lournal of Immunological Methods, 154(1992)179-184
179
© 1992ElsevierSciencePublishersB.V. All rightsreserved0022-1759/92/$05.00
JIM 06430
Antibody to recombinant marine tumor necrosis factor (TNF)
neutralizes guinea pig TNF Robin Ruppel-Kerr, Jamie A. Letup and Meryl H. Karol Department of Environmental and Occupational Health, Graduate School of Public Health, Universityof Pittsburgh, Pittsburgh, PA 1526L USA
(Received13November1991,revisedreceived16March 1992,accepted24April1992)
Cotton dust has been found to cause acute pulmonary inflammation and fever in humans and in a guinea pig model of byssinosis. Following 3 h inhalation of cotton dust particles, guinea pig macrophages were found to release ex vivo a factor(sJ toxic to WEHI fibrosarcoma cells. The cytotoxic factor(s) was also present in the bronchoaiveolar fluid. We sought to investigate the mechanism of the inflammatory response to determine whether the factor was TNF. Antibodies to murine TNF were produced by immunizing sets of rabbits using two protocols. All animals produced anti-TNF antibodies with titers of 1/1000-1/25,000. Sera from one set of animals completely neutralized the cytotoxicity of murine TNF toward the WEHI cell line. The antisera neutralized up to 93% of tlie cytotoxicity of guinea pig samples but only 54% of human recombinant TNF. These results identify TNF in pulmonary tissues of guinea pigs following exposure to cotton dust. Moreover, the studies indicate that rabbit antibodies to murine TNF can be used to detect the guinea pig cytokine. Key words: Tumornecrosisfactor, guineapig;Antibodycross-reactivity;Pulmonaryinflammation
Introduction
Tumor necrosis factor (TNF) has been found to have diverse activities in many biological species. The activities have been divided into categories based upon concentrations of TNF (Kunkel et al., 1989). At low levels, it appears to regulate cell and tissue homeostasis; at higher levels it may be involved in tissue remodeling and repair. In chronic processes where there may be prolonged synthesis of TNF, systemic effects, such
Correspondence to: M.H. Karol, Departmentof Environmental and OccupationalHealth, GraduateSchoolof Public Health, Universityof Pittsburgh,Pittsburgh,PA 15261,USA.
as fever, are likely. At the highest levels, cachexia, multiple organ failure, and death may occur. Exposure to cotton dust has been found to cause acute pulmonary inflammation and fever in both humans (Prausnitz, 1936; Rylander, 1990), and in an animal model (Ellakkani et al., 1984; Griffiths-Johnson et al., 1991). Large amounts of a factor cytotoxic to the TNF-sensitive WEHI cell line (Eskandari et al., 1990) were produced by macrophages and found in bronchoalveolar lavage from animals exposed to cotton dust (Ryan and Karol, 1991). In the current study, identification of the factor was sought in order to increase our understanding of the mechanisms underlying toxicity of cotton dust. Since there is considerable homology between cytokine structure in rodents (Tamatani et al.,
180 TABLE I PROTOCOLSFOR IMMUNIZATIONOF RABBITSWITH mrTNF Rabbits &B C, D
Day 1 21 42 1 23 43
Injectionmute i.d. s.c. s.c~ i.d. s.c. s.c.
TNF injected(/~g) S0.0 12.5 10.0 5.0 50.0 12.5
i.d., intraderraal;s.c. subcutaneous. 1989), immunologic cross-reaction of murine and guinea pig TNF was hypothesized. Methodology was developed to raise antibodies in rabbits to murine TNF (mrTNF). Antibodies were assessed for their ability to neutralize samples from guinea pigs exposed to cotton dust.
Materials and methods
Immunization of rabbits Four New Zealand white rabbits (Hazelton Laboratories, Denver, PA) were injected intradermally with murine recombinant TNF (Genentech Corp., San Francisco, CA) in complete Freund's adjuvant. As shown in Table I, two rabbits (A and 13) received 50 /~g mrTNF distributed in 15 sites along both sides of the cervical and thoracic regions of the back. Rabbits C and D each received 5 /~g mrTNF. All rabbits were boosted 21 days later by subcutaneous injection of 12.5 /zg mrTNF in incomplete adjuvant distributed in five injections, and at 42 days with 10 #g mrTNF, in five sites along the nape of the neck. Blood was drawn 7-10 days later following the second and third immunizations.
labeled goat anti-rabbit lgG (1/1000) (ICN Immunobiologicals, Lisle, IL). A 2,2-azinobis(ethylbenzothiazoline-6-sulfonicacid) (ABTS)-peroxide substrate (50 ~l/well) was added and the enzymatic reaction stopped 3 min later by the addition of 5% sodium dodecyi sulfate (SDS), (50 ~l/well). Absorbance was read at 410 nm using an MR600 microplate reader (Dynatech Laboratories, Inc., Chantilly, VA).
TNF Murine recombinent TNF was generously supplied by Genentech. Human rTNF (hrTNF) (Knoll Pharmaceuticals) was a gift from Dr. T. Whiteside, Pittsburgh Cancer Institute.
