HYBRIDOMA Volume 11, Number 2, 1992 Mary Ann Lieben, Inc., Publishers
Antibody
AP-282 Recognizes Marker for Human Polymorphonuclear Granules
Monoclonal
ANNE
MARTIN,1
BRIGITTE COLLET,1 CATHERINE BERNARD DRENOU,1 and L. TOUJAS1
a
PICOT,2
'Centre Régional de Lutte contre le Cancer, 35033 Rennes-Cedex, France ^Centre Régional de Transfusion Sanguine, Rennes, France
ABSTRACT
Hybridoma AP-282 was produced by fusing mouse plasmacytoma cells with splenocytes of mice immunized against purified human polymorphonuclear cells. The secreted monoclonal antibody (MAb), AP-282, a mouse IgGl, was found to react strongly with all neutrophilic granulocytes, their bone marrow precursors, weakly with blood monocytes and not with eosinophils. The anti¬ gen was resistant to formalin fixation but was destroyed by exposure to fixa¬ tives containing acetic acid. Using the APAAP technique, antibody AP-282 strongly labelled neutrophils on sections of frozen cut or paraffin embedded tissues. No staining was seen of non hematopoietic tissues. AP-282 recognized an internal antigen associated to cytoplasmic granules. Chemical investigations on dot blots of whole or of purified cellular extracts indicated that the antigen idenfied by MAb AP-282 was different from those recognized by usual antigranulocyte antibodies, i.e. myeloperoxidase, elastase, cathepsin G and lacto¬ ferrin. Thus, antibody AP-282 constitutes a new cytoplasmic marker of neu¬ trophils.
INTRODUCTION antibodies directed against human granulocytes have Numerous monoclonal been prepared. They react with antigens that may be expressed on cell surface or detected only in the cytoplasm. Internal antigens seem generally associated with products of cytoplasmic granules. The latter are of two main types: azurophil or primary granules which correspond to primary lysosomes and specific or secondary granules which are secondary lysosomes. Myeloperoxidase, cathepsin G, and elastase are the most characteristic components of azurophil granules whereas the presence of lactoferrin is often taken as a feature of specific granules (1-3). MAbs that react with elastase 135
G also called chymotrypsin-like canonie been prepared. The MAb described in the present work, AP-282, strongly labelled neutrophils in blood smears and paraffin embedded tissues. It recognizes a novel component of cytoplasmic granules.
(4), myeloperoxidase (5,6), cathepsin protein (7,8) and lactoferrin (8) have
MATERIALS AND METHODS
Production, Purification, Radiolabelling and Isotyping of MAb AP282
Balb/c mice were immunized by 3 subcutaneous inoculations, at one month intervals, of 5x1 fj7 granulocytes from healthy human donor. An intravenous
given 3 days before performing the fusion. The general proce¬ production and purification has been described previously immunized splenocytes were fused with SP2/0 plasmacytoma (9,10). Briefly, cells and the hybrids were screened by an indirect radioimmunological test lymphocytes from a chronic against human granulocytes and against both cell taken as leukemia controls, types being fixed with 0.25% lymphoid glutaraldehyde. After cloning by the limiting dilution technique, 3x10^ cells of the hybrid selected AP-282 were injected intra-peritoneally into Balb/c mice pretreated with Pristan oil (Aldrich, Strasbourg). The antibody was purified from ascitic fluid by ion exchange chromatography with a Mono-Q 10/10 column (FPLC, Pharmacia, Bois d'Arcy, France) and its concentration determined by spectrometry at 282 nm. Purified antibody was radiolabelled with Iodine-125 (S4 preparation, Office des Rayonnements Ionisants, Gif sur Yvette, France) by the Iodogen (Tetrachloro diphenyl glycouryl, Sigma, La Verpillière, France) technique and diluted in PBS containing 0.5% bovine serum albumin (PBS-BSA). The radioiodinated solution of MAb AP-282 contained 3.2 xlO^ cpm/ml and its binding to PMN was checked to compete with that of unlabelled AP-282. Antibody isotype in cloned hybrid by an ELISA technique using the MAb supernatants was determined kit A (Pharmingen). isotyping
injection
was
dure for MAb
Cells and Tissues. Granulocytes were obtained by centrifuging blood from healthy donors on a cushion of Monopoly resolving medium (Flow, Les Ulis, France). Blood mono¬ nuclear cells were isolated by centrifugation on a ficoll-metrizoate gradient (MSL separation medium, Eurobio, Paris). Blood derived macrophages were of the mononuclear cell fraction on Sepracell medium obtained by separation and by culturing purified monocytes for 6 days in Chexall (Clinisciences, Paris) II Teflon tissue culture bags (J-Bio, Les Ulis, France) with RPMI 1640 medium supplemented with 5% non heat-inactivated autologous serum. Cells of the monocyte-macrophage series were characterized either by their ability to phagocytize 0.81µ3 latex particles (Difco) during a 30 minute incubation at 37°C or by their reactivity with phycoerythrin-labelled anti-CD14 MAb (IOM2 Immunotech). Bone marrow smears were obtained from patients and the various tissues mentioned in "Results" were sampled during surgical opera¬ tions. The histiocytic cell line U937 was maintained in RPMI 1640 supple-
136
mented with
10% fetal bovine serum. Several fixatives were used. Cell suspensions were fixed, in one of the following, 0.25% glutaraldehyde, 70% ethanol, 70% ethanol/5% acetic acid or 4% solution of paraformaldehyde in Sòrensen buffer, pH 7.3. Solid tissues were fixed in one of the following, 10% formalin solution in 0.15M NaCl, Dubosq fixative (ethanol 60% v/v, formalin 10%, acetic acid 7% and picric acid 4g/l), or Bouin fixative. The tissues were, then, dehydrated in acetone, toluol and embedded in paraffin. For frozen sections, tissues were dipped in isopentane maintained at liquid nitrogen temperature, embedded in O.C.T. (Reichert-Jung, Paris) and cut into sections with a Cryocut microtome (Reichert-Jung). The sections, air dried on glass slides were fixed in acetone for 3 minutes.
