Pulmonary Pharmacology (1990) 3 171-179 01990 Longman Group UK Ltd
0952-0600/90/0003-0171/510 .00
PULMONARY PHARMACOLOGY
Antigen-Dependent Leukotriene Synthesis and Histamine Release from IgG1 Passively-Sensitized Guinea Pig Lungs Ex Vivo : Relationship Between Serum Levels of Antigen-Specific IgG1 and Mediator Synthesis/Release J . B . Cheng, J . S . Pillar, M . J . Conklyn, R . Breslow, J. T . Shirley, N . P. Gerard*, H . J . Showell Department of Immunology and Infectious Diseases, Central Research Division, Pfizer Inc ., Groton, Connecticut (USA) ; *Harvard Medical School, Beth Israel Hospital, Boston, Massachusetts (USA) SUMMARY . Naive guinea-pigs were passively sensitized with varying amounts of affinity column purified, homologous, anti-ovalbumin IgG1 (anti-OA IgG1) and then examined for a) the capacity of lung tissue to release mediators (histamine and LTB4 /LTD4 ) in response to antigen-challenge ex vivo and b) the attendant circulating levels of anti-OA IgG1 . Intraperitoneal administration of anti-OA IgG1 (0 .125-0.75 mg/kg) to guinea-pigs facilitated the synthesis of LTB 4 (8-25 ng/g lung) and LTD 4 (18-80 ng/g) and the release of histamine (1-7 ug/g) from lung tissue after exposure to 10 pg/ml of ovalbumin for 20 min ex vivo. Peak levels of mediators were found using 0 .5 mg/kg anti-OA IgG1 with an ED so =0 .35 mg/kg. LTD4 /LTB4 synthesis and histamine release were both antigen concentration- and time-dependent, and LT synthesis was observable in non-perfused lungs and in lungs perfused free of blood . Maximum sensitization occurred at 1-2 days post i .p. administration of anti-OA IgG1 and was maintained up to 7 days . Measurement of anti-OA IgG1 using an enzyme-linked immunosorbent assay demonstrated that circulating antibody levels were 2-6 µg/ml at the doses which caused sensitization . The level of anti-OA IgG1 found in passively sensitized animals was at least 100-fold less than that found in actively-sensitized guinea-pigs despite the similar magnitude in LTD4 /LTB4 synthesized and the amount of histamine released . Using purified antibody, the results demonstrate that in guinea-pigs, IgG1 can play a prominent role in regulating lung LT synthesis and histamine release, and that microgram per ml circulating levels of this antibody are sufficient to sensitize naive lungs .
ine from lungs of IgG1-sensitized guinea-pigs than those from IgE-sensitized lungs . Also, in vitro studies''' demonstrated that SRS-A was involved in the contraction of the IgG1-sensitized guinea-pig tracheal and parenchymal preparations but not in the IgE-sensitized tissue . These results continue to support the idea that IgG1 antibody plays an important role in immunologically regulating mediator production in guinea-pigs . Since mediator release/production from cells is linked to receptor occupancy, at equilibrium the level of occupancy should be dependent on the circulating level of antibody (up to the point of saturation) . Determination of the level of antibody should help us to understand the quantity of circulating IgG1 necessary for sensitization . However, there is little information in this regard . Moreover, a relationship between the circulating level of antigen-specific IgG1 and amount of mediators released from lung tissue remains to be demonstrated . An OA-specific homologous IgGl antibody has recently been isolated from hyperimmune sera of actively-sensitized guinea pigs . The antibody has been purified to homogeneity by chromatography .' Utiliz-
INTRODUCTION Lungs taken from actively- and passively-sensitized guinea-pigs synthesize leukotrienes (LTs) and release histamine upon antigen challenge ex vivo . This source of sensitized tissue has been used as a model system to study biochemical pathways of mediator generation and in addition how these pathways can be manipulated pharmacologically ." Recently, it became appreciated that the quantity of LTs generated could be affected by immunization protocols . Lung fragments of guinea-pigs injected with ovalbumin (OA) four times have been shown to produce more LTD 4 after antigen challenge than lung fragments of single OA injected animals .' In addition, both qualitatively and quantitatively mediator synthesis/release appear to vary depending upon whether IgG1 or IgE has been used to sensitize tissue . For example, Undem et al.' demonstrated substantially more release of slow-reacting substance of anaphylaxis (SRS-A) and histamCorrespondence to : Dr John B . Cheng, Department of Immunology and Infectious Diseases, Central Research Division, Pfizer Inc., Groton, CT 06340. 171
1 72 Pulmonary Pharmacology
ing purified IgG1 should permit a detailed examination of antigen-dependent mediator release/synthesis from passively-sensitized lungs . Moreover, an antigen-specific (anti-OA) IgG1 developed in the laboratory, should allow a quantitation of circulating levels of IgG1 . The information relating serum levels of IgG1 to antigen-induced response should also allow a determination of the in vivo potency of this antibody to sensitize lung tissues . A study was undertaken, therefore, to 1 . characterize OA-dependent LT synthesis and histamine release from IgG1 passively-sensitized lungs, 2 . determine the relationship between mediator production and circulating IgG1 levels in sensitized animals, and 3 . compare these findings with those obtained from lungs of actively-sensitized guinea-pigs .
3 mM) . SDS-PAGE electrophoresis analysis of the eluted sample revealed > 95% purity of the antibody (MW = 146 000) with no contamination with IgG2 antibody .' A passive cutaneous anaphylaxis (PCA) test in guinea-pigs .' 1-13 confirmed that the purified IgG1 antibody was heat-stable at 56°C for 4 h and persisted in the skin site for no more than 5 days . Development of a new enzyme-linked immunosorbent assay (ELISA) for the measurement of serum antiOA IgG1 In order to determine the circulating levels of anti-OA IgG1, a sensitive and antigen-specific ELISA was developed .' t-13 Table 1 shows the protocol for the ELISA . Guinea pig immunization and the chopped lung assay
METHODS Isolation and purification of an OA-specific IgG1 antibody from the serum of actively-sensitized guinea pigs
Male Hartley guinea-pigs (Charles River Laboratories, Wilmington, MA) were immunized with OA precipitated onto alum as described previously ."° The pooled hyperimmune sera was the source for the isolation of anti-OA IgG1 used in this study . As described previously' (procedures also shown in Fig. 1), IgG1 was separated from IgG2 by passing the sera through a Protein A-Sepharose 413 column and then isolated using an OA-linked Sepharose 4B column . The isolated sample was re-applied to the Protein A column, and the OA-specific IgG1 was eluted from the column using a pH gradient of 6 .0-5 .0 (glycine,
Passive immunization Male Hartley guinea pigs (age : 3-4 weeks) were sensitized by a single intraperitoneal injection of the purified anti-OA antibody (0 .175-0 .75 mg/kg) . Guinea-pigs were sacrificed at 48 h post injection unless otherwise specified . Active immunization This was performed according to the multiple antigen injection protocol described previously except that 5-8 day-old guinea pigs were used . 5,14 When a comparison of antigen-dependent mediator generation from passively- and actively-sensitized lungs was made, the final age of both groups was similar (3-5 weeks) . Chopped lung assay Guinea-pigs were anesthetized with pentobarbital (40-60 mg/kg) and exsanguinated by cutting the ab-
Pooled Antiserum
a
G
Protein A - Sepharose CL-4B
IgG2
a Ovalbumin - Sepharose 4B
G
IgG1 + IgG2
P a Effluent
a Protein A - Sepharose 4B CH pH
I
Effluent
i IgG1 P = Phosphate Buffered Saline, pH 7 .4 G = OA M Glycine, pH 3 .0 CP = Citrate Phosphate Buffer, pH 7 .4 pH = Citrate Phosphate Buffer, pH 6 .0 - 5 .0 Fig. 1-Isolation protocol for guinea-pig anti-OA IgG1 .
