Lactoferrin,. Eosinophil Peroxidase, and. Eosinophil Major Basic Protein. Keith M. Skubitz,. Neal P. Christiansen, and John R. Mendiola. Department of Medicine ...
Journal
of Leukocyte
Biology
46:109-118
(1989)
Preparation and Characterization of Monoclonal Antibodies to Human Neutrophil Cathepsin G, Lactoferrin, Eosinophil Peroxidase, and Eosinophil Major Basic Protein Keith
M. Skubitz,
Neal
P. Christiansen,
of Medicine, The University
Department
Cancer This
and John
of Minnesota Medical School, Center, Minneapolis
R. Mendiola and the Masonic
report
describes the production and characterization of five murine monoclonal react with granule proteins of human granulocytes. Monoclonal antibody AHN-1 1 (lgG2a) reacted specifically with neutrophiI cathepsin G; no reactivity with the homologous neutrophll neutral proteases, elastase, protelnase 3, or esterase N was detected. Antibodies AHN-9 (IgGi) and AHN-9.1 (lgG2b) each reacted with different epitopes on human lactoferrin, but not with the homologous protein transferrin. Two IgGi antibodies, AHE-1 and AHE-2, reacted specifically with eosinophils; AHE-1 reacted strongly with eosinophIl peroxidase but not eoslnophii major basic protein while AHE-2 recognized eosinophil major basic protein but not eoslnophil peroxidase. All five antibodies antibodies
that
could detect their respective tibodies antigens
antigens
in alcohol-fixed
cytospin
preparations.
an-
These
should
be useful for immunolocalizatlon and quantification of their respective as well as for other studies of the roles of these proteins in granulocyte function
and differentiation. Key words:
granulocyte,
protease,
immunochemical
INTRODUCTION Human granulocytes contain a variety of biologically active molecules located within several different types of granules. These granule contents are believed to play an important role in both normal host defense as well as in the pathogenesis of a variety of disease processes [1,9,11]. The two most widely studied granules of neutrophils are the primary and secondary granules. Primary granules contain a variety of proteases including elastase, cathepsin G, and the more recently described serine proteases, esterase N and proteinase 3 [1,4,14,28]. In some situations, these proteases may be released and play a role in the pathogenesis of a variety of diseases such as chronic lung disease [11,12,21,27], the adult respiratory distress syndrome [7,9,16,17], and destructive joint diseases [20]. While studies of the role of neutrophil proteases in pathologic states such as emphysema have focused on elastase, other neutrophil proteases may also play an important role and have not yet been thoroughly evaluated. Indeed, several of these proteases are structurally related. For example, cathepsin G, elastase, and proteinase 3 share a high degree of homology in their N-terminal amino acid sequences [29,31]. Thus, assays of cathepsin G activity often detect other neutrophil pro© 1989
Alan R. Lisa, Inc.
staining
teases, and polyclonal antisera to cathepsin G may also react with these related proteases. The major component of secondary granules is lactoferrin, a protein that exerts a regulatory influence on myelopoiesis by feedback inhibition, has an antibacterial action, and also can modulate several immune responses [2,25,26]. Eosinophils also contain several granule proteins, including some that are specific to the eosinophil. Among these are eosinophi! peroxidase and eosinophil major basic protein, which are believed to play important roles in defense against parasitic diseases as well as a role in the injury of normal tissue in some allergic and hypereosinophilic states [8]. This report describes the production and characterization of five murine monoclonal antibodies that react specifically with granule proteins. AHN-9 (an IgG 1) and AHN-9. 1 (an IgG2b) both react strongly with lactoferrin but not the related protein transferrin. The
Received
December
19,
1988;
accepted
February
24,
Neal P. Christiansen’s University of Tennessee,
current address is Department Memphis, TN 38163.
Reprint neapolis.
Skubitz,
requests: Keith MN 55455.
Box
325,
University
1989. of Medicine,
Hospital,
Min-
110
Skubitz
et al.
IgG2a monoclonal antibody AHN- 11 reacts with human neutrophil cathepsin G but not with the other major neutrophil neutral proteases: elastase, proteinase 3, or the newly purified neutral protease, esterase N [23,31]. Two eosinophil-specific monoclonal antibodies, AHE- 1 (IgGi) and AHE-2 (IgGl), react specifically with eosinophil peroxidase and eosinophil major basic protein, respectively.
