Veterinary Immunology and Immunopathology, 27 ( 1991 ) 337-350
337
Elsevier Science Publishers B.V., Amsterdam
Interaction of bovine neutrophils in Pasteurella haemolytica mediated damage to pulmonary endothelial cells M.A. Breider a, Sarmishtha K u m a r ~ and Richard E. Corstvet b "Department of Pathobiology, Collegeof VeterinaryMedicine, University of Tennessee, Knoxville, TN 37901, USA bDepartment of VeterinaryMicrobiology and Parasitology, Louisiana State University, Baton Rouge, LA 708O3, USA (Accepted 23 April 1990)
ABSTRACT Breider, M.A., Kumar, S. and Corstvet, R.E., 199 I. Interaction of bovine neutrophils in Pasteurella haemolytica mediated damage to pulmonary endothelial cells. Vet. Immunol. Immunopatho/., 27: 337-350. The purpose of these studies was to determine mechanisms of pulmonary tissue damage mediated by Pasteurella haemolytica and interaction with bovine neutrophils. Bovine pulmonary artery endothelial cell monolayers were treated with various combinations of P. haemolytica factors including bacterial culture supernatant (CS) and purified LPS, with and v~ithout bovine neutrophils. Damage to endothelial cells was monitored by 5~Cr release, cell detachment rate, and morphological changes. At
~ h pa©t_tr~atmant
I t D T ~ k o r . l r a r ~ , ' J l 'F,'Jr.*a,.a n . - , ' ~ , ' i , . , - . a r l , , a . ~, I ~ . ' l - l a .I.,,~i,-. ,,I.~,~.-,,-,,~ .i..~ ,-.,~ll~ I~ . . . . . . . . .
k.,
")*~
h PT both crude leukotoxin and LPS caused high levels of cytotoxicity and detachment. Neutrophils did not augment toxicity mediated by LPS, but actually protected endothelial cells from low levels of LPS. When the LPS component of CS was neutralized with polymyxin B, leukotoxin mediated neutrophil killing resulted in extensive endothelial cell damage. These results suggest that LPS may directly injure endothelial cells and this toxic effect may be reduced by neutrophils. However, neutrophil killing by leukotoxin may also contribute to endothelial cell damage in the absence of LPS.
INTRODUCTION
Bovine respiratory disease complex is one of the most important problems affecting cattle populations. The most common bacterial agent associated with bovine respiratory disease complex is Pasteurella haemolytica, biotype A, serotype l (Frank, ! 982). Upon colonizing the lung, P. haemolytica produces an acute fulminating lobar fibrinous pneumonia (Rehmtulla and Thomson, 1981 ). As early as 5 h post-inoculation, extensive alveolar and intedobular edema, fibrinous exudate, neutrophil (PMN) infiltrates, and hemorrhage occur, suggesting en0165-2427/91/$03.50
© 1991 - - Elsevier Science Publishers B.V.
338
M.A. BREIDERET AL.