Immunocytological and Histological Tests. Indirect radioimmunological tests were performed on cell suspensions as des¬ cribed previously for hybridoma supernatant screening (11). Briefly, the cells were incubated with the supernatants studied, washed and exposed to sheep anti-mouse Fab'2 radiolabelled with iodine-125 (Amersham Les Ulis, France). The radioactivity of the cell pellet was measured and compared with controls in which incubation with hybridoma supernatant had been omitted. A supernatant of hybrid W6/32 culture which contained a polymorphic antiHLA class I MAb, was used as positive control. Po66, an MAb of IgGl isotype previously studied in the laboratory (10,11) was used as a negative control. Indirect immunofluorescence tests on cell suspensions were also carried out, using a fluorescein-labelled sheep anti-mouse Fab'2 as second antibody layer (New England Nuclears, Dupont de Nemours France, Paris), (9). The results were analysed by flow cytometry (Ortho H 50 system). Cell light scattering properties at right angle (RAS) or forward angle (FAS) were utilized for the selection of unfixed monocyte, lymphocyte and granulocyte subpopulations. The results were expressed as percentages of positive cells by reference to control cells incubated with fluorescent Fab'2 alone and by the mean fluorescence intensity given in arbitrary units. Cell smears and tissue sections were stained by the APAAP (Alkaline Phos¬ phatase Anti-Alkaline Phosphatase) technique according to Cordell et al. (12). Briefly, sections were rehydrated and protein binding sites saturated with PBS-BSA. The slides were, then, exposed successively to the monoclonal antibody studied, rabbit anti-mouse immunoglobulin (Dako-Sebia, Issy les Moulineaux, France) and the APAAP complex (Dako). Alkaline phosphatase activity was revealed by incubation with a substrate containing fuchsin and the sections were counterstained with Harris haematoxylin. The ABC tech¬ nique of immunoperoxidase staining (13) was also used (Vectastain kit). APAAP and May-Grünwald-Giemsa (MGG) stainings were combined, according to Pryzwansky et al. (14). Briefly, bone marrow smears were stained with MGG and selected areas were photographed. After destaining, the slides were stained by the APAAP technique and the same areas photographed. Studies
Antigens from Cytosolic Preparations. Cytosols were prepared from 10& mononuclear cells and 1.5x10^ granulo¬ cytes. The cells were lysed by sonication and centrifuged at 10,000 g for 15 minutes. The supernatant was adjusted at 6mg/ml proteins, aliquoted and on
stored at -20°C.
137
tests, 2 µ samples of various dilutions of cytosol were CM membranes (Amersham, Les Ulis, France) The membranes were saturated for 4 hours in PBS-3% BSA and incubated with one of the following, sodium periodate, papain (Sigma), pepsin (Sigma), pronase (Calbiochem, France Biochem), trypsin-EDTA (Eurobio), collagenase II (Sigma) or DNAse I (Sigma), RNAse A (Sigma) or neuraminidase from Vibrio cholerae (Sigma). Digestion was carried out at 37°C for 2 hours in PBS-BSA buffer. After digestion, the membranes were washed and incubated for 4 hours with 1:30 diluted radiolabelled AP-282. The membranes were then washed 6 times with PBS-BSA and processed for autoradiography The reactivity of AP-282 with purified lysosomal enzyme extracts was carried out as follows. One hundred mg/ml aqueous solutions of myeloperoxidase from human leukocytes (Sigma), lactoferrin from human milk (Sigma), ca¬ thepsin G from human leukocytes (Sigma) and elastase from granulocytes kindly provided by Dr D.Y. Mason (Oxford) were deposited on Hybond CM membranes. After saturation in a PBS-BSA solution, the membranes were ex¬ posed to 1 mg/ml AP-282 solution or to various control antibodies: polyclonal rabbit anti-lactoferrin antibody (Dako), mouse anti-myeloperoxidase MP07 and anti-elastase NP57 MAbs, both generous gifts from Dr D.Y. Mason. After washing in PBS, the membranes were exposed to a rabbit antimouse antibody (Dako) for 30 minutes and to the APAAP complex (Dako). Alkaline phospha¬ tase was revealed by nitrobluetetrazolium (Sigma) and bromochlorylindolyl phosphate (Sigma) as described by McGadey (15).
enzymatic digestion deposited on Hybond For
RESULTS
Production of MAb AP-282. A cell fusion between SP2/0 plasmacytoma cells and mouse splenocytes immunized against purified human blood PMN gave rise to 93 hybrids, of which 13 reacted lymphocytes in a against human granulocytes but not against leukemic radioimmunological assay. The monoclonal antibody AP-282 was selected because it
strongly
neutrophils but not lymphocytes (Table 1 ). antibody AP-282 was found of IgGl kappa isotype.
labelled
Monoclonal
Table 1 : Reactivity of Mab AP282 with granulocytes and leukemic determined by an indirect radioimmunological assay.
B. LYMPHOCYTES
ANTIBODY
GRANULOCYTES
AP282
16037
226
5435
9804
506
186
Anti-HLA
None1
Results expressed in cpm bound to 10.E5 cells. anti-mouse Fab'2 alone.
138
lymphocytes as
(l):
cells incubated with radiolabelled
Reactivity of MAb AP-282 with Circulating Blood Cells. The reactivity of MAb AP-282 with blood PMN purified by centrifugation on a density gradient was investigated using an indirect immunofluorescence test and was analysed both microscopically and by flow cytometry. As can be
in seen Figure 1, unfixed cells did not react with AP-282 although they bound the anti-HLA MAb. By contrast, cells fixed with 70% ethanol were Concentrations stained. of of 1 strongly purified antibody µg/ml proved suf¬ ficient to detect the reaction. Various fixatives were tested. No labelling with AP-282 was seen after formalin fixation. This fixative caused poor permeabilization, as judged by trypan blue penetration and did not allow AP282 staining. The use of ethanol-acetic acid also gave poor results, probably because its acidic pH destroyed the antigen, as seen below.
II
I
i*"i«-"—"it¡-
L
TS
'
15
'
ÎT
*a
—
'
k
'
di
im
J
íaíi
75
îw
73
Í9»
~aíF
rss
" "
'
. SS
Ï5~
Figure 1. Reactivity of MAbs AP-282 and W6/32 with granulocytes, as measured by immunofluorescence and cell flow analysis. I : Unfixed granulocytes, II : Ethanol fixed granulocytes. A : AP-282 5 µg/ml, : AP-282 1 µg/ml, C : W6/32 hybridoma supernatant. -
Staining of blood smears fixed with ethanol by the APAAP technique, showed that MAb AP-282 reacted with 95% neutrophilic PMN, a result in agreement with flow cytometric measurements. Antibody concentrations as low as 0.1 mg/ml gave positive results (Figure 2). Eosinophils were unlabelled by AP282 Blood monocytes were clearly less reactive than PMN. When identified their by ability to phagocytize latex particles, only a low percentage stained positively. Macrophages derived from one week cultures of blood mononu¬ clear cells appeared as large cells exhibiting CD-14 antigen. About 10% of them had cytoplasmic granules intensively stained by AP-282. Similar results found with the macrophages of two ascitic malignant metastasis. The were histiocytic U937 cell line also bound the antibody. .
139
Figure 2. Left: APAAP staining of a smear of purified granulocytes with 0.1 AP-282. (G 1100). Right: control without exposure to MAb AP-282.
µg/ml
Tissue distribution of the antigen recognized by AP-282. The antibody was tested with bone marrow smears from patients. A non bone marrow malignant sample containing 15% promyelocytes was submitted to a double staining technique (14). Promyelocytes were unstained or weakly stained by AP-282, whereas myelocytes and more mature cells of the neutrophilic series were positive.