Mediator Release and Serum IgGI
1 73
Table 1 . Anti-ovalbumin IgGI ELISA Add 100 µl ovalbumin (0 .1 mg/ml carbonate buffer, pH 9 .6) per well (Dynatech Lab, Virginia) 2 . Blocking : Wash 2 x 200 Ed carbonate buffer Add 200 µl 1% BSA in carbonate buffer, 1 hr incubation, room temperature (RT) 3 . GP IgG1 binding Wash I x 200 µl carbonate buffer Wash 2 x 200 p1 PBS-T, pH 9 .8 Add 100 pl Sample/Standard in PBS-T, pH 9 .8, 1 hr incubation, RT 4. Rabbit Anti-GP IgG1 Wash 2 x 200 N1 PBS-T, pH 9.8 (biotinylated) Add 100 µl rabbit anti-GP IgG1 Binding : (biotinylated), 1 hr incubation, RT 5 . Complex binding: Wash 1 x 200 gl PBS-T, pH 9 .8 Wash 2 x 200 pl PBS-T, pH 7 .4 Add 100 gl complex solution* 30 min incubation, RT Wash 2 x 200 pl PBS-T, pH 7 .4 6. Development : Wash 1 x 200 p1 DEA buffer Add 100 pl nitrophenyl phosphate solution (1 mg/ml DEA), 10 min incubation, RT Add 25 µl 5N KOH Read OD410 *Complex formation : Combine 40 pl Streptavadin (1 mg/ml) and 40 µl biotinylated-alkaline phosphatase (0 .5 mg/ml) in 10 ml PBS-T, pH 7 .4, 30 min incubation, RT prior to use . 1 . Antigen binding :
Carbonate 1 .59 g Na2 C0 3 2 .93 g NaHCO 3 1000 ml H 20 adjust pH to 9 .6
PBS-T 8 .0 g NaCl 0 .2 g KH 2 P0 4 1 .15 g Na2 HPO 4 0 .2 g KCI 0 .5 ml Tween 20 1000 ml H20 adjust to pH 7 .4 or 9 .8
DEA 4.85 ml diethanolamine 50 mg MgC1 2 .6H 2 0 500 ml H2 0 adjust pH to 9 .8
dominal aorta and inferior vena cava . The apex of the heart was removed to ensure free flow of the perfusate before the lung was perfused via the pulmonary artery with 10-20 ml of cold Krebs-Ringer medium (see below) . The perfused lung was removed quickly . The bronchi and large bronchioles were trimmed away and the lung parenchyma was placed on a Mcllwain Chopper and cut into approximately 1 mm 3 tissue fragments . These were then washed five times with cold medium and weighed into reaction tubes containing 2 .5 ml of the medium and held on ice until assayed . The medium used through the procedure contained (millimolar) : NaCl, 117 .5 ; KCl, 5 .37 ; MgS04 .7H 2 0, 2 .52; CaC1 2 , 2.52 ; NaHCo 3 , 15 .51 ; NaH 2 PO4 , 1 .17 ; glucose, 5 .50 ; and sucrose, 13 .65 and was gassed with 95% O Z -5% CO 2 to bring the pH to 7.4 (25°C) . All lung fragments were prewarmed at 37 °C for 20 min in the presence of 1 gM propranolol' before OA challenge . Except where noted in the text and figure legends, assay tubes (total volume = 2 .5 ml) contained 0 .6 g of lung fragments, 10 .tg of OA, and the incubation was carried out for 20 min at 37°C . At the end of incubation, the medium was separated from the lung fragments by filtration using Whatman GF/C filter paper . After separation 0 .2 ml of filtrate was aliquoted, boiled and stored frozen at - 20°C for histamine measurements . The remaining filtrate (-2.2 ml) was used immediately for LTB 4 /LTD4 measurements . In some experiments, lung fragments
were prepared without perfusion or washing to remove blood. Extraction of 5-lipoxygenase products and reversephase HPLC analysis of the products The procedures were similar to the method reported previously 5,15 with slight modifications . In brief, 6 ml of acetonitrile was added to the filtrate to precipitate proteins . The supernatant was separated from the pellet by centrifugation at 14,000 x g for 10 min . The supernatant was decanted and acidified to pH 3 .5 with 21 ml of 1 .0 mM HCl before passage through a C 18 Sep-Pak cartridge (Water Associates, Framingham, MA) . The cartridge was washed with water (4 ml), 20% acetonitrile (4 ml) and then eluted with 70% acetonitrile (3 ml) . The eluted samples were dried under nitrogen . Each dry sample was reconstituted in 105 µl of water, followed by 55 µl of methanol . One hundred microliters of this mixture was then injected onto a Ranin Short One C18 column (3 µ, 10 x 0 .46 cm) and the products were separated isocratically using a mobile phase of 80% methanol-20% H2 O-0 .1% trifluoroacetic acid-0 .05% triethylamine at a flow rate of 1 ml/min . The peptidoleukotrienes and LTB 4 were detected by UV absorbance at 280 nm with a 18-µ1 flow cell spectrophotometer (Perkin-Elmer, Norwalk, CT LC95) . The peaks were quantitated by integration using a Perkin-Elmer LC1-100 integrator . LTB 4 metabolites,
1 74 Pulmonary Pharmacology
20-COOH-LTB4 (retention time : 3 .6 min) and 20OH-LTB 4 (4 .1 min), were separated from LTB 4 (14 .7 min) using a linear acetonitrile gradient system according to a previous method . 16 As shown previously' LTB 4 and LTD 4 were identified by coelution with authentic standards and ultraviolet spectral analysis, and the recovery of LTB4 and LTD 4 in the presence of lung tissues was 56% and 42%, respectively . The limit of detection for each LT by the HPLC measurement was approximately 0 .7 ng/g lung . In the absence of antigen, the amount of each LT produced from naive and actively- or passivelysensitized lung fragments was below this value . Measurements of histamine Histamine was determined using a double-isotopic enzymatic assay developed by Shaff and Beaven . 18 In brief, 5 ul of thawed sample or histamine standard solution (0, 5, 10, 25, 50, and 100 pg) was mixed with [3H]histamine (40 000 cpm/tube) and [14 C]SAM (50 000 cpm/tube) in 10 cm x 1 .5 cm borosilicate culture tubes . The reaction was initiated by incubating 50 ul of histamine-N-methyltransferase with the mixture at 37°C . After 90 min incubation, the reaction was stopped by the addition of 200 ul of unlabelled Nmethylhistamine (50 ug/200 ul) in 0 .8M perchloric acid . Histamine was then extracted into chloroform . The chloroform extract was washed once with 3 .3N NaOH, transferred to counting vials, evaporated to dryness and assayed for 3H and 14C activity . Antigendependent histamine release was corrected by subtracting the baseline histamine level (-0 .2 ug/g lung) in each experiment . Histamine-N-methyltransferase was prepared from rat kidneys ." Statistics All values are presented as the mean + SE. The data were analyzed by unpaired Student's t-test to determine if the differences between the amounts of mediators were significant at the 0 .05 level . For Figure 8, we performed a single linear regression to test significance of the correlation .
RESULTS 1. ELISA for measurement of serum anti-OA IgG1 Figure 2 illustrates the linearity of an IgG1 standard curve. Using this ELISA, it was possible to detect as little as 10 ng/ml anti-OA IgG I in guinea-pig serum . Serum per se had no affect on the standard curve . IgG1 levels in the samples were diluted and measured within the linear range (10-100 ng/ml) of the standard curve . IgG2 showed -30% cross-reactivity . 2. Characterization of leukotriene synthesis and histamine release in IgG1-passively sensitized guinea pig lung preparations Incubation of 0 .6 g/2 .5 ml of sensitized lung fragments with 10 ug/ml OA stimulated in a time-dependent fashion, the synthesis of LTB 4, LTD4 and the release of histamine (Fig . 3) . A significant amount of histamine was released as early as 2 min following antigen challenge . The generation of LTB 4 and LTD4 remained undetected until 2-4 min post antigen challenge . Peak responses for histamine release and LT synthesis occurred at 8 and 16 min respectively . HPLC analysis of the sample extracted from the filtrate of the antigen-challenged lungs at 20 min revealed no prominent peaks co-migrating with LTC 4 and LTE 4 or the known LTB 4 metabolites, 20-OHLTB 4 and 20-COOH-LTB4 . t 8, t 9 Figure 4 illustrates that OA stimulated LTB 4 and LTD4 synthesis and histamine release in the IgGI sensitized lung with an ED50 = -0 .3-1 jig/ml and a maximal effect at 10 .tg/ml . The latter amount and a 20 min incubation time were chosen for subsequent experiments . To determine the optimal amount of IgG1 for sensitization, guinea-pigs were injected i .p . with increasing amounts of the antibody . Two days after sensitization, the ability of lung fragments from these animals to synthesize LTB4, LTD 4 and release histamine was assessed ex vivo . As shown in Figure 5 the lowest
Chemicals LTB4, 20-COOH-LTB 4, 20-OH-LTB 4 and LTD4 were purchased from Biomol Inc . (Philadelphia, PA) . Ovalbumin (Grade V), histamine . 1-methylhistamine, biotin, streptavadin, alkaline phosphatase and ldpropanolol were all purchased from SIGMA Chemical Co. (St. Louis, MO). [2,5-3H-histamine dihydrochloride (48 Ci/mmol) and [ 14 C]-S-adenosyl-methionine [C 14-SAM] (58 .9 Ci/mmol) were obtained from New England Nuclear Co . (Boston, MA) . Rabbit anti-guinea pig (GP) IgG1 was purchased from Miles Scientific Co. (New Haven, CT) .
ANTI-OA IgG (ng)
Fig. 2-Anti-OA IgGI standard curve determined by the ELISA .
Mediator Release and Serum IgG1
c
3
60-
-6
50-
-5
1 75
40-
04
w
rn rn -3
-2
10-
LU Z 4 IN =
-1
0-
-0 0 2 4
16
32
64
TIME (min) 3-Time course of antigen-dependent leukotriene synthesis and histamine release in the IgG1 passively-sensitized lung . Guinea-pigs were passively immunized with a single intraperitoneal injection of anti-OA IgG1 and lung tissue was studied after 2 days . Lung fragments (0 .6 g/2 .5 ml) were incubated with 10 µg/ml OA for the time indicated . mean SE, n=4 . Fig.
s0-
8
so
6
T
o LT13 4 • LTD4 x Histamine 60-
-6
a a C 7 01
a
a
3
_ C
p'
40-
-4
-I
-6
C
J
40-
0
FN_
0 la H J
80
C
Cc
S 20-
m J
-2
20
-2
L0
0 0.125 0-
0
I
-5
.3 LOG OA (mg/mi)
4-Antigen concentration dependent leukotriene synthesis and histamine release in the IgG I passively-sensitized lung . Lung fragments (0.6 g/2.5 ml) were incubated with increasing concentrations of OA for 20 min mean ± SE, n = 4 . Fig.
amount of IgG1 administered (0 .125 mg/kg) resulted in the production of significant quantities of each mediator. The amount which caused 50% and maximal stimulation was estimated to be 0 .35 and 0 .5 mg/kg, respectively . As also shown in Figure 5, an amount greater than 0 .5 mg/kg failed to augment the levels of mediators further even though higher serum levels of anti-OA IgG1 were observed (see later) . To determine the duration of sensitization, guineapigs were injected i .p . with a single dose (0 .5 mg/kg) of the antibody and the OA-dependent response was determined at 1, 2, 3, 4, or 7 days post-immunization . As shown in Figure 6, a response was measurable as early as 1-day post-administration and lasted for 7
0 .25
0.5
0.75
ANTI-OA IgG1 (mg/kg)
Fig. 5-Antibody (anti-OA IgG1) dependent leukotriene synthesis and histamine release as a function of sensitization dose. Sensitization involved a single i.p . injection of increasing doses of the antibody and lung tissue was studied after 2 days . Lung fragments (0.6 g/2 .5 ml)were incubated with 10 µg/ml OA for 20 min mean ± SE, n=4 .