TNF bioassay TNF was assessed as described previously (Ryan and Karol, 1991). Briefly, WEHI 164 clone 13 mouse fibrosarcoma cells were cultured in RPMI 1640 with 2 mM L-glutamine, 100 U / m i penicillin, 100 /zg/ml streptomycin, and 10% heat-inactivated (30 min at 56 ° C) fetal calf serum and seeded into 96-well tissue culture plates (Flow Laboratories, McLean, VA) at 2 × 104 cells/well. Test samples (100/zl) were added and the plates incubated for 24 h at 37°C. 50 #! of 2 mg/ml 3-(4,5-dimethyithiazol-2-yl)-2,5-diphenyl tetrazolium bromide (Sigma Chemical Company, St. Louis, MO) was added, followed at 4 h by 150 p,l dimethyl sulfoxide. The plates were shaken for 30 rain at ambient temperature, then absorbance was measured at 570 nm. The absorbance was plotted vs. log sample dilution to obtain 50% cytotoxicity. A minimum of 15 wells/plate of control, untreated cells was used to assess cell viability. Each assay included mrTNF and hrTNF standards at an initial concentration of 100 ng/ml.
Guinea pig samples ELISA Microtiter plates (PVC, Fiow Laboratories) were coated by addition of 50 /~l of 2 /~g/mi mrTNF in 0.1 M carbonate buffer, pH 9.6 and incubated at 37°C for 3 h. Rabbit antiserum (f'mal volume 50 /~l) was added at an initial dilution of 1/100 and serial 2-fold dilutions were made. Plates were incubated at 37 °C for 1 h. Bound antibody was detec,ted using peroxidase-
Samples from guinea pigs exposed to cotton dust included: bronchoalveolar iavage fluid (BAL) collected immediately following 3 h of cotton dust inhalation (C-BAL); culture fluid from alveolar macrophages (AM) isolated from the BAL and cultured for 18 h in LPS-free media (less than 0.1 ng LPS/ml) (sample C-M), or in media which contained 10/~g/mi LPS (sample C-M-LPS). A minimum of two samples of each type of prepara-
181 TABLE
II
REGRESSION EQUATIONS FOR CYTOTOXICITY OF GUINEA PIG SAMPLES • C-lttL
1.0
0.5.
0.0
•
,
1
•
.
2
.
.
.
.
3
.
4
,
.
5
Log 10 Dilution
Fig. l. Cytotoxicity of guinea pig samples. Samples were assayed using WEHI 164 clone 13 cells. 100/tl samples were added to 2× 104 target cells. Representative data are shown. C-BAL, bronchoalveolar lavage fluid from a cotton dust exposed guinea pig; C-M, alveolar macrophage culture fluid supernatant from cotton dust-exposed guinea pig; M-LPS, culture fluid from LPS-stimulated AM obtained from a naive animal; C-M-LPS, samples similar to M-LPS obtained from a cotton dust-exposed guinea pig.
tion w a s tested. S a m p l e s f r o m naive animals, n o t e x p o s e d to cotton dust, served as controls.
Sample
Regression equation
C-BAL C-M C-M-LPS M-LPS
y = -0.346 +0.389x y = 0.0842+ 0.3356x y = - 0 . 2 7 3 +0.264x y = - 0.3201 + 0.2ff3x
Cytotoxic responses are shown in Fig. 1. Abbreviations:CBAL, bronchial alveolar lavage fluid from a guinea pig e x posed for 3 h to a cotton dust atmosphere of 33 mg/m3; C-M, culture fluid from alveolar macrophages (AM) isolated from guinea pigs exposed to cotton dust for 3 h; C-M-LPS, AM isolated as for C-M, but cultured in the presence of 10 p.g/ml LPS; M-LPS, AM as in C-M-LPS obtained from naive guinea pigs.
t h e lines were similar, r a n g i n g b e t w e e n 0.264 a n d 0.380. F r o m t h e s e data, t h e dilution o f s a m p l e which c a u s e d 80% cytotoxicity w a s obtained. T h e s e v a l u e s were u s e d for neutralization s t u d i e s with antibody.
Neutralization of cytotoxicity by rabbit anti-mrTNF antisera A n t i b o d i e s to m r T N F were raised in rabbits, a n d t h e a n t i s e r a were e v a l u a t e d u s i n g E L I S A .
Neutralization of TNF activity I n c r e a s i n g c o n c e n t r a t i o n s o f rabbit antim r T N F a n t i s e r u m were i n c u b a t e d with s a m p l e s c o n t a i n i n g T N F w h i c h h a d exhibited 80% cytotoxicity for W E H I cells. Neutralization was ass e s s e d by a d d i n g 100/~1 o f t h e m i x t u r e to wells c o n t a i n i n g 2 × 104 W E H I cells.
\ E
1.~0"
~ q~ ~
-o- Rabl~t A 4- ~ B .~- RJ~Y,t C
Statistics L i n e a r r e g r e s s i o n analysis w a s p e r f o r m e d using Cricket graphics (Cricket Software).
