastoma cells.

Effect of PMSF on the Ability of Proteins in Fraction B to Lyse Neuroblastoma Cells. PMSF inhibits the enzymatic activities of ca- thepsin-G and elastase (17, 25). Therefore we determined whether the lytic activity of the protein fraction that contained mainly these en- zymes (fraction B) could be inhibited with PMSE Since PMSF com- pletely prevented the material in fraction B from lysing neuroblastoma cells (Fig. 7), we concluded that the enzymatic activity of either cathepsin-G and/or elastase is required for lysis.

Ability of Purified Cathepsin-G and/or Elastase to Lyse Neu- roblastoma Cells. Since both cathepsin-G and elastase are major proteins of fraction B, it was determined whether either one or both are responsible for mediating the lysis of neuroblastoma cells. For this experiment enzymes were used which were purified from hu- man neutrophils. Purified cathepsin-G was found to lyse NMB-7 neuroblastoma cells effectively; however, purified elastase had very little if any lytic effect (Fig. 8). These data also indicate that elastase does not enhance the ability of cathepsin-G to lyse the NMB-7 neu- roblastoma cells. The ability of cathepsin-G to lyse neuroblastoma

a." 60 (/) +l r >, - - 40 .2 o

.= o~ 20

a

0.1 1

Concentration of ch 14.18 (]~g/ml)

18

d u~ is +1

�9 ~ 9 o'J

"0 u ~ 6 .g o .

~ 3

0 ........ . T . . 0 0.1 1

Concentration of ch ! 4.18 (~g/ml)

Fig. 3. GM-CSF enhancement of neutrophil-mediated lysis of chl4.18-coated neuro- blastoma cells (a) correlates with its ability to enhance neutrophil degranulation (b). Effector:target ratio for both assays was 100:1, and concentration of GM-CSF was 100 ng/ml. [], untreated neutrophils; 1~, GM-CSF treated neutrophils.

364

Research. on July 4, 2019. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

NEUTROPHIL-MEDIATED LYSIS OF NEUROBLASTOMA CELLS

MPO

ELASTASE CATHEPSIN-G

HNP-1

LYSOZYME

1 3 4 5 DISCUSSION

Fig. 4. Acid-urea polyacrylamide gel electrophoresis of acetate and acetic acid extracts of neutrophil granular proteins. Lane 1, 10 ~ag of unfractionated protein from acetate extracts of neutrophil granules; Lanes 2-5, contain 10 lag of fraction A-D, respectively. MPO, myeloperoxidase; HNP-I, defensin.

The results of this study indicate that degranulation is an important step in the process by which neutrophils lyse antibody-coated human neuroblastoma cells. This was shown both by a correlation of tumor cell lysis with neutrophil degranulation and by the inhibition of this lysis by agents known to block degranulation. Both cathepsin-G and defensins were implicated as components of neutrophil granules that are involved in this lytic process. In addition, we demonstrated that the mechanism by which GM-CSF enhances neutrophil-mediated ly- sis of antibody-coated neuroblastoma cells involves the ability of this cytokine to increase the release of lytic molecules from neutrophil granules.

A novel finding of this study is that lysis of antibody-coated neu- roblastoma cells by neutrophils can be mediated by cathepsin-G and defensins which are prepackaged in the cytoplasmic granules of these effector cells (26). We also established that the contents of neutrophil granules are important for the lysis of antibody-coated neuroblastoma cells. A direct correlation was found between the ability of neutrophils to lyse neuroblastoma cells and the amount of granular components released by these effector cells upon interaction with antibody-coated tumor cells. In addition, agents that block neutrophil degranulation were shown to inhibit the ability of neutrophils to lyse antibody- coated neuroblastoma cells. At the concentrations used in this study,

100

r 6O +t M r _>, 0

40 o o O. M

2O

None A B C UnfracUonated matedal

Fractions

Fig. 5. Lysis of NMB-7 neuroblastoma cells by acetate-extracted neutrophil granular proteins (A-C). The final concentration of each protein fraction used in the lytic assay was 50 lag/ml.