days . The level of LTB 4 /LTD4 or histamine at day 4 was significantly less than at day 2 or day 7 (p < 0 .05) . In an additional experiment, antigen-dependent mediator release/synthesis (histamine : 4 .8±0 .4ug/g lung ; LTB 4 : 20 ± 3 ng/g lung ; LTD4 ; 42 ± 10 ng/g lung; n = 4) was still measurable at 14 days after a single injection of 0 .5 mg/kg of IgG1 . Antigen-dependent LT synthesis in perfused and nonperfused preparations Circulating leukocytes from rats, rabbits, and humans synthesize significant amounts of LTB 4 and LTC 4 upon stimulation by various agonists . 2,4 In an attempt to determine if the antigen-dependent synthesis
1 76 Pulmonary Pharmacology 6o-
6
50-Sr
14-
5
N = 6 Y=11 .6x+0.16 0 = 0 .951
12-
4 /
10-
J O
30-
3
W J y
J O
0E Q a
2
20-
/
Oy Z a Q O
I-
J 10-
8-
6-
/
N W a Vi 4-
0
I 1
0 2
3
4
5
6
7
/ 2-
TIME POST SENSITIZATION (day)
X
Fig . 6-The time dependence of passive sensitization for leukotriene synthesis and release . Guinea-pigs were sensitized with 0 .5 mg/kg IgG1 for the time indicated . Lung fragments (0 .6 g/2 .5 ml) were incubated with 10 µg/ml OA for 20 min mean + SE, n=7-21 .
a
02
0
0 .4
0 .6
0 .8
1 .0
1 .2
ANTI-OA IgG1 DOSE (mg/kg, i .p .)
of LTs in the guinea pig lung preparation was dependent on circulating cells, the quantity of LTs found in perfused and non-perfused lung was determined . As shown in Figure 7, lungs that were not perfused with medium synthesized LTB and LTD in an antigen concentration-dependent manner, and the amounts extracted from the medium were close to those found in the perfused-lung preparations . In addition, HPLC analysis of the extracted sample obtained from the non-perfused lung preparation following antigen challenge revealed no distinct peaks co-migrating with 4
/
4
60
50-
Fig. 8-Dose-response relationship between circulating anti-OA IgG1 levels and the amount of IgG1 passively administered . Guinea-pigs were passively immunized with increasing amounts of anti-OA IgG1 and lung tissue studied 2 days later, mean ± SE, n = 8-16 .
known LTB COOH-LTB
4
metabolites (20-OH-LTB
4
or 20-
4)
3. Serum levels of anti-OA IgG1 in IgG1 passivelysensitized guinea pigs Using the above described ELISA, serum anti-OA IgG l levels in passively-sensitized guinea pigs were determined 2 days post-IgG1 administration (Fig . 8) A circulating level of - 1 ug/ml was found in animals sensitized with 0 .075 mg/kg antibody . With increasing amounts of antibody (up to 1 mg/kg), serum levels increased linearly . Figure 9 shows the time course of the circulating antibody levels of anti-OA IgG1 after i .p . administration of 0.5 mg/kg . A plasma level of - 5 ug/ml was detected at 6 h post-administration and this concentration remained constant for up to 7 days . In naive guinea-pigs, no circulating level of anti-OA IgGI was demonstrable .
10-
0-
4 . Leukotriene synthesis and histamine release in actively-sensitized guinea pig lung I .5
I -4
I .3
I -2
I -1 .5
LOG OA (mg/ml)
Figure 10 demonstrates that lungs taken from actively-sensitized guinea-pigs synthesize LTB4/LTD and release histamine as a function of time, and Figure I1 illustrates the antigen concentration response relationship. The time course and antigen concentration-response curve for histamine release and LTB4/D synthesis were similar to those obtained from the passively-sensitized lungs . More importantly, the amount of LTB , LTD synthesized and
4
Fig. 7-Comparison of antigen-dependent leukotriene synthesis in perfused and non-perfused lung preparations . For perfused preparations, lungs were perfused via the pulmonary artery with 20 ml cold Krebs-Ringer medium before the lung parenchyma was chopped into 1 mm' tissue fragments . The fragments were then washed 5 x with cold medium . For non-perfused preparations, the lung was neither perfused nor washed with the medium . The lung fragments (0 .6 g/2 .5 ml) were incubated with 10 µg/ml OA for 20 min mean ± SE, n = 4 .
4
4
4
Mediator Release and Serum IgG1 8-
1 77
-8
-6
-4
-2
.5
I -4
I -3
I -2
0
I -1 .5
LOG OA (mg/ml)
0
I
I
I
I
I
I
1
2
3
4
5
6
I 7
TIME POST SENSITIZATION (day)
Fig. 11-Antigen concentration-dependent leukotriene synthesis and histamine release in the actively-sensitized lung . Lung fragments (0 .6 g/2.5 ml) were incubated with increasing concentrations of OA for 20 min, mean ± SE, n = 4 . (38)
Fig. 9-Circulating anti-OA IgGI levels as a function of time post-sensitization. Guinea-pigs were passively immunized with 0 .5 mg/kg anti-OA IgGI and then serially bled at the times indicated, mean ± SE, n = 3-7 . 60
6
so
-6
3 d E rn
1
g 40
p
Y
-3
30
j0
m
-2
20
10-
-1
10
0-
PASSIVE SENSITIZATION 0
0 024
8
16
32
0 .5 mg/kg IgG1
64
ACTIVE SENSITIZATION 1 mg/kg OA followed by 0.25 x 3 mg/kg OA
TIME (min)
Fig. 10-Time course of antigen-dependent leukotriene synthesis and histamine release in the actively-sensitized lung . The active sensitization protocol involved multiple antigen injections as described previously . 1- ,14 In brief, guinea-pigs were injected s .c. with 1 mg/kg OA . A total of 14 days later, the animals were given weekly injections of 0 .25 mg/kg OA for 3 weeks and then sacrificed 3-8 days after the last injection . Lung fragments (0 .6 g/2 .5 ml) were incubated with 10 µg/ml OA for the times indicated, mean±SE, n=4.
histamine released was also similar between both passively- and actively-sensitized lungs . 5. Comparison of serum anti-OA IgG1 levels in passively- and actively-sensitized guinea pigs The serum antibody levels of 16 passively- and 38 actively-sensitized guinea-pigs were measured according to the procedure shown in Table 1 . Figure 12 illustrates that the levels of anti-OA IgG1 in activelysensitized guinea-pigs were 130-fold higher than those detected in passively-sensitized animals . In addition, the antibody levels relative to the mean value in the actively sensitized group (co-efficient of varia-
Fig. 12-Comparison of serum anti-OA IgG 1 levels in passivelyand actively-sensitized guinea-pigs . The immunization protocol involved either a single antibody injection (passive senzitization) or multiple antigen injections (active sensitization). Sensitization time was 2 days for passive sensitization and 31-36 days for active sensitization .
tion = 145 .2%) were more variable than those in the passively-sensitized group (31%) . Since the rabbit anti-GP IgG1 used in the ELISA shows -30% cross-reactivity with an OA-specific IgG2 subclass of antibody and anti-OA IgG2 is likely to be present in the active immunization procedure, the levels in actively-sensitized guinea-pigs could have been overestimated by 30% . Taking this into consideration, serum levels of the IgG1 antibody in actively-sensitized animals were still significantly (> 90-fold) higher (p < 0 .05) .
DISCUSSION In this study an antigen-specific ELISA was used to demonstrate that :
178
Pulmonary Pharmacology
a) the circulating levels of IgG1 increase linearly with the quantity of antibody passively administered to naive guinea pigs (2 days post i .p . dose) (Fig . 8) and b) the level is maintained for at least 7 days (Fig . 9) . Lung sensitization was observable at 0 .125 mg/kg IgG1 corresponding to a plasma level of -2 ug/ml (Figs . 5 & 8) . Moreover, the circulating antibody level at the dose (0 .5 mg/kg) that caused maximal sensitization was .- 6 ug/ml . These results indicate that 2-6 gg/ml of IgGI in plasma is sufficient to sensitize lung tissue . It appears that the antigen-dependent response is maximal at a circulating antibody level of 6 ug/ml, because increased IgG1 levels obtained using a higher amount of antibody to sensitize animals (0 .75 mg/kg) is not accompanied by enhanced mediator release/ synthesis (Fig . 5) . Both requirements for low plasma levels of IgGI for sensitization and the maintenance of antibody levels in circulation (long serum half-life) may partly explain the present and prior observation' 1-13,20,21 that once guinea-pigs are immunized with this subclass of antibody the sensitization lasts for 7 days and beyond . In the experiment where the degree of sensitization time was examined (Fig. 6), there was decreased mediator production at day 4 post-antibody administration . The reason for this decrease is currently unclear . As demonstrated by these studies, passively-sensitized lung tissue produced a significant amount of LTB 4/LTD 4 and histamine upon antigen challenge ex vivo . In fact, in respect to both kinetic and antigen dose-response parameters, the amount of histamine released and LTB4/D4 synthesized was comparable to that generated from lung fragments of actively-sensitized guinea pigs (Fig . 3 vs 10 & Fig . 4 vs 11) . Because IgG1 levels in the latter animals far exceed those required for maximal sensitization, an induction of this antibody subclass alone by the active immunization could account for the antigen-dependent response . The present studies have characterized mediator production from IgG1-sensitized lung fragments with regards to both the dose-response relationship for both antigen and antibody and the kinetics of their interaction . The profile and the quantity of antigenstimulated LT synthesis and histamine release by sensitized lungs agree with previously published results . 2, t 3,22,23 LTD 4 is a relatively stable terminal product of the 5-lipoxygenase pathway in the chopped lung preparation (Figs . 3 & 10) . At 20 min incubation time, no evidence for presence of LTC 4 and LTE4 (< 0 .7 ng/g lung) was found . This is consistent with the result obtained from actively-sensitized guinea-pig lungs .' LTB 4 has also been shown in this study to be another stable product because (1) 20-OH-LTB 4 or 20-COOH-LTB 4 could not be detected in the sample extracted from the supernatant of antigen-challenged lung and (2) the level of LTB4 (following antigen
challenge) was constant for up to 64 min . Guinea-pig lung fragments could have contained an insufficient number of polymorphonuclear leukocytes thus reducing its capacity to degrade LTB 4 as shown in a recent study using rabbit lungs .24 This possibility seems to be unlikely because each LTB 4 metabolite could again not be detected in non-perfused lung preparations . At 20 min post-antigen challenge, the molar ratio of LTD4/LTB4 generated from IgGI passivelysensitized lungs was about 3 (corrected according to recovery) . A similar ratio for these mediators was also released from actively-sensitized lungs (Figs . 10 & 11) . These results indicate that resident lung cells in guinea-pigs may have a greater capacity to synthesize peptidoleukotrienes than LTB 4 following immunological challenge . This possibility is supported by the presence of a highly specific LTC 4 synthetase in guinea-pig lung membrane . 25 A greater increase in both enzymatic transformation of LTA 4 to LTC4 and production of LTC 4 relative to LTB4 formation has also been seen in the lungs of rats 26,2 ' and rabbits .28 .29 While human lung tissue produced mostly LTD 4 in response to anti-IgE challenge ex vivo, 30-32 antigen challenge caused equal molar production of LTB4 and LTD 4 from human lungs that were passively-sensitized with hyperimmune serum in vitro . 33 No difference was observable between the response of the perfused and non-perfused preparations, suggesting that FcR-bearing resident cells such as macrophages 34,35 and possibly mast cells 36,3 ' rather than circulating cells are the cell types responsible for LT synthesis in the passively-sensitized guinea-pigs . However, this suggestion must be tentative because many circulating cells are fixed to the lung via specific adhesion molecules and thus may not wash off easily . In summary, the passive administration of purified antigen specific (OA) IgG1 to normal guinea-pigs allowed a demonstration of a direct relationship between the circulating level of antibody (anti-OA IgG1) and the capacity of antigen (OA) to induce histamine and leukotriene (LTB 4/LTD 4) synthesis from lung tissue ex vivo . The level of circulating antibody required to maximally sensitize lung tissue was approximately 6 gg/ml, corresponding to a plasma concentration of - 40 nM suggesting a high affinity binding of IgG1 to its receptors . That such low levels of antibody appear to be sufficient for maximum tissue sensitization raises interesting questions regarding the potential blocking effects of other classes of antigen-specific antibody 38 or non-specific IgG1 competing at surface receptors . These questions and others related to the distribution of FcR for IgG 1 in guinea-pig lung are currently under investigation . References 1 . Anderson P, Antigen-induced bronchial anaphylaxis in actively sensitized guinea pigs Anti-anaphylactic effects of sodium cromoglycate and aminophylline . Brit J Pharmacol 1980a; 69: 467-472 .