dothelial cell (EC) damage as a primary pathogenic event (Slocombe et al., 1985; Breider et al., 1988). Within 24-48 h lesions progress to vascular thrombosis, parenchymal necrosis, and eventual abscess formation (Rehmtulla and Thomson, 1981 ). Several virulence factors produced by P. haemolytica are suggested to be important in pathogenesis. P. haemolytica produces a heat labile exotoxin, termed leukotoxin (LT), that causes rapid death and lysis of ruminant leukocytes (Berggren et al., 1981 ). It has been suggested that LT killing of neutrophils in the lung may damage the microvasculature leading to edema, hemorrhage, and fibrinous exudate (Slocombe et al., 1985). However, the role of PMN in pathogenesis of the disease is not clear. One study demonstrated only slight pulmonary lesions in neutropenic calves inoculated with P. haemolytica (Slocombe et al., 1985), whereas another study demonstrated that P. haernolytica can produce significant alveolar edema, hemorrhage, and tissue damage in both PMN deficient and PMN sufficient calves (Breider et al., 1988). Bacterial lipopolysaccharide (LPS) is also suggested to be important in ruminant pulmonary tissue damage (Meyrick and Bringham, 1983). Bacterial LPS can directly damage bovine EC in vitro (Meyrick, 1976). Bovine leukocyte function can also be modulated with P. haemolytica LPS, however, the modulation is dependent on LPS concentration (Confer and Simons, 1986). The purpose of these investigations was to determine if PMN contribute to pulmonary EC damage following LT killing of PMN or modulation with LPS. Bovine pulmonary EC monolayers were treated with various combinations of bovine PMN, bacterial CS, and P. haemolytica LPS. The cells were monitored for damage by 5~Crrelease, cell detachment rate, and morphological changes. Our results demonstrate that PMN reduce EC damage caused by low levels of P. haemolytica LPS. However, when LPS is neutralized by polymyxin B, LT mediated killing of PMN leads to significant damage of EC. MATERIALS A N D M E T H O D S
Endothelial cell cultures Primary EC cultures were established from pulmonary arteries obtained from a local abattoir, using established techniques (Ryan and Maxwell, 1986 ). Briefly the arterial intimal surface was gently scraped with a scalpel aad the scraped cells were suspended in Dulbecco's Modified Eagle's Medium (DMEM) (Whittaker MA Bioproducts, WalkersviUe, MD) with 10% heatinactivated Ryans growth supplement (HI-RGS) (Dr. Una Ryan, U. of Miami, Miami, FL). The cells were grown in tissue culture oishes for 24-48 h and individual EC colonies were isolated using glass cloning rings (Bellco Glass, Inc., Vineland, N J) and 0.05% trypsin (Gibco Laboratories, Grand
NEUTROPHILS AND P. HAEMOLYTICA MEDIATED ENDOTHELIAL INJURY
339
Island, NY). The cell clones were expanded in tissue culture flasks and identified as EC by typical cobblestone morphology and the presence of anti-factor VIII antigen, using described methods (Ryan, 1986). The EC used in all experiments were of the same cell line and less than passage 20.
Bacterialfactors P. haemolytica culture supernatant (CS) was prepared using described techniques to optimize the production of P. haemolytica leukotoxin (Shewen and Wilkie, 1982). Briefly, P. haemolytica (biotype A, serotype 1 ) w a s grown on enriched blood agar for 18 h at 37 ° C followed by transfer of 7-10 bacterial colonies to brain heart infusion broth with 5.0% heat-inactivated fetal bovine serum (HI-FBS). Following a 4.5 h incubation at 37°C, the culture was centrifuged at l 0 000 g for 20 min and the supernatant was discarded. The bacterial pellet was reconstituted in RPMI- 1640 with 10% HI-FBS, incubated for 90 rain at 37 °C, and centrifuged at l 0 000 g for 20 rain. The supernatant was saved and filtered through a 0.22 micron filter (Costar, Cambridge, MA), aliquoted, and stored at - 70°C. The LT activity of the CS was determined to be 6855 toxic units/cc using a previously described neutral red uptake cytotoxicity assay (Greer and Shewen, 1986) of bovine lymphoma cells (obtained from R. Theilen, University of California, Davis, CA). The leukotoxic activity of a 10- l dilution of CS was eliminated using either heat denaturation (56 °C for 30 rain) or incubation for 60 rain at 37°F with LT neutralizing monoclonal antibody. The murine monoclonal leukotoxin neutralizing antibody was produced ustibody was characterized as IgM. Ascites fluid from mice, implanted with the LT-neutralizing hybridoma clone, was used in the described experiments. The neutralization of the L i" with heat-inactivation or monoclonal antibody was verified using the neutral red cytotoxicity assay and also a previously described PMN 51Cr release cytotoxicity assay (Mosier et al., 1986). The P. haemolytica LPS was extracted from bacteria using a phenol-water e~:~:'action method (McIntire et al., 1967) and was obtained from Robert Moore (Department of Microbiology, University of Tennessee, Knoxville, TN). The ¢ndotoxin activity of the LPS preparation and CS were quantitated using a chromogenic Limulus amebocyte lysate assay (Whittaker MA Bioproduct, Walkersville, MD). An E. coli endotoxin standard was used to standardize the levels of endotoxin in all bacterial products. The CS contained l04 cndotoxin units ( E U ) / m l and the P. haemolytica LPS contained 2 X 104 EU/~g. To neutralize the endotoxic activity of the CS or LPS, preparations were incubated for 60 min at 37°C with 1000 units/cc polymyxin B (Sigma Chemicals, St. Louis, MO ).