80

e~ 60

r +1

._o 40

o

2O

0 NMB-7 LAN-1 LAN-5

Fig. 6. Lysis of neuroblastoma cells by acetic acid-extracted neutrophil granular pro- teins which contained mostly defensins. The final concentrations of the proteins used in the lytic assay were 50 lag/ml ([7) and 25 lag/ml (l~l).

25

cells was not limited to NMB-7 cells, since this enzyme also lysed LAN-5 and to a smaller extent LAN-I neuroblastoma cells. The percentage of specific lysis mediated by 1 laM cathepsin-G was 34.72 + 1.07 (SD) for NMB-7; 35.39 + 1.19 for LAN-5; and 13.07 + 5.59 for LAN-1.

Effect of Purified Defensin HNP-1 on the Lysis of Neuroblas- toma Cells. Since defensins are present in the lytic fraction of the acetic acid-extracted material, we determined whether the purified defensin molecule HNP-1 is able to mediate the lysis of neuroblas- toma cells. As can be seen from the data shown in Fig. 9, HNP- 1 lysed neuroblastoma cells at concentrations ranging from 1.7 to 17 tUM (5--50 lag/ml). These concentrations are similar to those reported by others to be necessary for lysis of a variety of leukemic and lymphoma cells (9, 10). Our data suggest that in addition to cathepsin-G, de- fensins are also capable of lysing human neuroblastoma cells.

20

CO +I

w 15 u~ _>,

.2 o 10 0 Q. u)

N s

O ~ None 1 mM PMSF

Fig. 7. PMSF inhibits the lysis of neuroblastoma cells by the lytic fraction (fraction B) of the acetate-extracted neutrophil granular proteins. The final concentration of fraction B was 25 lag/ml, and that of PMSF was 1 laM.

365

Research. on July 4, 2019. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

N E U T R O P H I L - M E D I A T E D L Y S 1 S O F N E U R O B L A S T O M A C E L L S

d. u) -H a

cO

q-

O. o)

~ 1 30"

I �9 . , . ' i

0.0 0.5 1.0 1.5 2'.0 2.5

10'

Enzyme concentration (~M)

Fig. 8. Effect of cathepsin-G and elastase, purified from neutrophils, on the lysis of NMB-7 neuroblastoma cells, l l , cathepsin G-treated target cells; O, elastase-treated target cells; &, treatment with both enzymes.

50

40

30

2O

10

0 i I

0 5 10 15

Concentration of HNP-1 (j~M)

Fig. 9. Lysis of NMB-7 neuroblastoma cell by purified defensin protein HNP- 1.

i

20

these inhibitors did not affect the viability of the effector cells (data not shown). Nor have these inhibitory agents been found to interfere with the ability of neutrophils to conjugate with antibody-coated tu- mor cells (4). These same inhibitors of degranulation also do not affect the ability of neutrophils to form oxygen metabolites (27). Taken together these findings indicate that neutrophil granular proteins are required for these effector cells to lyse neuroblastoma cells.

The difference in magnitude of inhibition of lysis achieved by DMX, ISO, and DIDS may be due in part to the respective differences in their mechanisms of action. Thus whereas DMX and ISO increase intracellular levels of adenosine 3',5'-monophosphate and inhibit mi- crotubule formation (4), DIDS acts as an anion channel blocker (27). Thus it is possible that the differences observed in the ability to block lysis of neuroblastoma cells may be due in part to the mechanism by which the various inhibitors block neutrophil degranulation.

Since our study clearly demonstrates a correlation between degran- ulation and tumor cell lysis, it is conceivable that the role of the antibody molecule in neutrophil-mediated lysis of neuroblastoma cells may be to initiate a cascade of events in the effector cells, which eventually leads to the release of lytic molecules from neutrophil granules. In this regard, Kushner and Cheung (6) reported that 3G8, a monoclonal antibody directed against Fc3~III receptor, was able to

366

block the antibody-dependent lysis of neuroblastoma cells mediated by neutrophils.