Mediator Release and Serum IgG1 2 . Piper P J. Formation and actions of leukotrienes . Physiol . Rev. 1984 ; 64 : 744-761 . 3 . Muccitelli R M, Tucker S S, Hay D W P, Torphy T J, Wasserman M A . Is the guinea pig trachea a good in vitro model of human large and central airways? Comparison on leukotriene-, methacholine- and antigen-induced contractions . J Pharmacol Exp Therap 1987 ; 243 : 467-473 . 4 . Bach M K. Lipid mediators of hypersensitivity . Prog Allergy 1988 ; 44 : 10-98 . 5 . Cheng J B, Eskra J D, Pillar J . Comparison of antigen and Ca ++ -ionophore-induced peptidoleukotriene release from guinea pig lung preparations using highperformance liquid chromatography . J Pharmacol Exp Therap 1987a; 241 : 786-792. 6. Undem B J, Bucker C K, Harley P, Graziano F M . Smooth muscle contraction and release of histamine and slow-reacting substance of anaphylaxis in pulmonary tissues isolated from guinea pigs passively sensitized with IgG1 or IgE antibodies . Am Rev Resp Dis 1985 ; 131 : 260-266 . 7 . Regal J F. IgG vs IgE mediators of antigen-induced guinea pig tracheal contraction . Immunopharmacol. 1984; 8 : 111-119 . 8 . Regal J F. IgG vs. IgE mediators of antigen-induced guinea pig lung parenchymal contraction . Immunopharmacol . 1985 ; 10 : 137-146 . 9 . Conklyn M J, Showell H J. Purification of antigenspecific reaginic IgG1 from ovalbumin-sensitized guinea pigs. New England Pharmacologists 6th Annual Meeting (Abstract) . 987; p . 25 . 10 . Regal J F . Immunoglobulin G- and immunoglobulin E-mediated airway smooth muscle contraction in the guinea pig. J Pharmacol Exp Therap 983 ; 228 : 6-20 . 11 . Benacerraf F, Ovary Z, Bloch K J, Franklin E C. Properties of guinea pig 7S antibodies . I . Electrophoretic separation of two types of guinea-pig 7S antibodies . J Exp Med 963 ; 7 : 937-950. 12 . Ovary Z, Benacerraf B, Bloch K J . Properties of guinea pig antibodies . II Identification of antibodies involved in passive cutaneous and systemic anaphylaxis . J Exp Med 963 ; 7 : 95-964 . 13 . Graziano F M, Gundersen L, Larson L A, Harley P, Buckner C K . Receptor-specific mediation by immunoglobulin E of antigen-induced contraction of tracheal and lung parenchymal strips isolated from the guinea pig. J Clin Invest 1984 ; 73: 1215-1222 . 14 . Souhrada M, Souhrada J F . Potentiation of Na + -electrogenic pump of airway smooth muscle by sensitization. Respir Physiol 1982 ; 47 : 69-81 . 15 . Eskra J D, Pereira M J, Ernest M J . Solid-phase extraction and high-performance liquid chromatography analysis of lipoxygenase pathway products . Anal Biochem 1986; 154: 332-337 . 16 . Sweeney F J, Eskra J D, Carty T J. Development of system for evaluating 5-lipoxygenase inhibitors using human whole blood . Prostaglandins Leukotrienes Med 1987; 28 : 73-93 . 17 . Shaff R E, Beaven M A . Increased sensitivity of the enzymatic isotopic assay of histamine: Measurement of histamine in plasma and serum . Analyt Biochem . 1979; 94: 425-430 . 18 . Sun F F, McGuire J C . Metabolism of arachidonic acid by human neutrophils . Characterization of the enzymatic reactions that lead to the synthesis of leukotriene B4 . Biochem Biophys Acta 1984; 794 : 56-64 . 19. Shak S, Goldstein I M . w-Oxidation is the major pathway for catabolism of leukotriene B 4 in human polymorphonuclear leukocytes . J Biol Chem 1984; 259 :10181-10187 . 20 . Andersson P . Antigen-induced bronchial anaphylaxis in actively sensitized guinea pigs . Pattern of response in relation to immunization regimen . Allergy 1980b ; 35 : 65-71 . 21 . Fleisch J H, Spaethe S M . Improved methodology for evaluating release of mediators of anaphylaxis from guinea pig lung. Monogr Allergy 1979 ; 14: 185-188 . 22. Clancy R M, Hugh T E . The extraction of leukotri-
enes (LTC4, LTD, and LTE 4) from tissue fluids : The metabolism of these mediators from IgE-dependent hypersensitivity reaction in lung. Anal Biochem 1983; 133 : 30-39. 23 . Turner N C, Dollery C T . Release of arachidonic acid metabolite and histamine from sensitized guinea pig lung following antigen challenge . Brit J Pharmacol 1988 ; 93 : 751-758 . 24 . Grimminger F, Menger M, Becker G, Seeger W . Potentiation of leukotriene production following sequestration of neutrophils in isolated lungs : Indirect evidence for intercellular leukotriene A 4 transfer . Blood 1988a ; 72 : 1687-1692 . 25 . Yoshimoto T, Soberman R J, Spur B, Austen K F . Properties of highly purified leukotriene C 4 synthase of guinea pig lung. J Clin Invest 1988 ; 81 : 866-871 . 26 . Medina J F, Haeggstrom J, Kumlin M, Radmark O . Leukotriene A4 : Metabolism in different rat tissues . Biochem Biophys Acta 1988; 961 : 203-212 . 27. McDonnell T J, Chang S-W, Westcott J Y, Voelkel N F. Role of oxidants, eicosanoids, and neutrophils in amphotericin B lung injury in rats. J Appl Physiol 1988 ; 65 : 2195-2206. 28 . Schultz R, Seeger W . Release of leukotrienes into the perfusate of calcium-ionophore stimulated rabbit lungs : Influence of 5-lipoxygenase inhibitors. Biochem Pharmacol 1986; 35 : 183-193 . 29 . Grimminger F, Becker G, Seeger W . High yield enzymatic conversion of intravascular leukotriene A 4 in the blood-free perfused lungs . J Immunol 1988b (In Press) . 30 . Lewis R A, Austen K R, Drazen J M, Clark D A, Marfat A, Corey E J . Slow reacting substance of analphylaxis : Identification of leukotriene C-1 and D from human and rat sources . Proc Natl Acad Sci 1980; USA 77 : 3710-3714 . 31 . Kumlin M, Dahlen S E, Granstrom E . Differences in the formation and metabolism of leukotrienes B4 and C 4 in the human lung . Adv Prostaglandin Thromboxane Leukotriene Res 1987 ; 17: 1018-1022 . 32 . Vigano T, Toia A, Galli G, Berti F, Crivellari M T, Folco G C . Adenosine and eicosanoid release from immunologically challenged human lung fragments . Adv Protaglandin Thromboxane Leukotriene Res 1987; 17 : 992-996 . 33 . Salari H, Borgeat P, Fournier M, Hebert J, Pelletier G. Studies on the release of leukotrienes and histamine by human lung parenchymal and bronchial fragments upon immunologic and nonimmunologic stimulation : Effects of nordihydroguaiaretic acid, aspirin, and sodium cromoglycate . J Exp Med 1985 ; 162 : 1904-1915. 34 . Rankin J A, Hitchcock M, Merrill W, Bach M K, Brashler J R, Askenase P W . IgE-dependent release of leukotriene C4 from alveolar macrophages . Nature (Lond.) 1982 ; 297: 329-331 . 35 . Nagel G P, Young K R, Jr, Reynolds H Y . Receptors for human IgG subclasses on human alveolar macrophages . Am Rev Respir Dis 1984 ; 129 : 413-418 . 36. Peters S P, MacGlashan D W, Jr, Schulman E S, Schleimer R P, Hayes E D, Rokach J, Adkinson N F, Jr, Lichtenstein L M . Arachidonic acid metabolism in purified human lung mast cells . J Immunol 1984; 132 : 1972-1979 . 37 . Lewis R A, Austen K F . Mediation of local homostasis and inflammation by leukotrienes and other mast cell-dependent compounds . Nature 1981 ; (Lond .) 293 : 103-108. 38 . Yamauchi N, Ito K, Suko M, Ishii A, Miyamoto T . IgG2 antibodies block IgE antibody-induced asthma in guinea pigs . Int Archs Allergy AppI Immun 1986 ; 80: 76-80 . Date received: 6 August 1989 Date revised: 20 October 1989 Date accepted: 10 March 1990
1 79
0952-0600/90/0003-0171/510 .00
PULMONARY PHARMACOLOGY
Antigen-Dependent Leukotriene Synthesis and Histamine Release from IgG1 Passively-Sensitized Guinea Pig Lungs Ex Vivo : Relationship Between Serum Levels of Antigen-Specific IgG1 and Mediator Synthesis/Release J . B . Cheng, J . S . Pillar, M . J . Conklyn, R . Breslow, J. T . Shirley, N . P. Gerard*, H . J . Showell Department of Immunology and Infectious Diseases, Central Research Division, Pfizer Inc ., Groton, Connecticut (USA) ; *Harvard Medical School, Beth Israel Hospital, Boston, Massachusetts (USA) SUMMARY . Naive guinea-pigs were passively sensitized with varying amounts of affinity column purified, homologous, anti-ovalbumin IgG1 (anti-OA IgG1) and then examined for a) the capacity of lung tissue to release mediators (histamine and LTB4 /LTD4 ) in response to antigen-challenge ex vivo and b) the attendant circulating levels of anti-OA IgG1 . Intraperitoneal administration of anti-OA IgG1 (0 .125-0.75 mg/kg) to guinea-pigs facilitated the synthesis of LTB 4 (8-25 ng/g lung) and LTD 4 (18-80 ng/g) and the release of histamine (1-7 ug/g) from lung tissue after exposure to 10 pg/ml of ovalbumin for 20 min ex vivo. Peak levels of mediators were found using 0 .5 mg/kg anti-OA IgG1 with an ED so =0 .35 mg/kg. LTD4 /LTB4 synthesis and histamine release were both antigen concentration- and time-dependent, and LT synthesis was observable in non-perfused lungs and in lungs perfused free of blood . Maximum sensitization occurred at 1-2 days post i .p. administration of anti-OA IgG1 and was maintained up to 7 days . Measurement of anti-OA IgG1 using an enzyme-linked immunosorbent assay demonstrated that circulating antibody levels were 2-6 µg/ml at the doses which caused sensitization . The level of anti-OA IgG1 found in passively sensitized animals was at least 100-fold less than that found in actively-sensitized guinea-pigs despite the similar magnitude in LTD4 /LTB4 synthesized and the amount of histamine released . Using purified antibody, the results demonstrate that in guinea-pigs, IgG1 can play a prominent role in regulating lung LT synthesis and histamine release, and that microgram per ml circulating levels of this antibody are sufficient to sensitize naive lungs .