340
M.A. BREIDER ET AL.
Bovine neutrophils Neutrophils were isolated from the peripheral blood of an adult Charlois Hereford cross steer as previously described (Roth and Kaebede, 198 l a). The cell viability, determined with trypan blue, was > 98%. The PMN were used immediately in experiments following isolation.
Endothelial cell cytotoxicity assay Endothelial cell damage was quantitated using a previously described 51Cr release assay (Marten, 1984). Briefly, 105 bovine EC in l cc of DMEM- 10% RGS were plated into 24 well tissue culture plates. Upon reaching confluency in 24 h, the cells were labeled with l cc ofStCr ( 10/zCi/ml) of DMEM, 10% RGS) (ICN, Irvine, CA) for 24 h. The cells were then washed five times with Hank's buffered salt solution (HBSS) and variables added diluted in DMEM10% HI-FBS. Variables were tested ~.n quadruplets in each experiment and experiments were repeated to confirm results. At post-treatment (PT) times of 5 and 22 h, the plates were centrifuged and 200/ll of media was removed and measured for 5~Cr release using a gamma counter. The cytotoxic index for each variable was determined using the formula C i = (A-B/C-B)100 where A = variable release, B =spontaneous release, and C = total release.
Cellular detachment assay As a more sensitive indicator of EC damage, a previously described detachment assay (Vandenbrouche-Grauls et al., 1987) was also used to quantitate the degree of EC damage following administration of variables. The same experimental culture wells, as described for the 51Cr release assay were used. W,n, l l ~ u J , l.no. r ~ , , r n ~ , ~ ! r,,t:"l'h,~ 9 / ) / ) ,,I ~¢.,.n,=,A.;,-J ~r,,r 'l'l~,~ ,.-'.'~rl-r,,'l-~v;r,;'l-.~J, I,J©e,'J'~1 'l-h,~ lv~v ,~~, Al ll teJ ~L I , T A J L ~ VV J ~ . A ~ & ~ & I J L ~ , S • f~t,l ql.TA ~ J . J L ~ ~,-,~tJq~ jL~A ~ J i . AAAqv~JL~.~L JL%TA K.AJL~ ~ ' J L I , T L ~ 3 t . J L ~ J L L J A,~lI~ff , ll,l i . l . t
were washed three times with HBSS and treated with 500/zl of 1.0% Triton X-100. The total content from each well was then measured for 51Cr. The percent of cellular detachment was computed by the following formula: % detachment = 100 (control- variable )/contreL
Morphological observations At 5 and 22 h PT the cell monolayers were observed using an Olympus CK2 inverted phase contrast microscope.
Protocols To determine ifLT mediated PMN killing damaged EC, endothelial monolayers were treated with either CS, CS and PMN (106), heat-inactivated CS (HI-CS), HI-CS with PMN, CS preincubated 50 rain at 37°C with LT-neutralizing monoclonal antibody (CS-MAb), and CS-MAb with PMN. In additional experiments the LPS component of CS was neutralized by a 60 min 37°C incubation with polymyxin B (1000 units/ml) before adding to cell monolayers.