Several studies indicated that GM-CSF can enhance neutrophil- mediated lysis of antibody-coated neuroblastoma cells (5-7). Here we report for the first time that exposure of neutrophils to GM-CSF increases the cytolysis of antibody-coated tumor cells by enhancing the ability of these effector cells to degranulate. These findings are in general agreement with the observation made by Kowanko et al. (28) that enhanced killing of antibody-coated fungi, Torulopsis glabrata, by neutrophils exposed to GM-CSF was due in part to the ability of the cytokine to enhance degranulation.

Defensins are well-characterized Mr 3500-4000 proteins which constitute 5-8% of the neutrophil's total protein content (29) and 30-50% of the primary granule's proteins (30). Four of these mole- cules, which have been identified in human neutrophils (HNP-1, 2, 3, and 4) were found to have similar lytic ability (31) when tested against a variety of leukemic and lymphoma cells (9, 10). Here we demon- strate for the first time that a solid tumor like human neuroblastoma is also sensitive to one of the defensins, HNP-1, at concentrations rang- ing from 1.7 to 17 ~ (5-50 lag/ml). Although the exact mechanism by which defensin HNP-1 lyses neuroblastoma cells is not known, it may be similar to the one described by others for leukemia and lymphoma cells (10, 32, 33), where defensins form dimers which bind to the target cell membrane and create small ion channels (32). The effiux of such ions then disrupts the electric potential of the target cell's mem- brane, leading to cell disruption (33).

Okrent et al. (21) reported that a fractionate from an acetate extract of neutrophils, containing mainly cathepsin-G and elastase, lysed human umbilical cord endothelial cells. However, these investigators did not identify the molecule(s) involved in this process. We could demonstrate that cathepsin-G but not elastase contained in a similar fraction obtained from an acetate extract was responsible for the lysis of neuroblastoma cells, since only purified cathepsin-G was able to achieve this lysis.

One mechanism by which the enzymatic activity of cathepsin-G may mediate the lysis of neuroblastoma cells is by activating phos- pholipase A2. In this regard phospholipase A2 was reported to convert membrane phosphatidylcholine into detergent-like lysolecithin and as such lyse tumor cells (34), including neuroblastoma cells (35). Since neuroblastoma cells express phospholipase A2 (36, 37), it is conceiv- able that its activity may increase upon exposure of the zymogen form of the enzyme to cathepsin-G, which is able to produce more active molecules.

Although the detailed mechanisms by which either defensins or cathepsin-G can lyse neuroblastoma cells are not completely known, our findings indicate that these two molecules are involved in the lytic process, although they have completely different biological functions. Thus, while defensins form pores in the membranes of their target cells (32), the lytic effect of cathepsin-G appears to be linked to its enzymatic activity.

ACKNOWLEDGMENTS

We would like to thank Dr. Tomas Ganz for his advice on the isolation of neutrophil granular proteins.

REFERENCES

I. Clark, R. A., and Klebanoff, S. J. Studies on the mechanism of antibody-dependent polymorphonuclear leukocyte-mediated cytotoxicity. J. Immunol., 119: 1413-1418, 1977.

2. Hafeman, D. G., and Lucas, Z. J. Polymorphonuclear leukocyte-mediated antibody- dependent, cellular cytotoxicity against tumor cells: dependence on oxygen and the respiratory burst. J. Immunol., 123: 55-62, 1979.

3. Katz, P., Simone, C. B., Henkart, P. A., and Fauci, A. S. Mechanisms of antibody- dependent cellular cytotoxicity: the use of effector cells from chronic granulomatous disease patients as investigative probes. J. Clin. Invest., 65: 55-63, 1980.

4. Dallegri, F., Frumento, G., Ballestrero, A., Goretti, R., and Patrone, E Relationship

Research. on July 4, 2019. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

NEUTROPHIL-MEDIATED LYSIS OF NEUROBLASTOMA CELLS

between antibody-dependent tumour cell lysis and primary granule exocytosis by human neutrophils. Clin. Exp. Immunol., 70: 479-483, 1987.

5. Bruchelt, G., Handgretinger, R., Kimmig, A., Goeke, B., Siedner, B., Reisfeld, R. A., Nietbammer, D., and Treuner, J. Effect of granulocytes on human neuroblastoma cells measured by chemiluminescence and chromium-51 release assay. J. Biolumin. Chemilumin., 3: 93-96, 1989.