ine from lungs of IgG1-sensitized guinea-pigs than those from IgE-sensitized lungs . Also, in vitro studies''' demonstrated that SRS-A was involved in the contraction of the IgG1-sensitized guinea-pig tracheal and parenchymal preparations but not in the IgE-sensitized tissue . These results continue to support the idea that IgG1 antibody plays an important role in immunologically regulating mediator production in guinea-pigs . Since mediator release/production from cells is linked to receptor occupancy, at equilibrium the level of occupancy should be dependent on the circulating level of antibody (up to the point of saturation) . Determination of the level of antibody should help us to understand the quantity of circulating IgG1 necessary for sensitization . However, there is little information in this regard . Moreover, a relationship between the circulating level of antigen-specific IgG1 and amount of mediators released from lung tissue remains to be demonstrated . An OA-specific homologous IgGl antibody has recently been isolated from hyperimmune sera of actively-sensitized guinea pigs . The antibody has been purified to homogeneity by chromatography .' Utiliz-
INTRODUCTION Lungs taken from actively- and passively-sensitized guinea-pigs synthesize leukotrienes (LTs) and release histamine upon antigen challenge ex vivo . This source of sensitized tissue has been used as a model system to study biochemical pathways of mediator generation and in addition how these pathways can be manipulated pharmacologically ." Recently, it became appreciated that the quantity of LTs generated could be affected by immunization protocols . Lung fragments of guinea-pigs injected with ovalbumin (OA) four times have been shown to produce more LTD 4 after antigen challenge than lung fragments of single OA injected animals .' In addition, both qualitatively and quantitatively mediator synthesis/release appear to vary depending upon whether IgG1 or IgE has been used to sensitize tissue . For example, Undem et al.' demonstrated substantially more release of slow-reacting substance of anaphylaxis (SRS-A) and histamCorrespondence to : Dr John B . Cheng, Department of Immunology and Infectious Diseases, Central Research Division, Pfizer Inc., Groton, CT 06340. 171
1 72 Pulmonary Pharmacology
ing purified IgG1 should permit a detailed examination of antigen-dependent mediator release/synthesis from passively-sensitized lungs . Moreover, an antigen-specific (anti-OA) IgG1 developed in the laboratory, should allow a quantitation of circulating levels of IgG1 . The information relating serum levels of IgG1 to antigen-induced response should also allow a determination of the in vivo potency of this antibody to sensitize lung tissues . A study was undertaken, therefore, to 1 . characterize OA-dependent LT synthesis and histamine release from IgG1 passively-sensitized lungs, 2 . determine the relationship between mediator production and circulating IgG1 levels in sensitized animals, and 3 . compare these findings with those obtained from lungs of actively-sensitized guinea-pigs .
3 mM) . SDS-PAGE electrophoresis analysis of the eluted sample revealed > 95% purity of the antibody (MW = 146 000) with no contamination with IgG2 antibody .' A passive cutaneous anaphylaxis (PCA) test in guinea-pigs .' 1-13 confirmed that the purified IgG1 antibody was heat-stable at 56°C for 4 h and persisted in the skin site for no more than 5 days . Development of a new enzyme-linked immunosorbent assay (ELISA) for the measurement of serum antiOA IgG1 In order to determine the circulating levels of anti-OA IgG1, a sensitive and antigen-specific ELISA was developed .' t-13 Table 1 shows the protocol for the ELISA . Guinea pig immunization and the chopped lung assay
METHODS Isolation and purification of an OA-specific IgG1 antibody from the serum of actively-sensitized guinea pigs
Male Hartley guinea-pigs (Charles River Laboratories, Wilmington, MA) were immunized with OA precipitated onto alum as described previously ."° The pooled hyperimmune sera was the source for the isolation of anti-OA IgG1 used in this study . As described previously' (procedures also shown in Fig. 1), IgG1 was separated from IgG2 by passing the sera through a Protein A-Sepharose 413 column and then isolated using an OA-linked Sepharose 4B column . The isolated sample was re-applied to the Protein A column, and the OA-specific IgG1 was eluted from the column using a pH gradient of 6 .0-5 .0 (glycine,
Passive immunization Male Hartley guinea pigs (age : 3-4 weeks) were sensitized by a single intraperitoneal injection of the purified anti-OA antibody (0 .175-0 .75 mg/kg) . Guinea-pigs were sacrificed at 48 h post injection unless otherwise specified . Active immunization This was performed according to the multiple antigen injection protocol described previously except that 5-8 day-old guinea pigs were used . 5,14 When a comparison of antigen-dependent mediator generation from passively- and actively-sensitized lungs was made, the final age of both groups was similar (3-5 weeks) . Chopped lung assay Guinea-pigs were anesthetized with pentobarbital (40-60 mg/kg) and exsanguinated by cutting the ab-
Pooled Antiserum
a
G
Protein A - Sepharose CL-4B
IgG2
a Ovalbumin - Sepharose 4B
G
IgG1 + IgG2
P a Effluent
a Protein A - Sepharose 4B CH pH
I
Effluent
i IgG1 P = Phosphate Buffered Saline, pH 7 .4 G = OA M Glycine, pH 3 .0 CP = Citrate Phosphate Buffer, pH 7 .4 pH = Citrate Phosphate Buffer, pH 6 .0 - 5 .0 Fig. 1-Isolation protocol for guinea-pig anti-OA IgG1 .