341
NEUTROPHILS AND P. HAEMOLYTICA MEDIATED ENDOTHELIAL INJURY
A dose response of various dilutions of P. haemolytica LPS and CS was done with a constant number of PMN to determine if LPS: PMN ratio affected toxicity to EC. To determine ifPMN concentration affects EC damage, varying concentrations of PMN were added to monolayers simultaneously with CS. Similar PMN populations were also stimulated with opsonized zymosan granules (Roth and Kaeberle, 1981 b ) or phorbal myristate acetate ( 100 nM/ml media) (Sigma Chemical Co., St. Louis, MO) as previously described (Dobrina and Patriarca, 1986) to determine if stimulated bovine P MN damage EC in our system. Various dilutions of CS ( 10- ~, 10- 2, and l 0- 3) were preincubated at 60 min at 37°C with 1 × 10 6 cells/ml of either PMN, bovine turbinate epithelium (BTU) (provided by Dr. Leon Potgieter, Knoxville, TN), and bovine EC to determine if cells other than PMN diminished the LPS mediated EC damage. After a 37°C 60 min incubation the cells were removed from the CS and the supernatants were applied to a EC cytotoxic assay. In one treatment group, the PMN were retained in the CS following preincubation and applied to EC monolayer, to determine if the continued presence of PMN was necessary to mediate protection. The S~Cr release of the EC were monitored at 22 h PT.
Statistical analysis The cytotoxic index and percent cellular detachment for different variables were compared using Duncan's multiple range test or Student's t test to determine significant differences (P < 0.05 ) between groups.
60]
..-c.-
A ,oO.,o
50
B
g so
- -
(3
'i
a
b
c
d
e
f
g
h
a
b
c
d
e
f
g
h
Fig. 1. Cytotoxicity, based on (A) 5~Cr release and (B) detachment rate, of bovine endothelial cells 22 h following treatment with: (a) P. haemolytica culture supernatant (CS); (b) heatinactivated CS (HI-CS); (c) CS and neutrophiis: (d) HI-CS and neutrophils; (e) CS preincubated with anti-leukotoxin monoclonal antibody: (f) HI-CS and monoclonal antibody; (g) CS, monoclonal antibody and neutrophils, and (h) HI-CS preincubated with monoclonal antibody and neutrophils. Data reflect the mean and SEM of two separate experiments and a minimum of eight experimental wells. There was a significant difference between cytotoxic index values of a, b, c, d and e, g, h; and between mean detachment rate values of a, b, c, d and e, f, g, h (P
337
Elsevier Science Publishers B.V., Amsterdam
Interaction of bovine neutrophils in Pasteurella haemolytica mediated damage to pulmonary endothelial cells M.A. Breider a, Sarmishtha K u m a r ~ and Richard E. Corstvet b "Department of Pathobiology, Collegeof VeterinaryMedicine, University of Tennessee, Knoxville, TN 37901, USA bDepartment of VeterinaryMicrobiology and Parasitology, Louisiana State University, Baton Rouge, LA 708O3, USA (Accepted 23 April 1990)
ABSTRACT Breider, M.A., Kumar, S. and Corstvet, R.E., 199 I. Interaction of bovine neutrophils in Pasteurella haemolytica mediated damage to pulmonary endothelial cells. Vet. Immunol. Immunopatho/., 27: 337-350. The purpose of these studies was to determine mechanisms of pulmonary tissue damage mediated by Pasteurella haemolytica and interaction with bovine neutrophils. Bovine pulmonary artery endothelial cell monolayers were treated with various combinations of P. haemolytica factors including bacterial culture supernatant (CS) and purified LPS, with and v~ithout bovine neutrophils. Damage to endothelial cells was monitored by 5~Cr release, cell detachment rate, and morphological changes. At
~ h pa©t_tr~atmant
I t D T ~ k o r . l r a r ~ , ' J l 'F,'Jr.*a,.a n . - , ' ~ , ' i , . , - . a r l , , a . ~, I ~ . ' l - l a .I.,,~i,-. ,,I.~,~.-,,-,,~ .i..~ ,-.,~ll~ I~ . . . . . . . . .
k.,
")*~
h PT both crude leukotoxin and LPS caused high levels of cytotoxicity and detachment. Neutrophils did not augment toxicity mediated by LPS, but actually protected endothelial cells from low levels of LPS. When the LPS component of CS was neutralized with polymyxin B, leukotoxin mediated neutrophil killing resulted in extensive endothelial cell damage. These results suggest that LPS may directly injure endothelial cells and this toxic effect may be reduced by neutrophils. However, neutrophil killing by leukotoxin may also contribute to endothelial cell damage in the absence of LPS.