The distribution of the antigen recognized by Mab AP-282 among solid organs was studied on Cryostat tissue sections using the APAAP technique. The anti¬ body reacted with isolated cells present in blood vessels or forming inflam¬ matory clusters. Positive cells exhibited frequently a multilobulated nucleus characteristic of granulocytes. The reactivity of MAb AP-282 was investigated on sections of various paraffin-
embedded tissues: skin, gastric and colonie mucosa, lung, liver, kidney, thyroid gland, cardiac muscle, spleen, lymph node, acute abcesses and granulomas. The results clearly depended on the technique of tissue fixation. Blocks fixed with Bouin or Dubosq acidic fixatives did not show any positivity whereas, in tissues fixed with 10% formalin in saline (pH 7), isolated positive cells were clearly detected with concentrations of antibody of 1 µg/ml (Figure 3) or less. The reactive cells often showed the characteristic PMN nucleus. Some tissue macrophages also stained but Kupffer cells and alveolar macrophages were unlabelled. All other cells were
negative.
Cellular localization of the antigen identified by AP-282. In most instances, when cells were fixed with ethanol and stained as suspen¬ sions by immuno-fluorescence or on smears with APAAP or immunoperoxidase, the diffuse cytoplasmic staining was seen with AP-282. Sometimes nu¬ clear staining was seen. However, when cell suspensions of PMN, macro¬ U937 line or fixed with were phages paraformaldehyde, the antigen visualized the APAAP the or by immunoperoxidase technique appeared cytoplasmic and associated with granulations (Figure 2).
140
fîf*
Figure appear
H
Vf
·
'
3. APAAP staining of paraffin serial sections of thyroid gland. Granulocytes darkly stained cells (left) absent on the control (right) (G 230).
as
Biochemical characterization of antigen. A comparison of binding of radioiodinated MAb AP-282 with cytosolic prepa¬ rations from purified blood PMN and from purified blood mononuclear cells Radioiodinated AP-282 was incubated with cytosols deposi¬ was performed. ted on methylcellulose membranes. The antibody reacted much more strongly with the cytosol from PMN than with that of mononuclear cells, as also allowed to revealed by autoradiography. The dot blot technique determine the susceptibility of the antigen to various treatments. The reactivity of AP282 with PMN cytosol was altered essentially by low pH pretreatment and by pronase digestion. The alteration observed after pepsin digestion were presumably related to the low pH of the reaction buffer (Table
2). Table 2 : Dot blot reaction of PMN various enzymatic digestions.
Enzyme
cytosol
with Mab AP-282 after submission to
pH
Concentration
AP282
binding
+++ 7.2 2.8 2.8 250 u/ml Pepsine +++ 50 u/ml 6.0 Papaïne + Pronase 100 u/ml 7.2 +++ 7.2 Trypsine/EDTA2 DNase +++ 5.0 1000 u/ml +++ RNase 7.2 250 u/ml 4-++ 5.0 Neuraminidase 5 u/ml +++ 7.2 20 mM Periodate from autoradiography. (1): arbitrary quotation of AP-282 reactivity (2): Trypsin EDTA solution for cell culture detachment.
PBS Acidic buffer
141
1
ANTIBODIES
ANTIGENS
AP-282
Anti-myeloperoxic G
Ca
LM
My
________________
.; O
-
G
LM
ö
-
*
G
El
My -
LM
My
rt
El
Anti-elastase
La
-
*
AP-282
AP-282
Anti-lactoferrin
La
Figure 6: Reactivity of antigens deposited on nitrocellulose membranes with AP-282 and other antibodies. The antigens deposited were cytosolic preparations from granulocytes (G) and mononuclear cells (lymphocytes and monocytes: LM), purified Cathepsin-G (Ca), myeloperoxidase (My), elastase (El) and lactoferrin (La). The antibodies tested, in addition to MAb AP-282, were directed against myeloperoxidase, elastase and lactoferrin. Some purified substances present in the granules of granulocytes were allowed to react with AP-282 and other antibodies taken as controls. As in Figure 6, AP-282 did not bind to any of the substances tested: shown myeloperoxidase, elastase, cathepsine G or lactoferrin, although it reacted with PMN extract and the substances deposited were recognized by the corres¬ ponding antibodies. Anti-lactoferrin MAb was found incidently to react with the myeloperoxidase extract. The molecular weight of the AP-282 antigen could not be determined by im¬ munoprecipitation or Western blotting.
DISCUSSION was found to bind predominantly to PMN and their bone mar¬ precursors. Only weak labelling of monocytes and macrophages was seen. AP-282 did not stain any non hematopoietic cells. Since antibody AP-282 reacted with more than 95% blood PMN and stained these cells intensely at low concentrations on paraffin sections, AP-282 appears to be a reliable mar¬ ker for the identification of granulocytes on tissue sections or on blood and
MAb AP-282
row
142
bone marrow smears. The only reservation is that cells and tissues should be fixed adequately in non acidic fixatives. The lack of detection of the antigen after Dubosq fixative, indeed, was likely related to the acetic acid content of the reactant, which lowered the pH to 3.5. This hypothesis was sustained by dot blot investigations showing that the antigen was degradated by acidic pH. The site of localization of the antigen recognized by AP-282 was not the cell membrane, as suggested by the non reactivity of unfixed cells with the anti¬ body. The precise cytoplasmic localization raised some difficulties since it va¬ ried according to the fixation procedure. Diffusion phenomena of staining when studying antigens from PMN gra¬ seems to be frequently encountered nules. Results similar to those reported here have been found with the antielastase monoclonal antibody NP57 (4). Some fixatives particularly paraformaldehyde seemed more reliable than others and, in our study, showed the association of AP-282-identified antigen with cytoplasmic granules, in granu¬ locytes, monocytes, macrophages and U937 cells. The chemical nature of the antigen recognized by AP-282 has not been de¬ termined. A number of components have been now characterized in neutro¬ phil granules. In primary granules, myeloperoxidase, low molecular weight defensins (16) and several proteinases such as cathepsin G, elastase and rela¬ ted molecules like AGP7, azurocidin, p29b (17,18), proteinase 3 (19) and the antigen involved in Wegener's disease (20) have been identified. Lactoferrin, collagenase and gelatinase (21) were localized to secondary granules. The amino acid sequence of severü of these proteins has been recently establi¬ shed: myeloperoxidase (22), :Lthepsin G (8), antibiotic proteins (23) and We¬ gener's granulomatosis antiger. (20). Numerous aminoacid sequence homologies were found between the various compounds. For example, cathepsin G, elastase, proteinase-3, azurocydinô, AGP7 shared a high degree of similarity in their NH2-terminal amino acid sequence (18). This probably explains why MAbs directed against PMN may show a certain degree of cross reactivity when tested against different purified components of the granules (7). In this regard, it seems of interest that the epitope recognized by AP-282 was not ex¬ pressed on the more usual cytoplasmic markers of human granulocyte, elas¬ tase, myeloperoxidase, cathepsin G and lactoferrin although the specific anti¬ bodies taken as controls reacted positively. The reactivity of the polyclonal rabbit anti-lactoferrin antibody with the myeloperoxidase extract was observed. This might be related to an incomplete purification of one of the reactants. The weak or non expression of the antigen recognized by MAb AP282 on promyelocytes, in contrast with its presence on myelocytes and their descendants, suggests that AP-282 is a marker of secondary granules (14). Further investigations will be necessary to confirm this result, in particular comparisons of the localization of AP-282 with other known markers of secondary granules, like lactoferrin. This could be achieved with a panel of bone marrow samples by conventional techniques and double staining experiments in electron microscopy, as reported in (1-3). Antibody AP-282 may, thus, appear as a new reagent that can characterize granulocytes histologically and as a valuable tool for cytological studies de¬ signed for example to classify myeloid leukemias. Investigations to characterize the molecule(s) recognized by AP-282 are now required and might allow the identification of new granulocyte component(s).