R e s u l t s
Cytotoxicity of guine~ pig samples T h e cytotoxicity o f g u i n e a pig s a m p l e s is s h o w n in Fig. 1. T h e d a t a s u g g e s t a log linear relationship b e t w e e n t h e a m o u n t o f s a m p l e a n d t h e d e g r e e o f cytotoxicity. L i n e a r r e g r e s s i o n analysis o f t h e d a t a is p r e s e n t e d in T a b l e II. T h e slopes o f
Serum Dilution Fig. 2. ELISA for rabbit antibody to mrTNF. PVC plates were
coated using 50 p.l mrTNF (2 /zg/ml). Bound antibody was detected using peroxidase-labeled goat anti-rabbit IgG ( 1/ 1000) followed by ABTS/peroxide substrate.
182 TABLE 111 TITER AND NEUTRALIZING ACTIVITY OF RABBIT ANTIBODIES TO mrTNF Rabbit
Antibody titer
% neutralization a mrTNF g. pig sample
A B C
25000 25000 25000
95 91 100
100 100 23
D
1000
82
0
100
I# 2oi
el
rTNF (0.63 ng/ml) and guinea pig samples(M-LPS) were separately incubated with 1.7 p,! antiserum. "Murine
10 4
10 4
10 ~
10" 10° (ul)
10'
Antisemrn
These results are shown in Fig. 2. Sera from rabbits A, B, and C demonstrated titers of 1/25,000 whereas serum from rabbit D had a titer of 1/1000 (Table liD. Pre-immunization serum from all animals tested negative at 1/100 dilution. Serum samples from each of the rabbits were tested for their ability to neutralize mrTNF. Antisera neutralized between 82% and 100% of the cytotoxic activity of mrTNF (Table liD. The antisera were also able to neutralize the cytotoxieity of guinea pig samples, although to varying extents. A M were obtained from naive guinea pigs and cultured with l0 /~g/ml LPS (M-LPS). Antisera from rabbits A and B completely neutralized these guinea pig samples whereas sera from rabbits C and D were much less effective (Table liD. Only 23% neutralization was achieved by antiserum from rabbit C and 0% was obtained with antiserum from rabbit D. Neutralization of samples obtained from guinea pigs exposed to cotton dust was attempted. The neutralization achieved with antiserum from rabbit A is shown in Fig. 3. The antiserum reduced the cytotoxieity of each of the guinea pig samples. It was also effective against hrTNF. The location of the neutralization curves indicated that the smallest volume of antiserum was needed for neutralization of mrTNF, a larger volume for guinea pig samples and the greatest amount for human TNF. This relationship is depicted in Fig. 4. The amount of antiserum needed to reduce the cytotoxicity of each sample to 50% is shown. The antiserum was most potent with murine TNF, was
Fig. 3. Neutralization of cytotoxicity by rabbit antiserum. Increasing concentrations of antiserum (from rabbit A) were added to samples of TNF that gave 80% cytotoxicitywhen previously assayed (Fig. I). The volume in each well was constant. Amounts of TNF used: 0.20 ng/ml mrTNF, 0.19 ng/ml C-BAL, 0.18 ng/ml C-M, 0.53 ng/ml M-LPS, 1.38 ng/ml C-M-LPS and 1.56 ng/ml hrTNF. Concentrations of all guinea pig samples were calculated using a mrTNF standard. Representative data are shown. Abbreviations as in Fig. 1.
less potent with guinea pig samples, and least potent with human TNF. Approximately 15 times more antiserum was needed to neutralize the guinea pig samples than was needed for murine
e.o0
=.ram c.lm
c.n
ttl.ps
C-M.tPS
hrntF
Sample
Fig. 4. Volume of rabbit anti-mrTNF necessaw for 50% neutralization of TNF samples. Assays were performed and data analyzed as in Fig. 3. Values for guinea pig samples represent the average from two animals. Abbreviationsas in Fig. 1.
183 TNF, and approximately 30 times more antiserum was needed to neutralize human TNF compared with guinea pig samples. Comparison of the murine and human samples indicated that approximately 500 times more antiserum was needed to neutralize hrTNF than was required for mrTNF.