Mediator Release and Serum IgGI
1 73
Table 1 . Anti-ovalbumin IgGI ELISA Add 100 µl ovalbumin (0 .1 mg/ml carbonate buffer, pH 9 .6) per well (Dynatech Lab, Virginia) 2 . Blocking : Wash 2 x 200 Ed carbonate buffer Add 200 µl 1% BSA in carbonate buffer, 1 hr incubation, room temperature (RT) 3 . GP IgG1 binding Wash I x 200 µl carbonate buffer Wash 2 x 200 p1 PBS-T, pH 9 .8 Add 100 pl Sample/Standard in PBS-T, pH 9 .8, 1 hr incubation, RT 4. Rabbit Anti-GP IgG1 Wash 2 x 200 N1 PBS-T, pH 9.8 (biotinylated) Add 100 µl rabbit anti-GP IgG1 Binding : (biotinylated), 1 hr incubation, RT 5 . Complex binding: Wash 1 x 200 gl PBS-T, pH 9 .8 Wash 2 x 200 pl PBS-T, pH 7 .4 Add 100 gl complex solution* 30 min incubation, RT Wash 2 x 200 pl PBS-T, pH 7 .4 6. Development : Wash 1 x 200 p1 DEA buffer Add 100 pl nitrophenyl phosphate solution (1 mg/ml DEA), 10 min incubation, RT Add 25 µl 5N KOH Read OD410 *Complex formation : Combine 40 pl Streptavadin (1 mg/ml) and 40 µl biotinylated-alkaline phosphatase (0 .5 mg/ml) in 10 ml PBS-T, pH 7 .4, 30 min incubation, RT prior to use . 1 . Antigen binding :
Carbonate 1 .59 g Na2 C0 3 2 .93 g NaHCO 3 1000 ml H 20 adjust pH to 9 .6
PBS-T 8 .0 g NaCl 0 .2 g KH 2 P0 4 1 .15 g Na2 HPO 4 0 .2 g KCI 0 .5 ml Tween 20 1000 ml H20 adjust to pH 7 .4 or 9 .8
DEA 4.85 ml diethanolamine 50 mg MgC1 2 .6H 2 0 500 ml H2 0 adjust pH to 9 .8
dominal aorta and inferior vena cava . The apex of the heart was removed to ensure free flow of the perfusate before the lung was perfused via the pulmonary artery with 10-20 ml of cold Krebs-Ringer medium (see below) . The perfused lung was removed quickly . The bronchi and large bronchioles were trimmed away and the lung parenchyma was placed on a Mcllwain Chopper and cut into approximately 1 mm 3 tissue fragments . These were then washed five times with cold medium and weighed into reaction tubes containing 2 .5 ml of the medium and held on ice until assayed . The medium used through the procedure contained (millimolar) : NaCl, 117 .5 ; KCl, 5 .37 ; MgS04 .7H 2 0, 2 .52; CaC1 2 , 2.52 ; NaHCo 3 , 15 .51 ; NaH 2 PO4 , 1 .17 ; glucose, 5 .50 ; and sucrose, 13 .65 and was gassed with 95% O Z -5% CO 2 to bring the pH to 7.4 (25°C) . All lung fragments were prewarmed at 37 °C for 20 min in the presence of 1 gM propranolol' before OA challenge . Except where noted in the text and figure legends, assay tubes (total volume = 2 .5 ml) contained 0 .6 g of lung fragments, 10 .tg of OA, and the incubation was carried out for 20 min at 37°C . At the end of incubation, the medium was separated from the lung fragments by filtration using Whatman GF/C filter paper . After separation 0 .2 ml of filtrate was aliquoted, boiled and stored frozen at - 20°C for histamine measurements . The remaining filtrate (-2.2 ml) was used immediately for LTB 4 /LTD4 measurements . In some experiments, lung fragments
were prepared without perfusion or washing to remove blood. Extraction of 5-lipoxygenase products and reversephase HPLC analysis of the products The procedures were similar to the method reported previously 5,15 with slight modifications . In brief, 6 ml of acetonitrile was added to the filtrate to precipitate proteins . The supernatant was separated from the pellet by centrifugation at 14,000 x g for 10 min . The supernatant was decanted and acidified to pH 3 .5 with 21 ml of 1 .0 mM HCl before passage through a C 18 Sep-Pak cartridge (Water Associates, Framingham, MA) . The cartridge was washed with water (4 ml), 20% acetonitrile (4 ml) and then eluted with 70% acetonitrile (3 ml) . The eluted samples were dried under nitrogen . Each dry sample was reconstituted in 105 µl of water, followed by 55 µl of methanol . One hundred microliters of this mixture was then injected onto a Ranin Short One C18 column (3 µ, 10 x 0 .46 cm) and the products were separated isocratically using a mobile phase of 80% methanol-20% H2 O-0 .1% trifluoroacetic acid-0 .05% triethylamine at a flow rate of 1 ml/min . The peptidoleukotrienes and LTB 4 were detected by UV absorbance at 280 nm with a 18-µ1 flow cell spectrophotometer (Perkin-Elmer, Norwalk, CT LC95) . The peaks were quantitated by integration using a Perkin-Elmer LC1-100 integrator . LTB 4 metabolites,
1 74 Pulmonary Pharmacology
20-COOH-LTB4 (retention time : 3 .6 min) and 20OH-LTB 4 (4 .1 min), were separated from LTB 4 (14 .7 min) using a linear acetonitrile gradient system according to a previous method . 16 As shown previously' LTB 4 and LTD 4 were identified by coelution with authentic standards and ultraviolet spectral analysis, and the recovery of LTB4 and LTD 4 in the presence of lung tissues was 56% and 42%, respectively . The limit of detection for each LT by the HPLC measurement was approximately 0 .7 ng/g lung . In the absence of antigen, the amount of each LT produced from naive and actively- or passivelysensitized lung fragments was below this value . Measurements of histamine Histamine was determined using a double-isotopic enzymatic assay developed by Shaff and Beaven . 18 In brief, 5 ul of thawed sample or histamine standard solution (0, 5, 10, 25, 50, and 100 pg) was mixed with [3H]histamine (40 000 cpm/tube) and [14 C]SAM (50 000 cpm/tube) in 10 cm x 1 .5 cm borosilicate culture tubes . The reaction was initiated by incubating 50 ul of histamine-N-methyltransferase with the mixture at 37°C . After 90 min incubation, the reaction was stopped by the addition of 200 ul of unlabelled Nmethylhistamine (50 ug/200 ul) in 0 .8M perchloric acid . Histamine was then extracted into chloroform . The chloroform extract was washed once with 3 .3N NaOH, transferred to counting vials, evaporated to dryness and assayed for 3H and 14C activity . Antigendependent histamine release was corrected by subtracting the baseline histamine level (-0 .2 ug/g lung) in each experiment . Histamine-N-methyltransferase was prepared from rat kidneys ." Statistics All values are presented as the mean + SE. The data were analyzed by unpaired Student's t-test to determine if the differences between the amounts of mediators were significant at the 0 .05 level . For Figure 8, we performed a single linear regression to test significance of the correlation .
RESULTS 1. ELISA for measurement of serum anti-OA IgG1 Figure 2 illustrates the linearity of an IgG1 standard curve. Using this ELISA, it was possible to detect as little as 10 ng/ml anti-OA IgG I in guinea-pig serum . Serum per se had no affect on the standard curve . IgG1 levels in the samples were diluted and measured within the linear range (10-100 ng/ml) of the standard curve . IgG2 showed -30% cross-reactivity . 2. Characterization of leukotriene synthesis and histamine release in IgG1-passively sensitized guinea pig lung preparations Incubation of 0 .6 g/2 .5 ml of sensitized lung fragments with 10 ug/ml OA stimulated in a time-dependent fashion, the synthesis of LTB 4, LTD4 and the release of histamine (Fig . 3) . A significant amount of histamine was released as early as 2 min following antigen challenge . The generation of LTB 4 and LTD4 remained undetected until 2-4 min post antigen challenge . Peak responses for histamine release and LT synthesis occurred at 8 and 16 min respectively . HPLC analysis of the sample extracted from the filtrate of the antigen-challenged lungs at 20 min revealed no prominent peaks co-migrating with LTC 4 and LTE 4 or the known LTB 4 metabolites, 20-OHLTB 4 and 20-COOH-LTB4 . t 8, t 9 Figure 4 illustrates that OA stimulated LTB 4 and LTD4 synthesis and histamine release in the IgGI sensitized lung with an ED50 = -0 .3-1 jig/ml and a maximal effect at 10 .tg/ml . The latter amount and a 20 min incubation time were chosen for subsequent experiments . To determine the optimal amount of IgG1 for sensitization, guinea-pigs were injected i .p . with increasing amounts of the antibody . Two days after sensitization, the ability of lung fragments from these animals to synthesize LTB4, LTD 4 and release histamine was assessed ex vivo . As shown in Figure 5 the lowest
Chemicals LTB4, 20-COOH-LTB 4, 20-OH-LTB 4 and LTD4 were purchased from Biomol Inc . (Philadelphia, PA) . Ovalbumin (Grade V), histamine . 1-methylhistamine, biotin, streptavadin, alkaline phosphatase and ldpropanolol were all purchased from SIGMA Chemical Co. (St. Louis, MO). [2,5-3H-histamine dihydrochloride (48 Ci/mmol) and [ 14 C]-S-adenosyl-methionine [C 14-SAM] (58 .9 Ci/mmol) were obtained from New England Nuclear Co . (Boston, MA) . Rabbit anti-guinea pig (GP) IgG1 was purchased from Miles Scientific Co. (New Haven, CT) .
ANTI-OA IgG (ng)
Fig. 2-Anti-OA IgGI standard curve determined by the ELISA .
Mediator Release and Serum IgG1
c
3
60-
-6
50-
-5
1 75
40-
04
w
rn rn -3
-2
10-
LU Z 4 IN =
-1
0-
-0 0 2 4
16
32
64
TIME (min) 3-Time course of antigen-dependent leukotriene synthesis and histamine release in the IgG1 passively-sensitized lung . Guinea-pigs were passively immunized with a single intraperitoneal injection of anti-OA IgG1 and lung tissue was studied after 2 days . Lung fragments (0 .6 g/2 .5 ml) were incubated with 10 µg/ml OA for the time indicated . mean SE, n=4 . Fig.
s0-
8
so
6
T
o LT13 4 • LTD4 x Histamine 60-
-6
a a C 7 01
a
a
3
_ C
p'
40-
-4
-I
-6
C
J
40-
0
FN_
0 la H J
80
C
Cc
S 20-
m J
-2
20
-2
L0
0 0.125 0-
0
I
-5
.3 LOG OA (mg/mi)
4-Antigen concentration dependent leukotriene synthesis and histamine release in the IgG I passively-sensitized lung . Lung fragments (0.6 g/2.5 ml) were incubated with increasing concentrations of OA for 20 min mean ± SE, n = 4 . Fig.
amount of IgG1 administered (0 .125 mg/kg) resulted in the production of significant quantities of each mediator. The amount which caused 50% and maximal stimulation was estimated to be 0 .35 and 0 .5 mg/kg, respectively . As also shown in Figure 5, an amount greater than 0 .5 mg/kg failed to augment the levels of mediators further even though higher serum levels of anti-OA IgG1 were observed (see later) . To determine the duration of sensitization, guineapigs were injected i .p . with a single dose (0 .5 mg/kg) of the antibody and the OA-dependent response was determined at 1, 2, 3, 4, or 7 days post-immunization . As shown in Figure 6, a response was measurable as early as 1-day post-administration and lasted for 7
0 .25
0.5
0.75
ANTI-OA IgG1 (mg/kg)
Fig. 5-Antibody (anti-OA IgG1) dependent leukotriene synthesis and histamine release as a function of sensitization dose. Sensitization involved a single i.p . injection of increasing doses of the antibody and lung tissue was studied after 2 days . Lung fragments (0.6 g/2 .5 ml)were incubated with 10 µg/ml OA for 20 min mean ± SE, n=4 .