INTRODUCTION
Bovine respiratory disease complex is one of the most important problems affecting cattle populations. The most common bacterial agent associated with bovine respiratory disease complex is Pasteurella haemolytica, biotype A, serotype l (Frank, ! 982). Upon colonizing the lung, P. haemolytica produces an acute fulminating lobar fibrinous pneumonia (Rehmtulla and Thomson, 1981 ). As early as 5 h post-inoculation, extensive alveolar and intedobular edema, fibrinous exudate, neutrophil (PMN) infiltrates, and hemorrhage occur, suggesting en0165-2427/91/$03.50
© 1991 - - Elsevier Science Publishers B.V.
338
M.A. BREIDERET AL.
dothelial cell (EC) damage as a primary pathogenic event (Slocombe et al., 1985; Breider et al., 1988). Within 24-48 h lesions progress to vascular thrombosis, parenchymal necrosis, and eventual abscess formation (Rehmtulla and Thomson, 1981 ). Several virulence factors produced by P. haemolytica are suggested to be important in pathogenesis. P. haemolytica produces a heat labile exotoxin, termed leukotoxin (LT), that causes rapid death and lysis of ruminant leukocytes (Berggren et al., 1981 ). It has been suggested that LT killing of neutrophils in the lung may damage the microvasculature leading to edema, hemorrhage, and fibrinous exudate (Slocombe et al., 1985). However, the role of PMN in pathogenesis of the disease is not clear. One study demonstrated only slight pulmonary lesions in neutropenic calves inoculated with P. haemolytica (Slocombe et al., 1985), whereas another study demonstrated that P. haernolytica can produce significant alveolar edema, hemorrhage, and tissue damage in both PMN deficient and PMN sufficient calves (Breider et al., 1988). Bacterial lipopolysaccharide (LPS) is also suggested to be important in ruminant pulmonary tissue damage (Meyrick and Bringham, 1983). Bacterial LPS can directly damage bovine EC in vitro (Meyrick, 1976). Bovine leukocyte function can also be modulated with P. haemolytica LPS, however, the modulation is dependent on LPS concentration (Confer and Simons, 1986). The purpose of these investigations was to determine if PMN contribute to pulmonary EC damage following LT killing of PMN or modulation with LPS. Bovine pulmonary EC monolayers were treated with various combinations of bovine PMN, bacterial CS, and P. haemolytica LPS. The cells were monitored for damage by 5~Crrelease, cell detachment rate, and morphological changes. Our results demonstrate that PMN reduce EC damage caused by low levels of P. haemolytica LPS. However, when LPS is neutralized by polymyxin B, LT mediated killing of PMN leads to significant damage of EC. MATERIALS A N D M E T H O D S
Endothelial cell cultures Primary EC cultures were established from pulmonary arteries obtained from a local abattoir, using established techniques (Ryan and Maxwell, 1986 ). Briefly the arterial intimal surface was gently scraped with a scalpel aad the scraped cells were suspended in Dulbecco's Modified Eagle's Medium (DMEM) (Whittaker MA Bioproducts, WalkersviUe, MD) with 10% heatinactivated Ryans growth supplement (HI-RGS) (Dr. Una Ryan, U. of Miami, Miami, FL). The cells were grown in tissue culture oishes for 24-48 h and individual EC colonies were isolated using glass cloning rings (Bellco Glass, Inc., Vineland, N J) and 0.05% trypsin (Gibco Laboratories, Grand
NEUTROPHILS AND P. HAEMOLYTICA MEDIATED ENDOTHELIAL INJURY
339
Island, NY). The cell clones were expanded in tissue culture flasks and identified as EC by typical cobblestone morphology and the presence of anti-factor VIII antigen, using described methods (Ryan, 1986). The EC used in all experiments were of the same cell line and less than passage 20.