Discussion Inhalation of cotton dust is known to cause acute pulmonary inflammation. Neither the agent(s) within the dust responsible for the response, nor the mechanism of the inflammation has been established. We have developed an animal model (Eilakkani et al., 1984, 1987) to address both of these issues. In the model, exposure of animals to cotton dust was found to result in the release of large amounts of a cytotoxic factor in the lung (Ryan and Karol, 1991). Alveolar macrophages were found to produce the factor upon ex vivo culture (Ryan and Karol, 1991). The association of this factor with cotton dust inflammation was shown by the failure to detect its presence following exposure of guinea pigs to cellulose dust at the same concentration and having the same particle size (Griffiths-Johnson et al., 1991). The factor was assumed to be TNF from its cytotoxicity toward WEHI cells. This cell line has been shown to be sensitive to both TNF (TNF-a) and lymphotoxin (TNF-/3), but not to recombinant IL-1, IL.2, IL-6, or gamma interferon (Sbeehan et ai., 1989; Eskandari et al., 1990). We sought to identify the cytotoxic factor(s) using immunologic means. Immunization of rabbits with microgram quantities of mrTNF yielded high titers of antibody. Each of the antisera neutralized the cytotoxicity of mrTNF toward WEHI cells. Antisera were able to neutralize the cytotoxicity of guinea pig samples. Sera from rabbits A and B produced high titers of neutralizing antibody. Complete neutralization of an LPS-stimulated guinea pig AM sample was obtained with 1.7 p.l antiserum. Of interest was the finding that some antisera (those from rabbits A and B) had high neutralizing activity toward the guinea pig samples whereas
sera of the same titer from rabbits C and D had low neutralizing activity. The latter animals had received a different immunization protocol from rabbits A and B. The amount of serum needed to reduce guinea pig cytotoxicity by 50% ranged from 0.05/tl to 0.33/tl. In contrast, 5 p,! was required to neutralize hrTNF and 0.01 /~l was needed for mrTNF. These results suggest closer structural homology between guinea pig and murine TNF than between human and routine TNF. I13e structure of guinea pig TNF has been studied (Tamatani et ai., 1989). The 21 amino terminal residues have 77% homology with murine TNF and 71% homology with human TNF. However, guinea pig TNF was more effective then hrTNF in the induction of cytotoxic effects toward murine L929 cells (Tamatani et al., 1989). TNF-a is produced primarily by activated macrophages and T lymphocytes. In contrast, TNF-/3 appears to be produced only by T cells. The cytotoxicity demonstrated by the guinea pig bronchial alveolar lavages may thus reflect TNF-a and TNF-/3. In the mouse, these two cytokines show 35.5% sequence homology. Sheehan et al. (1989) found that hamster monoclonal antibodies to mrTNF-a inhibited all the in vitro TNF-dependent cytolytic activity of macrophages and T cells, implying cross-reactivity of the antisera with lymphotoxin. In our study, rabbit polyclonal antisera were effective in neutralizing the cytoxoticity of BAL a~d that of macrophages (Figs. 3 and 4). Since the former sample probably contained both TNF-a and TNF-/3, these results imply cross-reactivity of the antisera with TNF-/3. This report describes a procedure to raise antibodies which can cross-react with a guinea pig cytokine. The antibodies to TNF described here may be used to elucidate the role of TNF in the pulmonary inflammation associated with cotton dust exposure. The finding of cross-reactivity of anti-mrTNF antibodies with guinea pig TNF suggests that these antibodies can be used for immunohistochemical localization of TNF in the lung. In addition, it should now be poss~le to use affinity chromatography to isolate guinea pig TNF and initiate studies of TNF expression in inflammatory lung disease.
184
Acknowledgements T h i s investigation was s u p p o r t e d u n d e r C o o p erative A g r e e m e n t No. 43 YK-50050 b e t w e e n t h e U S D A a n d t h e University o f Pittsburgh, Drs. J a n e Roberts a n d C. K e n Bragg, project officers. T h e a u t h o r s t h a n k G e n e t e c h Inc. for t h e very g e n e r o u s gift o f m r T N F , a n d Lisa K, R y a n for providing t h e g u i n e a pig s a m p l e s . References Ellakkani, M.A., Alarie, Y., Weyel, D.A., Mazumdar, S. and Karol, M.H. (1984) Pulmonary reactions to inhaled cotton dust: an animal model for byssinosis. ToxicoL Appl. Pharmacol. 74, 267. Enakkani, M.A., Alarie, Y., Weyel, D. and Karol, M.H. (1987) Chronic pulmonary effects in guinea pigs from prolonged inhalation of cotton dust. Toxicol. Appl. Pharmacol. 88, 354. Eskandari, M.IC, Nguyen, D.T., Kunkel, S.L. and Remick, D.G. (1990) WEHI 164 subclone 13 assay for TNF: sensitivity, specificity, and reliability. Immunol. Invest. 19, 69.
Griffiths-Johnson, D., Ryan, L.K. and Karol, M.H. (1991) Development of an animal model for organic dust toxic syndrome. Inhal. Toxicol. 3, 405. Kunkcl, S.L., Remick, D.G., Streiter, R.M. and Larrick, J.W. (1989) Mechanisms that regulate the production and effects of tumor necrosis factor-alpha. Crit. Rev. lmmunol. 9, 93. Prausnitz, C. (1936) Investigations on respiratory dust disease in operatives in cotton industry. Medical Research Council, Spec. Rep. Ser. No. 212. His Majesty's Stationery Office, London, p. 60. Ryan, L.K. and Karol, M.H. (1991) Release of tumor necrosis factor in guinea pigs upon acute inhalation of cotton dust. Am. J. Respir. Cell. Mol. Biol. 5, 93. Rylander, R. (1990) Health effects of cotton dust exposures. Am. L Ind. Med. 17, 39. Sheehan, K.C.F., Ruddle, N.H. and Schreiber, R.D. (1989) Generation and characterization of hamster monoclonal antibodies that neutralize murine tumor necrosis factors. J. lmmunol. 142, 3884. Tamatani, T., Kimura, S., Hashimoto, T. and Onozaki, K. (1989) Purification of guinea pig tumor necrosis factor (TNF): comparison of its anti-proliferative and differentiatire activities for myeloid leukemic cell lines with those of recombinant human TNF. J. Biochem. 105, 55.