days . The level of LTB 4 /LTD4 or histamine at day 4 was significantly less than at day 2 or day 7 (p < 0 .05) . In an additional experiment, antigen-dependent mediator release/synthesis (histamine : 4 .8±0 .4ug/g lung ; LTB 4 : 20 ± 3 ng/g lung ; LTD4 ; 42 ± 10 ng/g lung; n = 4) was still measurable at 14 days after a single injection of 0 .5 mg/kg of IgG1 . Antigen-dependent LT synthesis in perfused and nonperfused preparations Circulating leukocytes from rats, rabbits, and humans synthesize significant amounts of LTB 4 and LTC 4 upon stimulation by various agonists . 2,4 In an attempt to determine if the antigen-dependent synthesis
1 76 Pulmonary Pharmacology 6o-
6
50-Sr
14-
5
N = 6 Y=11 .6x+0.16 0 = 0 .951
12-
4 /
10-
J O
30-
3
W J y
J O
0E Q a
2
20-
/
Oy Z a Q O
I-
J 10-
8-
6-
/
N W a Vi 4-
0
I 1
0 2
3
4
5
6
7
/ 2-
TIME POST SENSITIZATION (day)
X
Fig . 6-The time dependence of passive sensitization for leukotriene synthesis and release . Guinea-pigs were sensitized with 0 .5 mg/kg IgG1 for the time indicated . Lung fragments (0 .6 g/2 .5 ml) were incubated with 10 µg/ml OA for 20 min mean + SE, n=7-21 .
a
02
0
0 .4
0 .6
0 .8
1 .0
1 .2
ANTI-OA IgG1 DOSE (mg/kg, i .p .)
of LTs in the guinea pig lung preparation was dependent on circulating cells, the quantity of LTs found in perfused and non-perfused lung was determined . As shown in Figure 7, lungs that were not perfused with medium synthesized LTB and LTD in an antigen concentration-dependent manner, and the amounts extracted from the medium were close to those found in the perfused-lung preparations . In addition, HPLC analysis of the extracted sample obtained from the non-perfused lung preparation following antigen challenge revealed no distinct peaks co-migrating with 4
/
4
60
50-
Fig. 8-Dose-response relationship between circulating anti-OA IgG1 levels and the amount of IgG1 passively administered . Guinea-pigs were passively immunized with increasing amounts of anti-OA IgG1 and lung tissue studied 2 days later, mean ± SE, n = 8-16 .
known LTB COOH-LTB
4
metabolites (20-OH-LTB
4
or 20-
4)
3. Serum levels of anti-OA IgG1 in IgG1 passivelysensitized guinea pigs Using the above described ELISA, serum anti-OA IgG l levels in passively-sensitized guinea pigs were determined 2 days post-IgG1 administration (Fig . 8) A circulating level of - 1 ug/ml was found in animals sensitized with 0 .075 mg/kg antibody . With increasing amounts of antibody (up to 1 mg/kg), serum levels increased linearly . Figure 9 shows the time course of the circulating antibody levels of anti-OA IgG1 after i .p . administration of 0.5 mg/kg . A plasma level of - 5 ug/ml was detected at 6 h post-administration and this concentration remained constant for up to 7 days . In naive guinea-pigs, no circulating level of anti-OA IgGI was demonstrable .
10-
0-
4 . Leukotriene synthesis and histamine release in actively-sensitized guinea pig lung I .5
I -4
I .3
I -2
I -1 .5
LOG OA (mg/ml)
Figure 10 demonstrates that lungs taken from actively-sensitized guinea-pigs synthesize LTB4/LTD and release histamine as a function of time, and Figure I1 illustrates the antigen concentration response relationship. The time course and antigen concentration-response curve for histamine release and LTB4/D synthesis were similar to those obtained from the passively-sensitized lungs . More importantly, the amount of LTB , LTD synthesized and
4
Fig. 7-Comparison of antigen-dependent leukotriene synthesis in perfused and non-perfused lung preparations . For perfused preparations, lungs were perfused via the pulmonary artery with 20 ml cold Krebs-Ringer medium before the lung parenchyma was chopped into 1 mm' tissue fragments . The fragments were then washed 5 x with cold medium . For non-perfused preparations, the lung was neither perfused nor washed with the medium . The lung fragments (0 .6 g/2 .5 ml) were incubated with 10 µg/ml OA for 20 min mean ± SE, n = 4 .
4
4
4
Mediator Release and Serum IgG1 8-
1 77
-8
-6
-4
-2
.5
I -4
I -3
I -2
0
I -1 .5
LOG OA (mg/ml)
0
I
I
I
I
I
I
1
2
3
4
5
6
I 7
TIME POST SENSITIZATION (day)
Fig. 11-Antigen concentration-dependent leukotriene synthesis and histamine release in the actively-sensitized lung . Lung fragments (0 .6 g/2.5 ml) were incubated with increasing concentrations of OA for 20 min, mean ± SE, n = 4 . (38)
Fig. 9-Circulating anti-OA IgGI levels as a function of time post-sensitization. Guinea-pigs were passively immunized with 0 .5 mg/kg anti-OA IgGI and then serially bled at the times indicated, mean ± SE, n = 3-7 . 60
6
so
-6
3 d E rn
1
g 40
p
Y
-3
30
j0
m
-2
20
10-
-1
10
0-
PASSIVE SENSITIZATION 0
0 024
8
16
32
0 .5 mg/kg IgG1
64
ACTIVE SENSITIZATION 1 mg/kg OA followed by 0.25 x 3 mg/kg OA
TIME (min)
Fig. 10-Time course of antigen-dependent leukotriene synthesis and histamine release in the actively-sensitized lung . The active sensitization protocol involved multiple antigen injections as described previously . 1- ,14 In brief, guinea-pigs were injected s .c. with 1 mg/kg OA . A total of 14 days later, the animals were given weekly injections of 0 .25 mg/kg OA for 3 weeks and then sacrificed 3-8 days after the last injection . Lung fragments (0 .6 g/2 .5 ml) were incubated with 10 µg/ml OA for the times indicated, mean±SE, n=4.
histamine released was also similar between both passively- and actively-sensitized lungs . 5. Comparison of serum anti-OA IgG1 levels in passively- and actively-sensitized guinea pigs The serum antibody levels of 16 passively- and 38 actively-sensitized guinea-pigs were measured according to the procedure shown in Table 1 . Figure 12 illustrates that the levels of anti-OA IgG1 in activelysensitized guinea-pigs were 130-fold higher than those detected in passively-sensitized animals . In addition, the antibody levels relative to the mean value in the actively sensitized group (co-efficient of varia-
Fig. 12-Comparison of serum anti-OA IgG 1 levels in passivelyand actively-sensitized guinea-pigs . The immunization protocol involved either a single antibody injection (passive senzitization) or multiple antigen injections (active sensitization). Sensitization time was 2 days for passive sensitization and 31-36 days for active sensitization .
tion = 145 .2%) were more variable than those in the passively-sensitized group (31%) . Since the rabbit anti-GP IgG1 used in the ELISA shows -30% cross-reactivity with an OA-specific IgG2 subclass of antibody and anti-OA IgG2 is likely to be present in the active immunization procedure, the levels in actively-sensitized guinea-pigs could have been overestimated by 30% . Taking this into consideration, serum levels of the IgG1 antibody in actively-sensitized animals were still significantly (> 90-fold) higher (p < 0 .05) .
DISCUSSION In this study an antigen-specific ELISA was used to demonstrate that :
178
Pulmonary Pharmacology
a) the circulating levels of IgG1 increase linearly with the quantity of antibody passively administered to naive guinea pigs (2 days post i .p . dose) (Fig . 8) and b) the level is maintained for at least 7 days (Fig . 9) . Lung sensitization was observable at 0 .125 mg/kg IgG1 corresponding to a plasma level of -2 ug/ml (Figs . 5 & 8) . Moreover, the circulating antibody level at the dose (0 .5 mg/kg) that caused maximal sensitization was .- 6 ug/ml . These results indicate that 2-6 gg/ml of IgGI in plasma is sufficient to sensitize lung tissue . It appears that the antigen-dependent response is maximal at a circulating antibody level of 6 ug/ml, because increased IgG1 levels obtained using a higher amount of antibody to sensitize animals (0 .75 mg/kg) is not accompanied by enhanced mediator release/ synthesis (Fig . 5) . Both requirements for low plasma levels of IgGI for sensitization and the maintenance of antibody levels in circulation (long serum half-life) may partly explain the present and prior observation' 1-13,20,21 that once guinea-pigs are immunized with this subclass of antibody the sensitization lasts for 7 days and beyond . In the experiment where the degree of sensitization time was examined (Fig. 6), there was decreased mediator production at day 4 post-antibody administration . The reason for this decrease is currently unclear . As demonstrated by these studies, passively-sensitized lung tissue produced a significant amount of LTB 4/LTD 4 and histamine upon antigen challenge ex vivo . In fact, in respect to both kinetic and antigen dose-response parameters, the amount of histamine released and LTB4/D4 synthesized was comparable to that generated from lung fragments of actively-sensitized guinea pigs (Fig . 3 vs 10 & Fig . 4 vs 11) . Because IgG1 levels in the latter animals far exceed those required for maximal sensitization, an induction of this antibody subclass alone by the active immunization could account for the antigen-dependent response . The present studies have characterized mediator production from IgG1-sensitized lung fragments with regards to both the dose-response relationship for both antigen and antibody and the kinetics of their interaction . The profile and the quantity of antigenstimulated LT synthesis and histamine release by sensitized lungs agree with previously published results . 2, t 3,22,23 LTD 4 is a relatively stable terminal product of the 5-lipoxygenase pathway in the chopped lung preparation (Figs . 3 & 10) . At 20 min incubation time, no evidence for presence of LTC 4 and LTE4 (< 0 .7 ng/g lung) was found . This is consistent with the result obtained from actively-sensitized guinea-pig lungs .' LTB 4 has also been shown in this study to be another stable product because (1) 20-OH-LTB 4 or 20-COOH-LTB 4 could not be detected in the sample extracted from the supernatant of antigen-challenged lung and (2) the level of LTB4 (following antigen
challenge) was constant for up to 64 min . Guinea-pig lung fragments could have contained an insufficient number of polymorphonuclear leukocytes thus reducing its capacity to degrade LTB 4 as shown in a recent study using rabbit lungs .24 This possibility seems to be unlikely because each LTB 4 metabolite could again not be detected in non-perfused lung preparations . At 20 min post-antigen challenge, the molar ratio of LTD4/LTB4 generated from IgGI passivelysensitized lungs was about 3 (corrected according to recovery) . A similar ratio for these mediators was also released from actively-sensitized lungs (Figs . 10 & 11) . These results indicate that resident lung cells in guinea-pigs may have a greater capacity to synthesize peptidoleukotrienes than LTB 4 following immunological challenge . This possibility is supported by the presence of a highly specific LTC 4 synthetase in guinea-pig lung membrane . 25 A greater increase in both enzymatic transformation of LTA 4 to LTC4 and production of LTC 4 relative to LTB4 formation has also been seen in the lungs of rats 26,2 ' and rabbits .28 .29 While human lung tissue produced mostly LTD 4 in response to anti-IgE challenge ex vivo, 30-32 antigen challenge caused equal molar production of LTB4 and LTD 4 from human lungs that were passively-sensitized with hyperimmune serum in vitro . 33 No difference was observable between the response of the perfused and non-perfused preparations, suggesting that FcR-bearing resident cells such as macrophages 34,35 and possibly mast cells 36,3 ' rather than circulating cells are the cell types responsible for LT synthesis in the passively-sensitized guinea-pigs . However, this suggestion must be tentative because many circulating cells are fixed to the lung via specific adhesion molecules and thus may not wash off easily . In summary, the passive administration of purified antigen specific (OA) IgG1 to normal guinea-pigs allowed a demonstration of a direct relationship between the circulating level of antibody (anti-OA IgG1) and the capacity of antigen (OA) to induce histamine and leukotriene (LTB 4/LTD 4) synthesis from lung tissue ex vivo . The level of circulating antibody required to maximally sensitize lung tissue was approximately 6 gg/ml, corresponding to a plasma concentration of - 40 nM suggesting a high affinity binding of IgG1 to its receptors . That such low levels of antibody appear to be sufficient for maximum tissue sensitization raises interesting questions regarding the potential blocking effects of other classes of antigen-specific antibody 38 or non-specific IgG1 competing at surface receptors . These questions and others related to the distribution of FcR for IgG 1 in guinea-pig lung are currently under investigation . References 1 . Anderson P, Antigen-induced bronchial anaphylaxis in actively sensitized guinea pigs Anti-anaphylactic effects of sodium cromoglycate and aminophylline . Brit J Pharmacol 1980a; 69: 467-472 .