Bacterialfactors P. haemolytica culture supernatant (CS) was prepared using described techniques to optimize the production of P. haemolytica leukotoxin (Shewen and Wilkie, 1982). Briefly, P. haemolytica (biotype A, serotype 1 ) w a s grown on enriched blood agar for 18 h at 37 ° C followed by transfer of 7-10 bacterial colonies to brain heart infusion broth with 5.0% heat-inactivated fetal bovine serum (HI-FBS). Following a 4.5 h incubation at 37°C, the culture was centrifuged at l 0 000 g for 20 min and the supernatant was discarded. The bacterial pellet was reconstituted in RPMI- 1640 with 10% HI-FBS, incubated for 90 rain at 37 °C, and centrifuged at l 0 000 g for 20 rain. The supernatant was saved and filtered through a 0.22 micron filter (Costar, Cambridge, MA), aliquoted, and stored at - 70°C. The LT activity of the CS was determined to be 6855 toxic units/cc using a previously described neutral red uptake cytotoxicity assay (Greer and Shewen, 1986) of bovine lymphoma cells (obtained from R. Theilen, University of California, Davis, CA). The leukotoxic activity of a 10- l dilution of CS was eliminated using either heat denaturation (56 °C for 30 rain) or incubation for 60 rain at 37°F with LT neutralizing monoclonal antibody. The murine monoclonal leukotoxin neutralizing antibody was produced ustibody was characterized as IgM. Ascites fluid from mice, implanted with the LT-neutralizing hybridoma clone, was used in the described experiments. The neutralization of the L i" with heat-inactivation or monoclonal antibody was verified using the neutral red cytotoxicity assay and also a previously described PMN 51Cr release cytotoxicity assay (Mosier et al., 1986). The P. haemolytica LPS was extracted from bacteria using a phenol-water e~:~:'action method (McIntire et al., 1967) and was obtained from Robert Moore (Department of Microbiology, University of Tennessee, Knoxville, TN). The ¢ndotoxin activity of the LPS preparation and CS were quantitated using a chromogenic Limulus amebocyte lysate assay (Whittaker MA Bioproduct, Walkersville, MD). An E. coli endotoxin standard was used to standardize the levels of endotoxin in all bacterial products. The CS contained l04 cndotoxin units ( E U ) / m l and the P. haemolytica LPS contained 2 X 104 EU/~g. To neutralize the endotoxic activity of the CS or LPS, preparations were incubated for 60 min at 37°C with 1000 units/cc polymyxin B (Sigma Chemicals, St. Louis, MO ).
340
M.A. BREIDER ET AL.
Bovine neutrophils Neutrophils were isolated from the peripheral blood of an adult Charlois Hereford cross steer as previously described (Roth and Kaebede, 198 l a). The cell viability, determined with trypan blue, was > 98%. The PMN were used immediately in experiments following isolation.
Endothelial cell cytotoxicity assay Endothelial cell damage was quantitated using a previously described 51Cr release assay (Marten, 1984). Briefly, 105 bovine EC in l cc of DMEM- 10% RGS were plated into 24 well tissue culture plates. Upon reaching confluency in 24 h, the cells were labeled with l cc ofStCr ( 10/zCi/ml) of DMEM, 10% RGS) (ICN, Irvine, CA) for 24 h. The cells were then washed five times with Hank's buffered salt solution (HBSS) and variables added diluted in DMEM10% HI-FBS. Variables were tested ~.n quadruplets in each experiment and experiments were repeated to confirm results. At post-treatment (PT) times of 5 and 22 h, the plates were centrifuged and 200/ll of media was removed and measured for 5~Cr release using a gamma counter. The cytotoxic index for each variable was determined using the formula C i = (A-B/C-B)100 where A = variable release, B =spontaneous release, and C = total release.