179
© 1992ElsevierSciencePublishersB.V. All rightsreserved0022-1759/92/$05.00
JIM 06430
Antibody to recombinant marine tumor necrosis factor (TNF)
neutralizes guinea pig TNF Robin Ruppel-Kerr, Jamie A. Letup and Meryl H. Karol Department of Environmental and Occupational Health, Graduate School of Public Health, Universityof Pittsburgh, Pittsburgh, PA 1526L USA
(Received13November1991,revisedreceived16March 1992,accepted24April1992)
Cotton dust has been found to cause acute pulmonary inflammation and fever in humans and in a guinea pig model of byssinosis. Following 3 h inhalation of cotton dust particles, guinea pig macrophages were found to release ex vivo a factor(sJ toxic to WEHI fibrosarcoma cells. The cytotoxic factor(s) was also present in the bronchoaiveolar fluid. We sought to investigate the mechanism of the inflammatory response to determine whether the factor was TNF. Antibodies to murine TNF were produced by immunizing sets of rabbits using two protocols. All animals produced anti-TNF antibodies with titers of 1/1000-1/25,000. Sera from one set of animals completely neutralized the cytotoxicity of murine TNF toward the WEHI cell line. The antisera neutralized up to 93% of tlie cytotoxicity of guinea pig samples but only 54% of human recombinant TNF. These results identify TNF in pulmonary tissues of guinea pigs following exposure to cotton dust. Moreover, the studies indicate that rabbit antibodies to murine TNF can be used to detect the guinea pig cytokine. Key words: Tumornecrosisfactor, guineapig;Antibodycross-reactivity;Pulmonaryinflammation
Introduction
Tumor necrosis factor (TNF) has been found to have diverse activities in many biological species. The activities have been divided into categories based upon concentrations of TNF (Kunkel et al., 1989). At low levels, it appears to regulate cell and tissue homeostasis; at higher levels it may be involved in tissue remodeling and repair. In chronic processes where there may be prolonged synthesis of TNF, systemic effects, such
Correspondence to: M.H. Karol, Departmentof Environmental and OccupationalHealth, GraduateSchoolof Public Health, Universityof Pittsburgh,Pittsburgh,PA 15261,USA.
as fever, are likely. At the highest levels, cachexia, multiple organ failure, and death may occur. Exposure to cotton dust has been found to cause acute pulmonary inflammation and fever in both humans (Prausnitz, 1936; Rylander, 1990), and in an animal model (Ellakkani et al., 1984; Griffiths-Johnson et al., 1991). Large amounts of a factor cytotoxic to the TNF-sensitive WEHI cell line (Eskandari et al., 1990) were produced by macrophages and found in bronchoalveolar lavage from animals exposed to cotton dust (Ryan and Karol, 1991). In the current study, identification of the factor was sought in order to increase our understanding of the mechanisms underlying toxicity of cotton dust. Since there is considerable homology between cytokine structure in rodents (Tamatani et al.,
180 TABLE I PROTOCOLSFOR IMMUNIZATIONOF RABBITSWITH mrTNF Rabbits &B C, D
Day 1 21 42 1 23 43
Injectionmute i.d. s.c. s.c~ i.d. s.c. s.c.
TNF injected(/~g) S0.0 12.5 10.0 5.0 50.0 12.5
i.d., intraderraal;s.c. subcutaneous. 1989), immunologic cross-reaction of murine and guinea pig TNF was hypothesized. Methodology was developed to raise antibodies in rabbits to murine TNF (mrTNF). Antibodies were assessed for their ability to neutralize samples from guinea pigs exposed to cotton dust.
Materials and methods
Immunization of rabbits Four New Zealand white rabbits (Hazelton Laboratories, Denver, PA) were injected intradermally with murine recombinant TNF (Genentech Corp., San Francisco, CA) in complete Freund's adjuvant. As shown in Table I, two rabbits (A and 13) received 50 /~g mrTNF distributed in 15 sites along both sides of the cervical and thoracic regions of the back. Rabbits C and D each received 5 /~g mrTNF. All rabbits were boosted 21 days later by subcutaneous injection of 12.5 /zg mrTNF in incomplete adjuvant distributed in five injections, and at 42 days with 10 #g mrTNF, in five sites along the nape of the neck. Blood was drawn 7-10 days later following the second and third immunizations.
labeled goat anti-rabbit lgG (1/1000) (ICN Immunobiologicals, Lisle, IL). A 2,2-azinobis(ethylbenzothiazoline-6-sulfonicacid) (ABTS)-peroxide substrate (50 ~l/well) was added and the enzymatic reaction stopped 3 min later by the addition of 5% sodium dodecyi sulfate (SDS), (50 ~l/well). Absorbance was read at 410 nm using an MR600 microplate reader (Dynatech Laboratories, Inc., Chantilly, VA).
TNF Murine recombinent TNF was generously supplied by Genentech. Human rTNF (hrTNF) (Knoll Pharmaceuticals) was a gift from Dr. T. Whiteside, Pittsburgh Cancer Institute.