6. Kushner, B. H., and Cheung, N-K. V. GM-CSF enhances 3F8 monoclonal antibody- dependent cellular cytotoxicity against human melanoma and neuroblastoma. Blood, 73: 1936-1941, 1989.

7. Barker, E., Mueiler, B. M., Handgretinger, R., Herter, M., Yu, A. L., and Reisfeld, R. A. Effect of a chimeric anti-ganglioside GD2 antibody on cell-mediated lysis of human neuroblastoma cells. Cancer Res., 51: 144-149, 1991.

8. Weiss, S. L, and LoBuglio, A. E Biology of disease: phagocyte-generated oxygen metabolites and cellular injury. Lab. Invest., 47: 5-18, 1982.

9. Lichtenstein, A., Ganz, T., Selsted, M. E., and Lehrer, R. I. In vitro tumor cell cytolysis mediated by peptide defensins of human and rabbit granulocytes. Blood, 68: 1407-1410, 1986.

10. Lichtenstein, A. K., Ganz, T., Nguyen, T-M., Selsted, M. E., and Lehrer, R. I. Mechanism of target cytolysis by peptide defensins: target cell metabolic activities, possibly involving endocytosis, are crucial for expression of cytotoxicity. J. Immu- nol., 140: 2686-2694, 1988.

11. Borish, L., Rosenbaum, R., Albury, L., and Clark, S. Activation of neutrophils by recombinant interleukin 6. Cell. Immunol., 121: 280-289, 1989.

12. Borish, L., and Rocklin, R. Effects of leukocyte inhibitory factor (LIF) on neutrophil mediated antibody-dependent cellular cytotoxicity. J. Immunol., 138: 1480-1484, 1987.

13. Shen, L., Guyre, P. M., and Fanger, M. W. Polymorphonuclear leukocyte function triggered through the high affinity Fc receptor for monomeric IgG. J. Immunol., 139: 534-538, 1987.

14. Rand, T. H., Lopez, A. E, Gamble, J. R., and Vadas, M. A. Nedocromil sodium and cromolyn (sodium cromaglycate) selectively inhibit antibody-dependent granulocyte- mediated cytotoxicity. Int. Arch. Allergy Appl. Immunol., 87: 151-158, 1988.

15. Kushner, B. H., and Cheung, N-K. V. Clinically effective monoclonal antibody 3F8 mediates nonoxidative lysis of human neuroectodermal tumor cells by polymorpho- nuclear leukocytes. Cancer Res., 51: 4865-4870, 1991.

16. Smith, R. M., and Curnutte, J. T. Molecular basis of chronic granulomatous disease. Blood, 77: 673--686, 1991.

17. Mainardi, C. L., Hasty, D. L., Seyer, J. M., and Kang, A. H. Specific cleavage of human type III collagen by human polymorphonuclear leukocyte elastase. J. Biol. Chem., 255: 12006-12010, 1980.

18. Gillies, S. D., Lo, K-M., and Wesolowski, J. High-level expression of chimeric antibodies using adapted cDNA variable region cassettes. J. Immunol. Methods, 125: 191-202, 1989.

19. Haynes, A. E, and Fletcher, J. Neutrophil function tests. Baillieres Clin. Haematol., 3: 871-887, 1990.

20. Wright, D. G. Human neutrophil degranulation. Methods Enzymol., 162:538-551, 1988.

21. Okrent, D. G., Lichtenstein, A. K., and Ganz T. Direct cytotoxicity of polymorpho- nuclear leukocyte granule proteins to human lung-derived cells and endothelial cells.

Am. Rev. Respir. Dis., 141: 179-185, 1990. 22. Ganz, T., Selsted, M. E., Szklarek, D., Harwig, S. L., Daher, K., Bainton, D. E, and

Lehrer, R. I. Defensins: natural peptide antibiotics of human neutrophils. J. Clin. Invest., 76: 1427-1435, 1985.