Cellular detachment assay As a more sensitive indicator of EC damage, a previously described detachment assay (Vandenbrouche-Grauls et al., 1987) was also used to quantitate the degree of EC damage following administration of variables. The same experimental culture wells, as described for the 51Cr release assay were used. W,n, l l ~ u J , l.no. r ~ , , r n ~ , ~ ! r,,t:"l'h,~ 9 / ) / ) ,,I ~¢.,.n,=,A.;,-J ~r,,r 'l'l~,~ ,.-'.'~rl-r,,'l-~v;r,;'l-.~J, I,J©e,'J'~1 'l-h,~ lv~v ,~~, Al ll teJ ~L I , T A J L ~ VV J ~ . A ~ & ~ & I J L ~ , S • f~t,l ql.TA ~ J . J L ~ ~,-,~tJq~ jL~A ~ J i . AAAqv~JL~.~L JL%TA K.AJL~ ~ ' J L I , T L ~ 3 t . J L ~ J L L J A,~lI~ff , ll,l i . l . t
were washed three times with HBSS and treated with 500/zl of 1.0% Triton X-100. The total content from each well was then measured for 51Cr. The percent of cellular detachment was computed by the following formula: % detachment = 100 (control- variable )/contreL
Morphological observations At 5 and 22 h PT the cell monolayers were observed using an Olympus CK2 inverted phase contrast microscope.
Protocols To determine ifLT mediated PMN killing damaged EC, endothelial monolayers were treated with either CS, CS and PMN (106), heat-inactivated CS (HI-CS), HI-CS with PMN, CS preincubated 50 rain at 37°C with LT-neutralizing monoclonal antibody (CS-MAb), and CS-MAb with PMN. In additional experiments the LPS component of CS was neutralized by a 60 min 37°C incubation with polymyxin B (1000 units/ml) before adding to cell monolayers.
341
NEUTROPHILS AND P. HAEMOLYTICA MEDIATED ENDOTHELIAL INJURY
A dose response of various dilutions of P. haemolytica LPS and CS was done with a constant number of PMN to determine if LPS: PMN ratio affected toxicity to EC. To determine ifPMN concentration affects EC damage, varying concentrations of PMN were added to monolayers simultaneously with CS. Similar PMN populations were also stimulated with opsonized zymosan granules (Roth and Kaeberle, 1981 b ) or phorbal myristate acetate ( 100 nM/ml media) (Sigma Chemical Co., St. Louis, MO) as previously described (Dobrina and Patriarca, 1986) to determine if stimulated bovine P MN damage EC in our system. Various dilutions of CS ( 10- ~, 10- 2, and l 0- 3) were preincubated at 60 min at 37°C with 1 × 10 6 cells/ml of either PMN, bovine turbinate epithelium (BTU) (provided by Dr. Leon Potgieter, Knoxville, TN), and bovine EC to determine if cells other than PMN diminished the LPS mediated EC damage. After a 37°C 60 min incubation the cells were removed from the CS and the supernatants were applied to a EC cytotoxic assay. In one treatment group, the PMN were retained in the CS following preincubation and applied to EC monolayer, to determine if the continued presence of PMN was necessary to mediate protection. The S~Cr release of the EC were monitored at 22 h PT.
Statistical analysis The cytotoxic index and percent cellular detachment for different variables were compared using Duncan's multiple range test or Student's t test to determine significant differences (P < 0.05 ) between groups.
60]
..-c.-
A ,oO.,o
50
B
g so
- -
(3
'i
a
b
c
d
e
f
g
h
a
b
c
d
e
f
g
h
Fig. 1. Cytotoxicity, based on (A) 5~Cr release and (B) detachment rate, of bovine endothelial cells 22 h following treatment with: (a) P. haemolytica culture supernatant (CS); (b) heatinactivated CS (HI-CS); (c) CS and neutrophiis: (d) HI-CS and neutrophils; (e) CS preincubated with anti-leukotoxin monoclonal antibody: (f) HI-CS and monoclonal antibody; (g) CS, monoclonal antibody and neutrophils, and (h) HI-CS preincubated with monoclonal antibody and neutrophils. Data reflect the mean and SEM of two separate experiments and a minimum of eight experimental wells. There was a significant difference between cytotoxic index values of a, b, c, d and e, g, h; and between mean detachment rate values of a, b, c, d and e, f, g, h (P