TNF bioassay TNF was assessed as described previously (Ryan and Karol, 1991). Briefly, WEHI 164 clone 13 mouse fibrosarcoma cells were cultured in RPMI 1640 with 2 mM L-glutamine, 100 U / m i penicillin, 100 /zg/ml streptomycin, and 10% heat-inactivated (30 min at 56 ° C) fetal calf serum and seeded into 96-well tissue culture plates (Flow Laboratories, McLean, VA) at 2 × 104 cells/well. Test samples (100/zl) were added and the plates incubated for 24 h at 37°C. 50 #! of 2 mg/ml 3-(4,5-dimethyithiazol-2-yl)-2,5-diphenyl tetrazolium bromide (Sigma Chemical Company, St. Louis, MO) was added, followed at 4 h by 150 p,l dimethyl sulfoxide. The plates were shaken for 30 rain at ambient temperature, then absorbance was measured at 570 nm. The absorbance was plotted vs. log sample dilution to obtain 50% cytotoxicity. A minimum of 15 wells/plate of control, untreated cells was used to assess cell viability. Each assay included mrTNF and hrTNF standards at an initial concentration of 100 ng/ml.
Guinea pig samples ELISA Microtiter plates (PVC, Fiow Laboratories) were coated by addition of 50 /~l of 2 /~g/mi mrTNF in 0.1 M carbonate buffer, pH 9.6 and incubated at 37°C for 3 h. Rabbit antiserum (f'mal volume 50 /~l) was added at an initial dilution of 1/100 and serial 2-fold dilutions were made. Plates were incubated at 37 °C for 1 h. Bound antibody was detec,ted using peroxidase-
Samples from guinea pigs exposed to cotton dust included: bronchoalveolar iavage fluid (BAL) collected immediately following 3 h of cotton dust inhalation (C-BAL); culture fluid from alveolar macrophages (AM) isolated from the BAL and cultured for 18 h in LPS-free media (less than 0.1 ng LPS/ml) (sample C-M), or in media which contained 10/~g/mi LPS (sample C-M-LPS). A minimum of two samples of each type of prepara-
181 TABLE
II
REGRESSION EQUATIONS FOR CYTOTOXICITY OF GUINEA PIG SAMPLES • C-lttL
1.0
0.5.
0.0
•
,
1
•
.
2
.
.
.
.
3
.
4
,
.
5
Log 10 Dilution
Fig. l. Cytotoxicity of guinea pig samples. Samples were assayed using WEHI 164 clone 13 cells. 100/tl samples were added to 2× 104 target cells. Representative data are shown. C-BAL, bronchoalveolar lavage fluid from a cotton dust exposed guinea pig; C-M, alveolar macrophage culture fluid supernatant from cotton dust-exposed guinea pig; M-LPS, culture fluid from LPS-stimulated AM obtained from a naive animal; C-M-LPS, samples similar to M-LPS obtained from a cotton dust-exposed guinea pig.
tion w a s tested. S a m p l e s f r o m naive animals, n o t e x p o s e d to cotton dust, served as controls.
Sample
Regression equation
C-BAL C-M C-M-LPS M-LPS
y = -0.346 +0.389x y = 0.0842+ 0.3356x y = - 0 . 2 7 3 +0.264x y = - 0.3201 + 0.2ff3x
Cytotoxic responses are shown in Fig. 1. Abbreviations:CBAL, bronchial alveolar lavage fluid from a guinea pig e x posed for 3 h to a cotton dust atmosphere of 33 mg/m3; C-M, culture fluid from alveolar macrophages (AM) isolated from guinea pigs exposed to cotton dust for 3 h; C-M-LPS, AM isolated as for C-M, but cultured in the presence of 10 p.g/ml LPS; M-LPS, AM as in C-M-LPS obtained from naive guinea pigs.
t h e lines were similar, r a n g i n g b e t w e e n 0.264 a n d 0.380. F r o m t h e s e data, t h e dilution o f s a m p l e which c a u s e d 80% cytotoxicity w a s obtained. T h e s e v a l u e s were u s e d for neutralization s t u d i e s with antibody.
Neutralization of cytotoxicity by rabbit anti-mrTNF antisera A n t i b o d i e s to m r T N F were raised in rabbits, a n d t h e a n t i s e r a were e v a l u a t e d u s i n g E L I S A .
Neutralization of TNF activity I n c r e a s i n g c o n c e n t r a t i o n s o f rabbit antim r T N F a n t i s e r u m were i n c u b a t e d with s a m p l e s c o n t a i n i n g T N F w h i c h h a d exhibited 80% cytotoxicity for W E H I cells. Neutralization was ass e s s e d by a d d i n g 100/~1 o f t h e m i x t u r e to wells c o n t a i n i n g 2 × 104 W E H I cells.
\ E
1.~0"
~ q~ ~
-o- Rabl~t A 4- ~ B .~- RJ~Y,t C
Statistics L i n e a r r e g r e s s i o n analysis w a s p e r f o r m e d using Cricket graphics (Cricket Software).