MATERIALS Preparation
AND METHODS of Peripheral
Blood
Cells
Neutrophils, red blood cells, lymphocytes, monocytes, and platelets were prepared from heparinized (2 units/mi) human venous blood as described [24]. Eosinophils from a patient with the hypereosinophilic syndrome were a gift of Dr. D. Hurd, University of Minnesota Medical School, Minneapolis. Immunization Hybridoma
and Production Cell Lines
of
Monoclonal antibodies were prepared as previously described [24]. Briefly, for the production of antibodies AHN-9, AHN-9.l, and AHN-ll, 8 wk-old female BALB/cJ mice (Jackson Laboratories, Bar Harbor, ME) were immunized by intrapentoneal injection of 1O human neutrophils in Freund’s complete adjuvant (GIBCO Laboratories, Grand Island, NY) and subsequently subcutaneously, at weekly intervals, with l0 neutrophils in incomplete Freund’s adjuvant (GIBCO). One week following the third immunization, mice received 1 o neutrophils in phosphate-buffered saline, pH 7.2 (PBS) intraperitoneally. Four days later, their spleen cells were fused with murine myeloma P3-X63-Ag8.653 cells at a ratio of four spleen cells per myeloma cell, as previously described [24]. Hybridoma cell cultures secreting antibodies that bound to neutrophils were twice cloned in soft agar as described [24]. Immunoglobulin isotype was determined by using a commercially available isotyping kit (Bio Rad, Richmond, CA). A similar procedure was used for the production of AHE- 1, AHE-2, and the CG and EN series of antibodies, except that in the case of AHE- 1, eosinophils from a patient with the hypereosinophilic syndrome were used instead of neutrophils. For the production of AHE-2, 50 g of protein purified from these hypereosinophilic eosinophils by wheat germ affinity chromatography [5] was used in place of io neutrophils for immunization, and 0.5 ig of this protein preparation was used in each solid-phase assay for supernatant screening. For the production of the CG and EN series of antibodies, 10-50 ig of purified cathepsin G or esterase N, respectively, was used in place of l0 neutrophils for each immunization, the first and last being intraperitoneal immunizations with other immuniza-
tions being subcutaneous, and 0.5 p.g of these proteins was used in each solid-phase antibody-binding assay for supernatant screening. Antibody-Binding
Assays
Antibodies were assayed for cell-binding activity by a solid-phase assay, as described [24]. Briefly, target cells were suspended at 106/ml in PBS and 0.02% NaN3, and aliquots were applied to wells of a Linbro 96-well microtiter plate and dried overnight at 23#{176}C. Two hundred microliters of PBS containing 50 mg/mi BSA and 0.02% NaN3 was then added to each well and incubated for 4 h at 23#{176}C. This solution was then removed, and 50 l of supernatant to be tested was added and incubated for 1 h at 23#{176}C. Excess antibody was removed by washing three times with 200 ii of ice-cold PBS containing 1 mg/mi BSA and 0.1% Triton X-l0O; ‘251-labeled sheep antimouse immunoglobulin (100,000 cpm, specific activity 10-50 pCi/g, Amersham, Arlington Heights, IL) was then added in 50 p.! of buffer containing 20 mM Tris-HC1, pH 7.6, 1 mM EDTA, 100 mM NaC1, and 20 mg/mi BSA, and incubated for 1 h at 23#{176}C. After washing as above, 100 pA of 2 M NaOH was added and after 30 mm at 23#{176}C, the solution was transferred to glass tubes and the radioactivity was counted. Identical assay procedures were also performed in which various purified proteins, as indicated, in 50 pA aliquots were substituted for target cells. Plasmin, elastase, lactoferrin, transferrin, thrombin, and kallikrein were obtained from Sigma. Human neutrophil elastase, cathepsin G, proteinase 3, and esterase N, purified as previously described [10,14,23,31], were gifts from Drs. B. Gray and N. Wehner, University of Minnesota, Minneapolis. Eosinophil peroxidase and eosinophil major basic protein were gifts of Dr. A. Slungaard, University of Minnesota, Minneapolis.