143
ACKNOWLEDGMENTS: The authors wish to thank Drs D. Mason and K. Pulford for their kind assist¬ ance and skilled advice. Work supported by Comité d'Ille et Vilaine de la Ligue Nationale Française .
REFERENCES
1. Calafat, J., Goldschmeding, R., Ringeling, P.L., Janssen, H. and Van der Schoot, CE. (1990) In situ localization by double labeling immuno electron microscopy of anti-neutrophil cytoplasmic auto antibodies in neutrophils and monocytes. Blood. 75: 242-250.
2. Cramer, E.M., Beesley, J.E., Pulford, K.A., Breton-Gorius, J. and Mason, D.Y. (1989) Co-localization of elastase and myeloperoxidase in human blood and bone marrow neutrophils using a monoclonal antibody and immunogold. Am. J. Pathol. 134, 1275-1284. 3. Esaguy, ., Aguas, A.P. and lactoferrin in human neutrophils neity. J. Leuk. Biol. 46,51-62.
Silva, M.T. (1989) High resolution localization of :
labeling
of
secondary granules
and cell
heteroge¬
4.Pulford, K.A.F., Erber W.N., Crick, J.A., I Olsson, I., Micklem, K.J. Gatter, K.C., and Mason, D.Y. (1988) Use of monoclonal antibodies against human neutro¬ phil elastase in normal and leukaemic myeloid cells. J. Clin. Pathol. 41, 853-860. ,
5.Homma, T., Suzuki, K., Kudo, Y., Inagawa, M., Mizuno, S., Yamaguchi, K. and M. (1989) Preparation and characterization of monoclonal antibodies human myeloperoxidase. Arch. Biochem. Biophys. 273, 189-196.
Tagawa,
against
6. Zaki, S.R., Chan,W.C, McKolanis, J. and Austin, G.E. (1989) Production and characterization of monoclonal antibodies to human myeloperoxidase. Clin.ImmunolImmunopathol. 50, 283-297. 7. Skubitz, K.M., Christiansen, N.P. and Mendiola, J.R. (1989) Preparation and cha¬ racterization of monoclonal antibodies to human neutrophil cathepsin G, lactoferrin, eosinophil peroxidase, and eosinophil major basic protein. J. Leukoc. Biol. 46, 109118.
8. Hohn, P.A., Popescu, N.C., Hanson, R.D., Salvesen, G. and Ley, T.J. (1989) Genomic organization and chromosomal localization of the human cathepsin G gene. J. Biol. Chem. 264, 13412-13419. 9. Martin, ., Lecorre, R., Pellen, P., Bourel, D., Merdrignac, G, Genetet, B,.and Toujas, L. (1986) A monoclonal antibody reacting with a mouse specific epitope of Macl antigen. Tissue antigens. 28, 14-23. .
10. Martin, ., Pellen, P., Youinou, P., Collet, B., Desrues, .,Bourel, D., Dazord,
144
L. and Toujas, L. (1989) Characterization of the antigen identified by P066, a mo¬ noclonal antibody raised against a lung squamous cell carcinoma. Cancer Immunol. Immunother. 29, 118-124.
11. Dazord, L., Bourel, D., Martin, ., Lecorre, R., Bourguet, P., Bohy, J., Saccavini, J.C., Délavai, P., Louvet, M. and Toujas, L. (1987) A monoclonal antibody (Po66) directed against human lung squamous cell carcinoma. Immunolocalization of tumour xenografts in nude mice. Cancer Immunol. Immunother, 24, 263-268.
12. Cordell, J.L., Falini, B., Erber W.N., Ghosh, A.K., Abdulaziz, Z., Macdonald, S., Pulford, K.A.F., Stein, F. and Mason, D.Y. (1984) Immunoenzymatic labeling of
monoclonal antibodies using immune complexes of alkaline phosphatase and mono¬ clonal anti-alkaline phosphatase (APAAP complexes). J. Histochem. Cytochem. 32, 219-229.
13. Hsu, S., Raine, I. and Fänger, H. (1981) The use of avidin-biotin peroxidase complex (ABC) in immunoperoxidase technique. J. Histochem., 29, 577-581. 14.
Pryzwansky, K.B., Rausch, P.G., Spitznagel, J.K. and Herion, J.C. (1979) Immunocytochemical distinction between primary and secondary granule formation in developping human neutrophils: correlation with Romanansky stains. Blood, 53,179-185.
15. McGadey, J. (1970) A tetrazolium method for Histochem. 23, 180-184.
non
specific alkaline phosphatase.
16. Lehrer, R.I., Barton, ., Daher, . ., Harwig, S.S., Ganz, T.and Selsted, .E. (1988) Interaction of human defensins with Escherichia coli. Mechanism of bacteri¬ cidal activity. J Clin Invest. 84,553-561. 17. Campanelli, D., Detmers, P.A., Nathan, CF. and Gabay, J.E. (1990) Azurocidin and a homologous serine protease from neutrophils. Differential antimicrobial and proteolytic properties. J. Clin. Invest. 85, 904-915. 18. Wilde, CG, Snable J.L., Griffith, J.E. and Scott, R.W. (1990) Characterization of two azurophil granule proteases with active-site homology to neutrophil elastase. J. Biol. Chem. 265, 2038-2041. 19. Kao, R.C., Wehner, N.G., Skubitz, K.M., Gray. B.H, and Hoidal JR. (1988) Proteinase 3: a distinct polymorphonuclear leukocyte protease which produced em¬ physema. J. Clin. Invest. 82, 1963-1973. 20. Lüdemann, J., Utecht, B. and Gross, W.L. (1990) Anti-neutrophil cytoplasm an¬ tibodies in Wegener's granulomatosis recognize an elastinolytic enzyme. J. Exp. Med.
171,357-362.
21. Hibbs, M.S. and Bainton, D.F. (1989) Human neutrophil ge latinase is nent of specific granules. J. Clin. Invest. 84, 1395-1402. 22.
a
compo¬
Wright, J., Bastian, ., Davis, T.A., Zuo, C, Yoshimoto, S., Orme-Johnson, 145
W.
A. Structural characterization of the isoenzymatic forms of hu¬ myeloperoxidase, evaluation of the iron containing prosthetic group. Blood, 75,
and Tauber man
(1990)
238-241.
Gabay, JE., Scott, RW„ Campanelli, D., Griffith, J., Wilde C, Marra M.M., Seeger, M. and Nathan, CF. (1989) Antibiotic proteins of human polymorphonu¬ clear leukocytes. Proc. Nati. Acad. Sci. 86, 5610-5615. 23.