R e s u l t s
Cytotoxicity of guine~ pig samples T h e cytotoxicity o f g u i n e a pig s a m p l e s is s h o w n in Fig. 1. T h e d a t a s u g g e s t a log linear relationship b e t w e e n t h e a m o u n t o f s a m p l e a n d t h e d e g r e e o f cytotoxicity. L i n e a r r e g r e s s i o n analysis o f t h e d a t a is p r e s e n t e d in T a b l e II. T h e slopes o f
Serum Dilution Fig. 2. ELISA for rabbit antibody to mrTNF. PVC plates were
coated using 50 p.l mrTNF (2 /zg/ml). Bound antibody was detected using peroxidase-labeled goat anti-rabbit IgG ( 1/ 1000) followed by ABTS/peroxide substrate.
182 TABLE 111 TITER AND NEUTRALIZING ACTIVITY OF RABBIT ANTIBODIES TO mrTNF Rabbit
Antibody titer
% neutralization a mrTNF g. pig sample
A B C
25000 25000 25000
95 91 100
100 100 23
D
1000
82
0
100
I# 2oi
el
rTNF (0.63 ng/ml) and guinea pig samples(M-LPS) were separately incubated with 1.7 p,! antiserum. "Murine
10 4
10 4
10 ~
10" 10° (ul)
10'
Antisemrn
These results are shown in Fig. 2. Sera from rabbits A, B, and C demonstrated titers of 1/25,000 whereas serum from rabbit D had a titer of 1/1000 (Table liD. Pre-immunization serum from all animals tested negative at 1/100 dilution. Serum samples from each of the rabbits were tested for their ability to neutralize mrTNF. Antisera neutralized between 82% and 100% of the cytotoxic activity of mrTNF (Table liD. The antisera were also able to neutralize the cytotoxieity of guinea pig samples, although to varying extents. A M were obtained from naive guinea pigs and cultured with l0 /~g/ml LPS (M-LPS). Antisera from rabbits A and B completely neutralized these guinea pig samples whereas sera from rabbits C and D were much less effective (Table liD. Only 23% neutralization was achieved by antiserum from rabbit C and 0% was obtained with antiserum from rabbit D. Neutralization of samples obtained from guinea pigs exposed to cotton dust was attempted. The neutralization achieved with antiserum from rabbit A is shown in Fig. 3. The antiserum reduced the cytotoxieity of each of the guinea pig samples. It was also effective against hrTNF. The location of the neutralization curves indicated that the smallest volume of antiserum was needed for neutralization of mrTNF, a larger volume for guinea pig samples and the greatest amount for human TNF. This relationship is depicted in Fig. 4. The amount of antiserum needed to reduce the cytotoxicity of each sample to 50% is shown. The antiserum was most potent with murine TNF, was
Fig. 3. Neutralization of cytotoxicity by rabbit antiserum. Increasing concentrations of antiserum (from rabbit A) were added to samples of TNF that gave 80% cytotoxicitywhen previously assayed (Fig. I). The volume in each well was constant. Amounts of TNF used: 0.20 ng/ml mrTNF, 0.19 ng/ml C-BAL, 0.18 ng/ml C-M, 0.53 ng/ml M-LPS, 1.38 ng/ml C-M-LPS and 1.56 ng/ml hrTNF. Concentrations of all guinea pig samples were calculated using a mrTNF standard. Representative data are shown. Abbreviations as in Fig. 1.
less potent with guinea pig samples, and least potent with human TNF. Approximately 15 times more antiserum was needed to neutralize the guinea pig samples than was needed for murine
e.o0
=.ram c.lm
c.n
ttl.ps
C-M.tPS
hrntF
Sample
Fig. 4. Volume of rabbit anti-mrTNF necessaw for 50% neutralization of TNF samples. Assays were performed and data analyzed as in Fig. 3. Values for guinea pig samples represent the average from two animals. Abbreviationsas in Fig. 1.
183 TNF, and approximately 30 times more antiserum was needed to neutralize human TNF compared with guinea pig samples. Comparison of the murine and human samples indicated that approximately 500 times more antiserum was needed to neutralize hrTNF than was required for mrTNF.