23. Weiss, J., Elsbach, P., Olsson, I., and Oderberg, H. Purification and characterization of a potent bactericidal and membrane active protein from the granules of human polymorphonuclear leukocytes. J. Biol. Chem., 253: 2664-2672, 1978.

24. Boros, P., Odin, J. A., Muryoi, T., Masur, S. K., Bona, C., and Unkeless, J. C. IgM Anti-Fc3,R autoantibodies trigger neutrophil degranulation. J. Exp. Med., 173: 1473- 1482, 1991.

25. Starkey, P. M., and Barrett, A. J. Human cathepsin G: catalytic and immunological properties. Biochem. J., 155: 273-278, 1976.

26. Wright, D. W. The neutrophil as a secretory organ of host defense. In: J. I. Gallin and A. S. Fauci (eds.), Advances in Host Defense Mechanisms, Vol. 1, pp. 75-95. New York: Raven Press, 1982.

27. Korchak, H. M., Eisenstat, B. A., Hoffstein, S. T., Dunham, B. P., and Weissmann, G. Anion channel blockers inhibit lysosomal enzyme secretion from human neutrophils without affecting generation of superoxide anion. Proc. Natl. Acad. Sci. USA, 77: 2721-2725, 1980.

28. Kowanko, I. C., Ferrante, A., Harvey, D. P., and Carman, K. L. Granulocyte-mac- rophage colony-stimulating factor augments neutrophil killing of Torulopsis glabrata and stimulates neutrophil respiratory burst and degranulation. Clin. Exp. Immunol., 83: 225-230, 1991.

29. Spitznagel, J. K. Antibiotic proteins of human neutrophils. J. Clin. Invest., 86: 1381-1386, 1990.

30. Lehrer, R. I., and Ganz, T. Antimicrobial polypeptides of human neutrophils. Blood, 76: 2169-2181, 1990.

31. Lehrer, R. I., Ganz, T., and Selsted, M. E. Defensins: endogenous antibiotic peptides of animal cells. Cell, 64: 229-230, 1991.

32. Hill, C. P., Yee, J., Selsted, M. E., and Eisenberg, D. Crystal structure of defensin HNP-3, an amphiphilic dimer: mechanisms of membrane permeabilization. Science (Washington DC), 251: 1481-1485, 1991.

33. Lichtenstein, A. Mechanism of mammalian cell lysis mediated by peptide defensins: evidence for an initial alteration of the plasma membrane. J. Clin. Invest., 88: 93-100, 1991.

34. Frye, L. D., and Friou, G. J. Inhibition of mammalian cytotoxic cells by phosphati- dylcholine and its analogue. Nature (Lond.), 258: 333-335, 1975.

35. Chwetzoff, S., Tsunasawa, S., Sakiyama, E, and Menez, A. Nigexine, a phospholipase A2 from cobra venom with cytotoxic properties not related to esterase activity: purification, amino acid sequence, and biological properties. J. Biol. Chem., 264: 13289-13297, 1989.

36. Morash, S. C., Cook, H. W., and Spence, M. W. Lysophosphatidylcholine as an intermediate in phosphatidylcholine metabolism and glycerophosphocholine synthe- sis in cultured cells: an evaluation of the role of 1-acyl- and 2-acyl-tysophosphati- dylcholine. Biochim. Biophys. Acta, 1004: 221-229, 1989.

37. Hirashima, Y., Mills, J. S., Yates, A. J., and Horrocks, L. A. Phospholipase A2 activities with a plasmalogen substrate in brain and in neural tumor cells: a sensitive and specific assay using pyrenesulfonyl-labeled plasmenylethanolamine. Biochim. Biophys. Acta, 1047: 35-40, 1990.

367

Research. on July 4, 2019. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

1993;53:362-367. Cancer Res   Edward Barker and Ralph A. Reisfeld  Neuroblastoma CellsA Mechanism for Neutrophil-mediated Lysis of Human

  Updated version

  http://cancerres.aacrjournals.org/content/53/2/362

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/53/2/362To request permission to re-use all or part of this article, use this link

Research. on July 4, 2019. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from