Address reprint requests to: A. MARTIN Laboratoire d'Immunologie-Immunothérapie Centre Eugène Marquis 35033 RENNES CEDEX (France) -
146
Antibody
AP-282 Recognizes Marker for Human Polymorphonuclear Granules
Monoclonal
ANNE
MARTIN,1
BRIGITTE COLLET,1 CATHERINE BERNARD DRENOU,1 and L. TOUJAS1
a
PICOT,2
'Centre Régional de Lutte contre le Cancer, 35033 Rennes-Cedex, France ^Centre Régional de Transfusion Sanguine, Rennes, France
ABSTRACT
Hybridoma AP-282 was produced by fusing mouse plasmacytoma cells with splenocytes of mice immunized against purified human polymorphonuclear cells. The secreted monoclonal antibody (MAb), AP-282, a mouse IgGl, was found to react strongly with all neutrophilic granulocytes, their bone marrow precursors, weakly with blood monocytes and not with eosinophils. The anti¬ gen was resistant to formalin fixation but was destroyed by exposure to fixa¬ tives containing acetic acid. Using the APAAP technique, antibody AP-282 strongly labelled neutrophils on sections of frozen cut or paraffin embedded tissues. No staining was seen of non hematopoietic tissues. AP-282 recognized an internal antigen associated to cytoplasmic granules. Chemical investigations on dot blots of whole or of purified cellular extracts indicated that the antigen idenfied by MAb AP-282 was different from those recognized by usual antigranulocyte antibodies, i.e. myeloperoxidase, elastase, cathepsin G and lacto¬ ferrin. Thus, antibody AP-282 constitutes a new cytoplasmic marker of neu¬ trophils.
INTRODUCTION antibodies directed against human granulocytes have Numerous monoclonal been prepared. They react with antigens that may be expressed on cell surface or detected only in the cytoplasm. Internal antigens seem generally associated with products of cytoplasmic granules. The latter are of two main types: azurophil or primary granules which correspond to primary lysosomes and specific or secondary granules which are secondary lysosomes. Myeloperoxidase, cathepsin G, and elastase are the most characteristic components of azurophil granules whereas the presence of lactoferrin is often taken as a feature of specific granules (1-3). MAbs that react with elastase 135
G also called chymotrypsin-like canonie been prepared. The MAb described in the present work, AP-282, strongly labelled neutrophils in blood smears and paraffin embedded tissues. It recognizes a novel component of cytoplasmic granules.
(4), myeloperoxidase (5,6), cathepsin protein (7,8) and lactoferrin (8) have
MATERIALS AND METHODS
Production, Purification, Radiolabelling and Isotyping of MAb AP282
Balb/c mice were immunized by 3 subcutaneous inoculations, at one month intervals, of 5x1 fj7 granulocytes from healthy human donor. An intravenous
given 3 days before performing the fusion. The general proce¬ production and purification has been described previously immunized splenocytes were fused with SP2/0 plasmacytoma (9,10). Briefly, cells and the hybrids were screened by an indirect radioimmunological test lymphocytes from a chronic against human granulocytes and against both cell taken as leukemia controls, types being fixed with 0.25% lymphoid glutaraldehyde. After cloning by the limiting dilution technique, 3x10^ cells of the hybrid selected AP-282 were injected intra-peritoneally into Balb/c mice pretreated with Pristan oil (Aldrich, Strasbourg). The antibody was purified from ascitic fluid by ion exchange chromatography with a Mono-Q 10/10 column (FPLC, Pharmacia, Bois d'Arcy, France) and its concentration determined by spectrometry at 282 nm. Purified antibody was radiolabelled with Iodine-125 (S4 preparation, Office des Rayonnements Ionisants, Gif sur Yvette, France) by the Iodogen (Tetrachloro diphenyl glycouryl, Sigma, La Verpillière, France) technique and diluted in PBS containing 0.5% bovine serum albumin (PBS-BSA). The radioiodinated solution of MAb AP-282 contained 3.2 xlO^ cpm/ml and its binding to PMN was checked to compete with that of unlabelled AP-282. Antibody isotype in cloned hybrid by an ELISA technique using the MAb supernatants was determined kit A (Pharmingen). isotyping
injection
was
dure for MAb
Cells and Tissues. Granulocytes were obtained by centrifuging blood from healthy donors on a cushion of Monopoly resolving medium (Flow, Les Ulis, France). Blood mono¬ nuclear cells were isolated by centrifugation on a ficoll-metrizoate gradient (MSL separation medium, Eurobio, Paris). Blood derived macrophages were of the mononuclear cell fraction on Sepracell medium obtained by separation and by culturing purified monocytes for 6 days in Chexall (Clinisciences, Paris) II Teflon tissue culture bags (J-Bio, Les Ulis, France) with RPMI 1640 medium supplemented with 5% non heat-inactivated autologous serum. Cells of the monocyte-macrophage series were characterized either by their ability to phagocytize 0.81µ3 latex particles (Difco) during a 30 minute incubation at 37°C or by their reactivity with phycoerythrin-labelled anti-CD14 MAb (IOM2 Immunotech). Bone marrow smears were obtained from patients and the various tissues mentioned in "Results" were sampled during surgical opera¬ tions. The histiocytic cell line U937 was maintained in RPMI 1640 supple-
136
mented with
10% fetal bovine serum. Several fixatives were used. Cell suspensions were fixed, in one of the following, 0.25% glutaraldehyde, 70% ethanol, 70% ethanol/5% acetic acid or 4% solution of paraformaldehyde in Sòrensen buffer, pH 7.3. Solid tissues were fixed in one of the following, 10% formalin solution in 0.15M NaCl, Dubosq fixative (ethanol 60% v/v, formalin 10%, acetic acid 7% and picric acid 4g/l), or Bouin fixative. The tissues were, then, dehydrated in acetone, toluol and embedded in paraffin. For frozen sections, tissues were dipped in isopentane maintained at liquid nitrogen temperature, embedded in O.C.T. (Reichert-Jung, Paris) and cut into sections with a Cryocut microtome (Reichert-Jung). The sections, air dried on glass slides were fixed in acetone for 3 minutes.