50ij
Antibodies Mouse antihuman neutrophil antiserum was prepared from CS7BL/6J mice as described [24]. Monoclonal antibody P3, and IgG1 with no known specificity, was obtained from the supernatant of the P3-X63-Ag8 cell line. Rabbit antimouse immunoglobulin was a gift of Dr. M. Wick, University of Minnesota Medical School. The antihuman neutrophil elastase monocloncal antibody AHN-lO (IgGi) has been previously described [23]. The source of monoclonal antibodies was either hybrid cell culture supernatants or ascites fluid diluted in PBS as indicated. All sera and ascites were heat inactivated at 56#{176}C for 30 mm and clarified by centrifugation at 50,000g at 4#{176}C for 1 h before use. Immunocytochemical Immunocytochemical cytospmn preparations
Staining studies were performed by using the alkaline phosphatase-
on
Mnn8l8nal AntiWin t Oanulvti
Prtslna
111
4
,#{149}.#{149}
Fit. 1, lmmunQchQmicul laiflIfl 91 p9rlptIEaI I99U cII y antibody AllN-il. Cytospln preparations were reacted wIth control culture supematant (panel A) or AHN-11 (panels B and C) and then staIned by using the APAAP technique as described In the text. The cells In panel A were from the bottom of a FicollHypaque gradient (neutrophil rich) to demonstrate lack of reactivity with control supernatant. All neutrophils, but not eoslnophils, In this preparation reacted with AHN-1 1 (not Shown). The cells In panels B and C were from the interface of a Ficoll-
mouse-anti-alkaline-phosphatase (APAAP) technique as previously described [6,23]. Briefly, cytospin preparations were immersed in ethanol immediately after preparation and stored at -20#{176}Cuntil use. The slides were then rehydrated by dipping 40 times in each of the following: two changes of ethanol, one change of 95% ethanol, one change of 70% ethanol, and three changes of distilled water. The slides were then blocked with 2% normal rabbit serum in PBS for 10 mm at 23#{176}C. Appropriately diluted primary antibody (2 drops) was then applied and the slide was incubated at 23#{176}C for 2 h. The slide was then rinsed first with PBS and thereafter with 0.05 M Tris-HC1, pH 7.6. Rabbit antimouse immunoglobulin diluted 1:40 with 0.1% BSA in PBS (35 p.1) was then added and the slide was incubated for 20 mm at 23#{176}C in a humidified chamber. All subsequent incubations were performed at 23#{176}C in a humidified chamber. The slide was then rinsed with 0.05 M Tris-HC1, pH 7.6, and 35 p.1 of APAAP complex (Dako, Santa Barbara, CA) diluted 1:25 with 0.05 M Tris-HC1, pH 7.6, was added. The slide was then incubated for 40 mm and washed, and 35 p.1 of rabbit antimouse immunoglobulin (1:40) was added. After a 20 mm incubation, the slide was washed and 35 pA of APAAP complex (1:25) was added, and the slide was incubated for another 20 mm. The slide was then rinsed and freshly prepared substrate [2.6 mg Naphthol ASMX phosphate (Sigma), 200 p.1 N,N-dimethylformamide (Sigma), 9.8 ml of 0.5 M TrisHC1, pH 8.2, 2.4 mg Levamisole (Sigma), and 20 mg of Fast Red TR (Sigma)] was added. After incubation for 15 mm, the slide was washed with water and counterstained with Harris Hematoxylin for 60 sec, washed with
Mypaqus WraUlant (lymphocytemU monocyts rich)toallowox. aminatlon of cells other than neutrophils and eosinophlls. AHN11 staIned neutrophlls (arrows) but not lymphocytes, monocytes, eosInophils, or platelets. Since this exposure was designed to optimally demonstrate the APAAP reaction product, the morphology of the neutrophils (identified by arrows in panels B and C) is not clear; appropriate illumination demonstrated the characteristic nuclear structure of neutrophils. Bar represents 20 jim.
water for 5 mm, dipped in saturated Li2CO3 five times, and washed in water for 2 mm. Coverslips were then mounted by using Glycergel (Dako) and the preparations were examined by light microscopy.
RESULTS Characterization
of Antibody
AHN-1 1
Antibody AHN-l 1 bound to normal human neutrophils by solid-phase radioimmunoassay but not to platelets or erythrocytes (data not shown). Immunocytochemical studies demonstrated that monoclonal antibody AHN-l 1 reacted strongly with neutrophils but did not react with other peripheral blood cells, including platelets, lymphocytes, or eosinophils (Fig. 1). Faint immunostaining of some monocytes was occasionally detected. No binding of AHN-ll to the surface of live normal human peripheral blood cells could be detected by using flow cytofluorimetry, implying that the antigen was not expressed on the surface of the cells. AHN- 11 was an IgG2a molecule, as determined by the use of class-specific antisera in the solid-phase immunoassay. In an effort to characterize the antigens recognized by these antibodies, fractions of primary granules, secondary granules, plasma membranes, and cytoplasm were prepared from neutrophils by the technique of Borregaard [3] and used as antigens in the solid-phase-binding assay. AHN-l 1 reacted nearly exclusively with the pnmary granule fraction as determined by the solid-phasebinding assay (data not shown). These results led us to test the reactivity of AHN- 11 with elastase and cathepsin G, the two major proteases of primary granules. AHN- 11
112
Skubitz
et al.