Discussion Inhalation of cotton dust is known to cause acute pulmonary inflammation. Neither the agent(s) within the dust responsible for the response, nor the mechanism of the inflammation has been established. We have developed an animal model (Eilakkani et al., 1984, 1987) to address both of these issues. In the model, exposure of animals to cotton dust was found to result in the release of large amounts of a cytotoxic factor in the lung (Ryan and Karol, 1991). Alveolar macrophages were found to produce the factor upon ex vivo culture (Ryan and Karol, 1991). The association of this factor with cotton dust inflammation was shown by the failure to detect its presence following exposure of guinea pigs to cellulose dust at the same concentration and having the same particle size (Griffiths-Johnson et al., 1991). The factor was assumed to be TNF from its cytotoxicity toward WEHI cells. This cell line has been shown to be sensitive to both TNF (TNF-a) and lymphotoxin (TNF-/3), but not to recombinant IL-1, IL.2, IL-6, or gamma interferon (Sbeehan et ai., 1989; Eskandari et al., 1990). We sought to identify the cytotoxic factor(s) using immunologic means. Immunization of rabbits with microgram quantities of mrTNF yielded high titers of antibody. Each of the antisera neutralized the cytotoxicity of mrTNF toward WEHI cells. Antisera were able to neutralize the cytotoxicity of guinea pig samples. Sera from rabbits A and B produced high titers of neutralizing antibody. Complete neutralization of an LPS-stimulated guinea pig AM sample was obtained with 1.7 p.l antiserum. Of interest was the finding that some antisera (those from rabbits A and B) had high neutralizing activity toward the guinea pig samples whereas
sera of the same titer from rabbits C and D had low neutralizing activity. The latter animals had received a different immunization protocol from rabbits A and B. The amount of serum needed to reduce guinea pig cytotoxicity by 50% ranged from 0.05/tl to 0.33/tl. In contrast, 5 p,! was required to neutralize hrTNF and 0.01 /~l was needed for mrTNF. These results suggest closer structural homology between guinea pig and murine TNF than between human and routine TNF. I13e structure of guinea pig TNF has been studied (Tamatani et ai., 1989). The 21 amino terminal residues have 77% homology with murine TNF and 71% homology with human TNF. However, guinea pig TNF was more effective then hrTNF in the induction of cytotoxic effects toward murine L929 cells (Tamatani et al., 1989). TNF-a is produced primarily by activated macrophages and T lymphocytes. In contrast, TNF-/3 appears to be produced only by T cells. The cytotoxicity demonstrated by the guinea pig bronchial alveolar lavages may thus reflect TNF-a and TNF-/3. In the mouse, these two cytokines show 35.5% sequence homology. Sheehan et al. (1989) found that hamster monoclonal antibodies to mrTNF-a inhibited all the in vitro TNF-dependent cytolytic activity of macrophages and T cells, implying cross-reactivity of the antisera with lymphotoxin. In our study, rabbit polyclonal antisera were effective in neutralizing the cytoxoticity of BAL a~d that of macrophages (Figs. 3 and 4). Since the former sample probably contained both TNF-a and TNF-/3, these results imply cross-reactivity of the antisera with TNF-/3. This report describes a procedure to raise antibodies which can cross-react with a guinea pig cytokine. The antibodies to TNF described here may be used to elucidate the role of TNF in the pulmonary inflammation associated with cotton dust exposure. The finding of cross-reactivity of anti-mrTNF antibodies with guinea pig TNF suggests that these antibodies can be used for immunohistochemical localization of TNF in the lung. In addition, it should now be poss~le to use affinity chromatography to isolate guinea pig TNF and initiate studies of TNF expression in inflammatory lung disease.
184
Acknowledgements T h i s investigation was s u p p o r t e d u n d e r C o o p erative A g r e e m e n t No. 43 YK-50050 b e t w e e n t h e U S D A a n d t h e University o f Pittsburgh, Drs. J a n e Roberts a n d C. K e n Bragg, project officers. T h e a u t h o r s t h a n k G e n e t e c h Inc. for t h e very g e n e r o u s gift o f m r T N F , a n d Lisa K, R y a n for providing t h e g u i n e a pig s a m p l e s . References Ellakkani, M.A., Alarie, Y., Weyel, D.A., Mazumdar, S. and Karol, M.H. (1984) Pulmonary reactions to inhaled cotton dust: an animal model for byssinosis. ToxicoL Appl. Pharmacol. 74, 267. Enakkani, M.A., Alarie, Y., Weyel, D. and Karol, M.H. (1987) Chronic pulmonary effects in guinea pigs from prolonged inhalation of cotton dust. Toxicol. Appl. Pharmacol. 88, 354. Eskandari, M.IC, Nguyen, D.T., Kunkel, S.L. and Remick, D.G. (1990) WEHI 164 subclone 13 assay for TNF: sensitivity, specificity, and reliability. Immunol. Invest. 19, 69.
Griffiths-Johnson, D., Ryan, L.K. and Karol, M.H. (1991) Development of an animal model for organic dust toxic syndrome. Inhal. Toxicol. 3, 405. Kunkcl, S.L., Remick, D.G., Streiter, R.M. and Larrick, J.W. (1989) Mechanisms that regulate the production and effects of tumor necrosis factor-alpha. Crit. Rev. lmmunol. 9, 93. Prausnitz, C. (1936) Investigations on respiratory dust disease in operatives in cotton industry. Medical Research Council, Spec. Rep. Ser. No. 212. His Majesty's Stationery Office, London, p. 60. Ryan, L.K. and Karol, M.H. (1991) Release of tumor necrosis factor in guinea pigs upon acute inhalation of cotton dust. Am. J. Respir. Cell. Mol. Biol. 5, 93. Rylander, R. (1990) Health effects of cotton dust exposures. Am. L Ind. Med. 17, 39. Sheehan, K.C.F., Ruddle, N.H. and Schreiber, R.D. (1989) Generation and characterization of hamster monoclonal antibodies that neutralize murine tumor necrosis factors. J. lmmunol. 142, 3884. Tamatani, T., Kimura, S., Hashimoto, T. and Onozaki, K. (1989) Purification of guinea pig tumor necrosis factor (TNF): comparison of its anti-proliferative and differentiatire activities for myeloid leukemic cell lines with those of recombinant human TNF. J. Biochem. 105, 55.