Immunocytological and Histological Tests. Indirect radioimmunological tests were performed on cell suspensions as des¬ cribed previously for hybridoma supernatant screening (11). Briefly, the cells were incubated with the supernatants studied, washed and exposed to sheep anti-mouse Fab'2 radiolabelled with iodine-125 (Amersham Les Ulis, France). The radioactivity of the cell pellet was measured and compared with controls in which incubation with hybridoma supernatant had been omitted. A supernatant of hybrid W6/32 culture which contained a polymorphic antiHLA class I MAb, was used as positive control. Po66, an MAb of IgGl isotype previously studied in the laboratory (10,11) was used as a negative control. Indirect immunofluorescence tests on cell suspensions were also carried out, using a fluorescein-labelled sheep anti-mouse Fab'2 as second antibody layer (New England Nuclears, Dupont de Nemours France, Paris), (9). The results were analysed by flow cytometry (Ortho H 50 system). Cell light scattering properties at right angle (RAS) or forward angle (FAS) were utilized for the selection of unfixed monocyte, lymphocyte and granulocyte subpopulations. The results were expressed as percentages of positive cells by reference to control cells incubated with fluorescent Fab'2 alone and by the mean fluorescence intensity given in arbitrary units. Cell smears and tissue sections were stained by the APAAP (Alkaline Phos¬ phatase Anti-Alkaline Phosphatase) technique according to Cordell et al. (12). Briefly, sections were rehydrated and protein binding sites saturated with PBS-BSA. The slides were, then, exposed successively to the monoclonal antibody studied, rabbit anti-mouse immunoglobulin (Dako-Sebia, Issy les Moulineaux, France) and the APAAP complex (Dako). Alkaline phosphatase activity was revealed by incubation with a substrate containing fuchsin and the sections were counterstained with Harris haematoxylin. The ABC tech¬ nique of immunoperoxidase staining (13) was also used (Vectastain kit). APAAP and May-Grünwald-Giemsa (MGG) stainings were combined, according to Pryzwansky et al. (14). Briefly, bone marrow smears were stained with MGG and selected areas were photographed. After destaining, the slides were stained by the APAAP technique and the same areas photographed. Studies
Antigens from Cytosolic Preparations. Cytosols were prepared from 10& mononuclear cells and 1.5x10^ granulo¬ cytes. The cells were lysed by sonication and centrifuged at 10,000 g for 15 minutes. The supernatant was adjusted at 6mg/ml proteins, aliquoted and on
stored at -20°C.
137
tests, 2 µ samples of various dilutions of cytosol were CM membranes (Amersham, Les Ulis, France) The membranes were saturated for 4 hours in PBS-3% BSA and incubated with one of the following, sodium periodate, papain (Sigma), pepsin (Sigma), pronase (Calbiochem, France Biochem), trypsin-EDTA (Eurobio), collagenase II (Sigma) or DNAse I (Sigma), RNAse A (Sigma) or neuraminidase from Vibrio cholerae (Sigma). Digestion was carried out at 37°C for 2 hours in PBS-BSA buffer. After digestion, the membranes were washed and incubated for 4 hours with 1:30 diluted radiolabelled AP-282. The membranes were then washed 6 times with PBS-BSA and processed for autoradiography The reactivity of AP-282 with purified lysosomal enzyme extracts was carried out as follows. One hundred mg/ml aqueous solutions of myeloperoxidase from human leukocytes (Sigma), lactoferrin from human milk (Sigma), ca¬ thepsin G from human leukocytes (Sigma) and elastase from granulocytes kindly provided by Dr D.Y. Mason (Oxford) were deposited on Hybond CM membranes. After saturation in a PBS-BSA solution, the membranes were ex¬ posed to 1 mg/ml AP-282 solution or to various control antibodies: polyclonal rabbit anti-lactoferrin antibody (Dako), mouse anti-myeloperoxidase MP07 and anti-elastase NP57 MAbs, both generous gifts from Dr D.Y. Mason. After washing in PBS, the membranes were exposed to a rabbit antimouse antibody (Dako) for 30 minutes and to the APAAP complex (Dako). Alkaline phospha¬ tase was revealed by nitrobluetetrazolium (Sigma) and bromochlorylindolyl phosphate (Sigma) as described by McGadey (15).
enzymatic digestion deposited on Hybond For
RESULTS
Production of MAb AP-282. A cell fusion between SP2/0 plasmacytoma cells and mouse splenocytes immunized against purified human blood PMN gave rise to 93 hybrids, of which 13 reacted lymphocytes in a against human granulocytes but not against leukemic radioimmunological assay. The monoclonal antibody AP-282 was selected because it
strongly
neutrophils but not lymphocytes (Table 1 ). antibody AP-282 was found of IgGl kappa isotype.
labelled
Monoclonal
Table 1 : Reactivity of Mab AP282 with granulocytes and leukemic determined by an indirect radioimmunological assay.
B. LYMPHOCYTES
ANTIBODY
GRANULOCYTES
AP282
16037
226
5435
9804
506
186
Anti-HLA
None1
Results expressed in cpm bound to 10.E5 cells. anti-mouse Fab'2 alone.
138
lymphocytes as
(l):
cells incubated with radiolabelled
Reactivity of MAb AP-282 with Circulating Blood Cells. The reactivity of MAb AP-282 with blood PMN purified by centrifugation on a density gradient was investigated using an indirect immunofluorescence test and was analysed both microscopically and by flow cytometry. As can be
in seen Figure 1, unfixed cells did not react with AP-282 although they bound the anti-HLA MAb. By contrast, cells fixed with 70% ethanol were Concentrations stained. of of 1 strongly purified antibody µg/ml proved suf¬ ficient to detect the reaction. Various fixatives were tested. No labelling with AP-282 was seen after formalin fixation. This fixative caused poor permeabilization, as judged by trypan blue penetration and did not allow AP282 staining. The use of ethanol-acetic acid also gave poor results, probably because its acidic pH destroyed the antigen, as seen below.
II
I
i*"i«-"—"it¡-
L
TS
'
15
'
ÎT
*a
—
'
k
'
di
im
J
íaíi
75
îw
73
Í9»
~aíF
rss
" "
'
. SS
Ï5~
Figure 1. Reactivity of MAbs AP-282 and W6/32 with granulocytes, as measured by immunofluorescence and cell flow analysis. I : Unfixed granulocytes, II : Ethanol fixed granulocytes. A : AP-282 5 µg/ml, : AP-282 1 µg/ml, C : W6/32 hybridoma supernatant. -
Staining of blood smears fixed with ethanol by the APAAP technique, showed that MAb AP-282 reacted with 95% neutrophilic PMN, a result in agreement with flow cytometric measurements. Antibody concentrations as low as 0.1 mg/ml gave positive results (Figure 2). Eosinophils were unlabelled by AP282 Blood monocytes were clearly less reactive than PMN. When identified their by ability to phagocytize latex particles, only a low percentage stained positively. Macrophages derived from one week cultures of blood mononu¬ clear cells appeared as large cells exhibiting CD-14 antigen. About 10% of them had cytoplasmic granules intensively stained by AP-282. Similar results found with the macrophages of two ascitic malignant metastasis. The were histiocytic U937 cell line also bound the antibody. .
139
Figure 2. Left: APAAP staining of a smear of purified granulocytes with 0.1 AP-282. (G 1100). Right: control without exposure to MAb AP-282.
µg/ml
Tissue distribution of the antigen recognized by AP-282. The antibody was tested with bone marrow smears from patients. A non bone marrow malignant sample containing 15% promyelocytes was submitted to a double staining technique (14). Promyelocytes were unstained or weakly stained by AP-282, whereas myelocytes and more mature cells of the neutrophilic series were positive.