10000
E
E
0. C.)
C,)
0.
8000
C
C
6000
0 C
4000 0
0 .0
0 .0
2000
C
4
C
4
0
C I,
a. a -C
a C.)
C
C.,
C,
a a a
-
uJ
a a a a
uJ
C
C
C
a
E0
a a
a
E 0
.-
-
a
0
.
a -
a
uJ
0 I VVVJ
o a
0. C,)
Fig. 2. ReactIon of AHN-1 1 with purified proteins as determined by radioimmunoassay. One-half microgram of each purified protein, as indicated, was adsorbed to microtiter wells; the wells were then blocked with BSA; and AHN-1 1 bindIng was determined as described in the text. EPO, eosinophii peroxldase; MBP, eoslnophil major basic protein. Specific binding (cpm bound using culture supernatant of the AHN-11 hybridoma as the test antlbody-cpm bound using culture media as the test supernatant) is shown as the mean of three separate determinations ±SD. in each case, the negative control culture
media gave a signal of 5
0
0 .0 C
4
‘-
‘-
o
a
-
C
_I
I-
E
.0 0.
w
-
a.
-
w
w
()
Fig. 5. ReactIon of AHN-9 with purified proteins as determined by radlolmmunoassay. One-half microgram of each purified protein, as indicated, was allowed to adsorb to microtlter wells; the wells were then blocked with BSA; and AHN-9 binding was then determined as described in the text. EPO, eosInophil peroxidase; MBP, eosinophil major basic protein. Specific binding (cpm bound using culture supernatant of the AHN-9 hybridoma as the test antlbody-cpm bound using culture medIa as the test supernatant) Is shown as the mean of three separate determinations ± SD. In each case, the negative control culture media gave a signal of 5
V
Characterization
of Antibody
AHE-1
Antibody AHE- 1 reacted strongly with eosinophils as demonstrated by immunocytochemical staining but did not react with other peripheral blood cells, including platelets, neutrophils, monocytes, lymphocytes, or red blood cells (Fig. 7A,B). No binding of AHE-l to the surface of live normal human peripheral blood cells could be detected by using flow cytofluorimetry. AHE- 1 was of the IgGl subtype as determined by the use of class-specific antisera in the solid-phase immunoassay. In an effort to determine the antigen recognized by AHE1, solid-phase-binding assays were performed with two purified eosmnophil granule proteins. AHE- 1 reacted strongly with eosinophil peroxidase but not with eosinophil major basic protein (Fig. 8). To further test the specificity of AHE-l for eosinophil peroxidase, a variety of other proteins were also tested; AHE- 1 did not react with elastase, cathepsin 0, esterase N, thrombin, plasmin, kallikrein, lactoferrin, or transferrin.
0 .0
C C
-
‘-
o
a
-
6
Characterization
C
4
C)
a .J
C
a a a a
C
a
a. a
.
a
I-
C.)
a UI
C .0
E 0 I-
z
C
a
E
a a a a
a a a
0
a
a Lu
a
Lu
Fig. 6. Reaction of AHN-9.1 with purified proteins as determined by radiolmmunoassay. One-half microgram of each purIfied proteIn, as Indicated, was allowed to adsorb to microtiter wells; the wells were then blocked with BSA; and AHN-9.1 binding was then determined as described in the text. EPO, eosinophil peroxldase; MBP, eosinophil major basic protein. Specific binding (cpm bound using culture hybridoma as the test antibody-cpm
supernatant
of the AHN-9.1
bound using culture media as the test supernatant) is shown as the mean of three separate determinations ± SD. In each case, the negative control culture media gave a signal of 5
>5
V
V
0
0
.0
.0
C
C
4
C
4 0
a. Lu
a
a
a
a a a Lu
0 C -
a PS
Z
.E
a
a
E
a
.C
C
C
E
a
a
C
o a UI
-
a
-
o
a
a
a -
_i
PS
-
! aa o
0 a a
z C
a.
C
a -J
.PS
a C.)
a a a a a
w
C
C
E
E a
.
a
0 I-
a
Fig. 8. ReactIon of AHE-1 with purified proteins as determined by radloimmunoassay. One-half microgram of each purified protein, as indicated, was allowed to adsorb to microtiter wells; the wells were then blocked with BSA; and AHE-1 binding was then determined as described in the text. EPO, eosinophil peroxldase; MBP, eosinophil major basic protein. Specific binding (cpm bound using culture supernatant of the AHE-1 hybridoma as the test antibody-cpm bound using culture media as the test supernatant) is shown as the mean of three separate determInations ± SD. In each case, the negative control culture media gave a signal of