Mediator Release and Serum IgG1 2 . Piper P J. Formation and actions of leukotrienes . Physiol . Rev. 1984 ; 64 : 744-761 . 3 . Muccitelli R M, Tucker S S, Hay D W P, Torphy T J, Wasserman M A . Is the guinea pig trachea a good in vitro model of human large and central airways? Comparison on leukotriene-, methacholine- and antigen-induced contractions . J Pharmacol Exp Therap 1987 ; 243 : 467-473 . 4 . Bach M K. Lipid mediators of hypersensitivity . Prog Allergy 1988 ; 44 : 10-98 . 5 . Cheng J B, Eskra J D, Pillar J . Comparison of antigen and Ca ++ -ionophore-induced peptidoleukotriene release from guinea pig lung preparations using highperformance liquid chromatography . J Pharmacol Exp Therap 1987a; 241 : 786-792. 6. Undem B J, Bucker C K, Harley P, Graziano F M . Smooth muscle contraction and release of histamine and slow-reacting substance of anaphylaxis in pulmonary tissues isolated from guinea pigs passively sensitized with IgG1 or IgE antibodies . Am Rev Resp Dis 1985 ; 131 : 260-266 . 7 . Regal J F. IgG vs IgE mediators of antigen-induced guinea pig tracheal contraction . Immunopharmacol. 1984; 8 : 111-119 . 8 . Regal J F. IgG vs. IgE mediators of antigen-induced guinea pig lung parenchymal contraction . Immunopharmacol . 1985 ; 10 : 137-146 . 9 . Conklyn M J, Showell H J. Purification of antigenspecific reaginic IgG1 from ovalbumin-sensitized guinea pigs. New England Pharmacologists 6th Annual Meeting (Abstract) . 987; p . 25 . 10 . Regal J F . Immunoglobulin G- and immunoglobulin E-mediated airway smooth muscle contraction in the guinea pig. J Pharmacol Exp Therap 983 ; 228 : 6-20 . 11 . Benacerraf F, Ovary Z, Bloch K J, Franklin E C. Properties of guinea pig 7S antibodies . I . Electrophoretic separation of two types of guinea-pig 7S antibodies . J Exp Med 963 ; 7 : 937-950. 12 . Ovary Z, Benacerraf B, Bloch K J . Properties of guinea pig antibodies . II Identification of antibodies involved in passive cutaneous and systemic anaphylaxis . J Exp Med 963 ; 7 : 95-964 . 13 . Graziano F M, Gundersen L, Larson L A, Harley P, Buckner C K . Receptor-specific mediation by immunoglobulin E of antigen-induced contraction of tracheal and lung parenchymal strips isolated from the guinea pig. J Clin Invest 1984 ; 73: 1215-1222 . 14 . Souhrada M, Souhrada J F . Potentiation of Na + -electrogenic pump of airway smooth muscle by sensitization. Respir Physiol 1982 ; 47 : 69-81 . 15 . Eskra J D, Pereira M J, Ernest M J . Solid-phase extraction and high-performance liquid chromatography analysis of lipoxygenase pathway products . Anal Biochem 1986; 154: 332-337 . 16 . Sweeney F J, Eskra J D, Carty T J. Development of system for evaluating 5-lipoxygenase inhibitors using human whole blood . Prostaglandins Leukotrienes Med 1987; 28 : 73-93 . 17 . Shaff R E, Beaven M A . Increased sensitivity of the enzymatic isotopic assay of histamine: Measurement of histamine in plasma and serum . Analyt Biochem . 1979; 94: 425-430 . 18 . Sun F F, McGuire J C . Metabolism of arachidonic acid by human neutrophils . Characterization of the enzymatic reactions that lead to the synthesis of leukotriene B4 . Biochem Biophys Acta 1984; 794 : 56-64 . 19. Shak S, Goldstein I M . w-Oxidation is the major pathway for catabolism of leukotriene B 4 in human polymorphonuclear leukocytes . J Biol Chem 1984; 259 :10181-10187 . 20 . Andersson P . Antigen-induced bronchial anaphylaxis in actively sensitized guinea pigs . Pattern of response in relation to immunization regimen . Allergy 1980b ; 35 : 65-71 . 21 . Fleisch J H, Spaethe S M . Improved methodology for evaluating release of mediators of anaphylaxis from guinea pig lung. Monogr Allergy 1979 ; 14: 185-188 . 22. Clancy R M, Hugh T E . The extraction of leukotri-
enes (LTC4, LTD, and LTE 4) from tissue fluids : The metabolism of these mediators from IgE-dependent hypersensitivity reaction in lung. Anal Biochem 1983; 133 : 30-39. 23 . Turner N C, Dollery C T . Release of arachidonic acid metabolite and histamine from sensitized guinea pig lung following antigen challenge . Brit J Pharmacol 1988 ; 93 : 751-758 . 24 . Grimminger F, Menger M, Becker G, Seeger W . Potentiation of leukotriene production following sequestration of neutrophils in isolated lungs : Indirect evidence for intercellular leukotriene A 4 transfer . Blood 1988a ; 72 : 1687-1692 . 25 . Yoshimoto T, Soberman R J, Spur B, Austen K F . Properties of highly purified leukotriene C 4 synthase of guinea pig lung. J Clin Invest 1988 ; 81 : 866-871 . 26 . Medina J F, Haeggstrom J, Kumlin M, Radmark O . Leukotriene A4 : Metabolism in different rat tissues . Biochem Biophys Acta 1988; 961 : 203-212 . 27. McDonnell T J, Chang S-W, Westcott J Y, Voelkel N F. Role of oxidants, eicosanoids, and neutrophils in amphotericin B lung injury in rats. J Appl Physiol 1988 ; 65 : 2195-2206. 28 . Schultz R, Seeger W . Release of leukotrienes into the perfusate of calcium-ionophore stimulated rabbit lungs : Influence of 5-lipoxygenase inhibitors. Biochem Pharmacol 1986; 35 : 183-193 . 29 . Grimminger F, Becker G, Seeger W . High yield enzymatic conversion of intravascular leukotriene A 4 in the blood-free perfused lungs . J Immunol 1988b (In Press) . 30 . Lewis R A, Austen K R, Drazen J M, Clark D A, Marfat A, Corey E J . Slow reacting substance of analphylaxis : Identification of leukotriene C-1 and D from human and rat sources . Proc Natl Acad Sci 1980; USA 77 : 3710-3714 . 31 . Kumlin M, Dahlen S E, Granstrom E . Differences in the formation and metabolism of leukotrienes B4 and C 4 in the human lung . Adv Prostaglandin Thromboxane Leukotriene Res 1987 ; 17: 1018-1022 . 32 . Vigano T, Toia A, Galli G, Berti F, Crivellari M T, Folco G C . Adenosine and eicosanoid release from immunologically challenged human lung fragments . Adv Protaglandin Thromboxane Leukotriene Res 1987; 17 : 992-996 . 33 . Salari H, Borgeat P, Fournier M, Hebert J, Pelletier G. Studies on the release of leukotrienes and histamine by human lung parenchymal and bronchial fragments upon immunologic and nonimmunologic stimulation : Effects of nordihydroguaiaretic acid, aspirin, and sodium cromoglycate . J Exp Med 1985 ; 162 : 1904-1915. 34 . Rankin J A, Hitchcock M, Merrill W, Bach M K, Brashler J R, Askenase P W . IgE-dependent release of leukotriene C4 from alveolar macrophages . Nature (Lond.) 1982 ; 297: 329-331 . 35 . Nagel G P, Young K R, Jr, Reynolds H Y . Receptors for human IgG subclasses on human alveolar macrophages . Am Rev Respir Dis 1984 ; 129 : 413-418 . 36. Peters S P, MacGlashan D W, Jr, Schulman E S, Schleimer R P, Hayes E D, Rokach J, Adkinson N F, Jr, Lichtenstein L M . Arachidonic acid metabolism in purified human lung mast cells . J Immunol 1984; 132 : 1972-1979 . 37 . Lewis R A, Austen K F . Mediation of local homostasis and inflammation by leukotrienes and other mast cell-dependent compounds . Nature 1981 ; (Lond .) 293 : 103-108. 38 . Yamauchi N, Ito K, Suko M, Ishii A, Miyamoto T . IgG2 antibodies block IgE antibody-induced asthma in guinea pigs . Int Archs Allergy AppI Immun 1986 ; 80: 76-80 . Date received: 6 August 1989 Date revised: 20 October 1989 Date accepted: 10 March 1990
1 79