The distribution of the antigen recognized by Mab AP-282 among solid organs was studied on Cryostat tissue sections using the APAAP technique. The anti¬ body reacted with isolated cells present in blood vessels or forming inflam¬ matory clusters. Positive cells exhibited frequently a multilobulated nucleus characteristic of granulocytes. The reactivity of MAb AP-282 was investigated on sections of various paraffin-
embedded tissues: skin, gastric and colonie mucosa, lung, liver, kidney, thyroid gland, cardiac muscle, spleen, lymph node, acute abcesses and granulomas. The results clearly depended on the technique of tissue fixation. Blocks fixed with Bouin or Dubosq acidic fixatives did not show any positivity whereas, in tissues fixed with 10% formalin in saline (pH 7), isolated positive cells were clearly detected with concentrations of antibody of 1 µg/ml (Figure 3) or less. The reactive cells often showed the characteristic PMN nucleus. Some tissue macrophages also stained but Kupffer cells and alveolar macrophages were unlabelled. All other cells were
negative.
Cellular localization of the antigen identified by AP-282. In most instances, when cells were fixed with ethanol and stained as suspen¬ sions by immuno-fluorescence or on smears with APAAP or immunoperoxidase, the diffuse cytoplasmic staining was seen with AP-282. Sometimes nu¬ clear staining was seen. However, when cell suspensions of PMN, macro¬ U937 line or fixed with were phages paraformaldehyde, the antigen visualized the APAAP the or by immunoperoxidase technique appeared cytoplasmic and associated with granulations (Figure 2).
140
fîf*
Figure appear
H
Vf
·
'
3. APAAP staining of paraffin serial sections of thyroid gland. Granulocytes darkly stained cells (left) absent on the control (right) (G 230).
as
Biochemical characterization of antigen. A comparison of binding of radioiodinated MAb AP-282 with cytosolic prepa¬ rations from purified blood PMN and from purified blood mononuclear cells Radioiodinated AP-282 was incubated with cytosols deposi¬ was performed. ted on methylcellulose membranes. The antibody reacted much more strongly with the cytosol from PMN than with that of mononuclear cells, as also allowed to revealed by autoradiography. The dot blot technique determine the susceptibility of the antigen to various treatments. The reactivity of AP282 with PMN cytosol was altered essentially by low pH pretreatment and by pronase digestion. The alteration observed after pepsin digestion were presumably related to the low pH of the reaction buffer (Table
2). Table 2 : Dot blot reaction of PMN various enzymatic digestions.
Enzyme
cytosol
with Mab AP-282 after submission to
pH
Concentration
AP282
binding
+++ 7.2 2.8 2.8 250 u/ml Pepsine +++ 50 u/ml 6.0 Papaïne + Pronase 100 u/ml 7.2 +++ 7.2 Trypsine/EDTA2 DNase +++ 5.0 1000 u/ml +++ RNase 7.2 250 u/ml 4-++ 5.0 Neuraminidase 5 u/ml +++ 7.2 20 mM Periodate from autoradiography. (1): arbitrary quotation of AP-282 reactivity (2): Trypsin EDTA solution for cell culture detachment.
PBS Acidic buffer
141
1
ANTIBODIES
ANTIGENS
AP-282
Anti-myeloperoxic G
Ca
LM
My
________________
.; O
-
G
LM
ö
-
*
G
El
My -
LM
My
rt
El
Anti-elastase
La
-
*
AP-282
AP-282
Anti-lactoferrin
La
Figure 6: Reactivity of antigens deposited on nitrocellulose membranes with AP-282 and other antibodies. The antigens deposited were cytosolic preparations from granulocytes (G) and mononuclear cells (lymphocytes and monocytes: LM), purified Cathepsin-G (Ca), myeloperoxidase (My), elastase (El) and lactoferrin (La). The antibodies tested, in addition to MAb AP-282, were directed against myeloperoxidase, elastase and lactoferrin. Some purified substances present in the granules of granulocytes were allowed to react with AP-282 and other antibodies taken as controls. As in Figure 6, AP-282 did not bind to any of the substances tested: shown myeloperoxidase, elastase, cathepsine G or lactoferrin, although it reacted with PMN extract and the substances deposited were recognized by the corres¬ ponding antibodies. Anti-lactoferrin MAb was found incidently to react with the myeloperoxidase extract. The molecular weight of the AP-282 antigen could not be determined by im¬ munoprecipitation or Western blotting.
DISCUSSION was found to bind predominantly to PMN and their bone mar¬ precursors. Only weak labelling of monocytes and macrophages was seen. AP-282 did not stain any non hematopoietic cells. Since antibody AP-282 reacted with more than 95% blood PMN and stained these cells intensely at low concentrations on paraffin sections, AP-282 appears to be a reliable mar¬ ker for the identification of granulocytes on tissue sections or on blood and
MAb AP-282
row
142
bone marrow smears. The only reservation is that cells and tissues should be fixed adequately in non acidic fixatives. The lack of detection of the antigen after Dubosq fixative, indeed, was likely related to the acetic acid content of the reactant, which lowered the pH to 3.5. This hypothesis was sustained by dot blot investigations showing that the antigen was degradated by acidic pH. The site of localization of the antigen recognized by AP-282 was not the cell membrane, as suggested by the non reactivity of unfixed cells with the anti¬ body. The precise cytoplasmic localization raised some difficulties since it va¬ ried according to the fixation procedure. Diffusion phenomena of staining when studying antigens from PMN gra¬ seems to be frequently encountered nules. Results similar to those reported here have been found with the antielastase monoclonal antibody NP57 (4). Some fixatives particularly paraformaldehyde seemed more reliable than others and, in our study, showed the association of AP-282-identified antigen with cytoplasmic granules, in granu¬ locytes, monocytes, macrophages and U937 cells. The chemical nature of the antigen recognized by AP-282 has not been de¬ termined. A number of components have been now characterized in neutro¬ phil granules. In primary granules, myeloperoxidase, low molecular weight defensins (16) and several proteinases such as cathepsin G, elastase and rela¬ ted molecules like AGP7, azurocidin, p29b (17,18), proteinase 3 (19) and the antigen involved in Wegener's disease (20) have been identified. Lactoferrin, collagenase and gelatinase (21) were localized to secondary granules. The amino acid sequence of severü of these proteins has been recently establi¬ shed: myeloperoxidase (22), :Lthepsin G (8), antibiotic proteins (23) and We¬ gener's granulomatosis antiger. (20). Numerous aminoacid sequence homologies were found between the various compounds. For example, cathepsin G, elastase, proteinase-3, azurocydinô, AGP7 shared a high degree of similarity in their NH2-terminal amino acid sequence (18). This probably explains why MAbs directed against PMN may show a certain degree of cross reactivity when tested against different purified components of the granules (7). In this regard, it seems of interest that the epitope recognized by AP-282 was not ex¬ pressed on the more usual cytoplasmic markers of human granulocyte, elas¬ tase, myeloperoxidase, cathepsin G and lactoferrin although the specific anti¬ bodies taken as controls reacted positively. The reactivity of the polyclonal rabbit anti-lactoferrin antibody with the myeloperoxidase extract was observed. This might be related to an incomplete purification of one of the reactants. The weak or non expression of the antigen recognized by MAb AP282 on promyelocytes, in contrast with its presence on myelocytes and their descendants, suggests that AP-282 is a marker of secondary granules (14). Further investigations will be necessary to confirm this result, in particular comparisons of the localization of AP-282 with other known markers of secondary granules, like lactoferrin. This could be achieved with a panel of bone marrow samples by conventional techniques and double staining experiments in electron microscopy, as reported in (1-3). Antibody AP-282 may, thus, appear as a new reagent that can characterize granulocytes histologically and as a valuable tool for cytological studies de¬ signed for example to classify myeloid leukemias. Investigations to characterize the molecule(s) recognized by AP-282 are now required and might allow the identification of new granulocyte component(s).
143
ACKNOWLEDGMENTS: The authors wish to thank Drs D. Mason and K. Pulford for their kind assist¬ ance and skilled advice. Work supported by Comité d'Ille et Vilaine de la Ligue Nationale Française .
REFERENCES
1. Calafat, J., Goldschmeding, R., Ringeling, P.L., Janssen, H. and Van der Schoot, CE. (1990) In situ localization by double labeling immuno electron microscopy of anti-neutrophil cytoplasmic auto antibodies in neutrophils and monocytes. Blood. 75: 242-250.
2. Cramer, E.M., Beesley, J.E., Pulford, K.A., Breton-Gorius, J. and Mason, D.Y. (1989) Co-localization of elastase and myeloperoxidase in human blood and bone marrow neutrophils using a monoclonal antibody and immunogold. Am. J. Pathol. 134, 1275-1284. 3. Esaguy, ., Aguas, A.P. and lactoferrin in human neutrophils neity. J. Leuk. Biol. 46,51-62.
Silva, M.T. (1989) High resolution localization of :
labeling
of
secondary granules
and cell
heteroge¬
4.Pulford, K.A.F., Erber W.N., Crick, J.A., I Olsson, I., Micklem, K.J. Gatter, K.C., and Mason, D.Y. (1988) Use of monoclonal antibodies against human neutro¬ phil elastase in normal and leukaemic myeloid cells. J. Clin. Pathol. 41, 853-860. ,
5.Homma, T., Suzuki, K., Kudo, Y., Inagawa, M., Mizuno, S., Yamaguchi, K. and M. (1989) Preparation and characterization of monoclonal antibodies human myeloperoxidase. Arch. Biochem. Biophys. 273, 189-196.
Tagawa,
against
6. Zaki, S.R., Chan,W.C, McKolanis, J. and Austin, G.E. (1989) Production and characterization of monoclonal antibodies to human myeloperoxidase. Clin.ImmunolImmunopathol. 50, 283-297. 7. Skubitz, K.M., Christiansen, N.P. and Mendiola, J.R. (1989) Preparation and cha¬ racterization of monoclonal antibodies to human neutrophil cathepsin G, lactoferrin, eosinophil peroxidase, and eosinophil major basic protein. J. Leukoc. Biol. 46, 109118.
8. Hohn, P.A., Popescu, N.C., Hanson, R.D., Salvesen, G. and Ley, T.J. (1989) Genomic organization and chromosomal localization of the human cathepsin G gene. J. Biol. Chem. 264, 13412-13419. 9. Martin, ., Lecorre, R., Pellen, P., Bourel, D., Merdrignac, G, Genetet, B,.and Toujas, L. (1986) A monoclonal antibody reacting with a mouse specific epitope of Macl antigen. Tissue antigens. 28, 14-23. .
10. Martin, ., Pellen, P., Youinou, P., Collet, B., Desrues, .,Bourel, D., Dazord,
144
L. and Toujas, L. (1989) Characterization of the antigen identified by P066, a mo¬ noclonal antibody raised against a lung squamous cell carcinoma. Cancer Immunol. Immunother. 29, 118-124.
11. Dazord, L., Bourel, D., Martin, ., Lecorre, R., Bourguet, P., Bohy, J., Saccavini, J.C., Délavai, P., Louvet, M. and Toujas, L. (1987) A monoclonal antibody (Po66) directed against human lung squamous cell carcinoma. Immunolocalization of tumour xenografts in nude mice. Cancer Immunol. Immunother, 24, 263-268.
12. Cordell, J.L., Falini, B., Erber W.N., Ghosh, A.K., Abdulaziz, Z., Macdonald, S., Pulford, K.A.F., Stein, F. and Mason, D.Y. (1984) Immunoenzymatic labeling of
monoclonal antibodies using immune complexes of alkaline phosphatase and mono¬ clonal anti-alkaline phosphatase (APAAP complexes). J. Histochem. Cytochem. 32, 219-229.
13. Hsu, S., Raine, I. and Fänger, H. (1981) The use of avidin-biotin peroxidase complex (ABC) in immunoperoxidase technique. J. Histochem., 29, 577-581. 14.
Pryzwansky, K.B., Rausch, P.G., Spitznagel, J.K. and Herion, J.C. (1979) Immunocytochemical distinction between primary and secondary granule formation in developping human neutrophils: correlation with Romanansky stains. Blood, 53,179-185.
15. McGadey, J. (1970) A tetrazolium method for Histochem. 23, 180-184.
non
specific alkaline phosphatase.
16. Lehrer, R.I., Barton, ., Daher, . ., Harwig, S.S., Ganz, T.and Selsted, .E. (1988) Interaction of human defensins with Escherichia coli. Mechanism of bacteri¬ cidal activity. J Clin Invest. 84,553-561. 17. Campanelli, D., Detmers, P.A., Nathan, CF. and Gabay, J.E. (1990) Azurocidin and a homologous serine protease from neutrophils. Differential antimicrobial and proteolytic properties. J. Clin. Invest. 85, 904-915. 18. Wilde, CG, Snable J.L., Griffith, J.E. and Scott, R.W. (1990) Characterization of two azurophil granule proteases with active-site homology to neutrophil elastase. J. Biol. Chem. 265, 2038-2041. 19. Kao, R.C., Wehner, N.G., Skubitz, K.M., Gray. B.H, and Hoidal JR. (1988) Proteinase 3: a distinct polymorphonuclear leukocyte protease which produced em¬ physema. J. Clin. Invest. 82, 1963-1973. 20. Lüdemann, J., Utecht, B. and Gross, W.L. (1990) Anti-neutrophil cytoplasm an¬ tibodies in Wegener's granulomatosis recognize an elastinolytic enzyme. J. Exp. Med.
171,357-362.
21. Hibbs, M.S. and Bainton, D.F. (1989) Human neutrophil ge latinase is nent of specific granules. J. Clin. Invest. 84, 1395-1402. 22.
a
compo¬
Wright, J., Bastian, ., Davis, T.A., Zuo, C, Yoshimoto, S., Orme-Johnson, 145
W.
A. Structural characterization of the isoenzymatic forms of hu¬ myeloperoxidase, evaluation of the iron containing prosthetic group. Blood, 75,
and Tauber man
(1990)
238-241.
Gabay, JE., Scott, RW„ Campanelli, D., Griffith, J., Wilde C, Marra M.M., Seeger, M. and Nathan, CF. (1989) Antibiotic proteins of human polymorphonu¬ clear leukocytes. Proc. Nati. Acad. Sci. 86, 5610-5615. 23.
Address reprint requests to: A. MARTIN Laboratoire d'Immunologie-Immunothérapie Centre Eugène Marquis 35033 RENNES CEDEX (France) -
146
