Acute Effects of Interleukin-2 on Lung Mechanics and Airway Responsiveness in Rats1- 4
PAOLO M. RENZI, TAO DU, SANTO SAPIENZA, NAI SAN WANG, and JAMES G. MARTIN
Introduction
The discovery of interleukin-2 (IL-2; T-cell growth factor) was an important breakthrough in T-celland tumor immunology. The availability of large quantities of recombinant human IL-2 has led to the use of this lymphokine in several clinical conditions. IL-2 has been employed for the treatment of immunodeficiency (1) and in association with lymphokine-activated killer cells for the treatment of cancer (2). However, major side effects have been reported. Pulmonary complications of treatment include the induction of pulmonary edema and, less often, bronchospasm (2). The development of bronchospasm after the administration of IL-2 for the treatment of cancer, in conjunction with in vitro evidence of abnormal regulation of IL-2 production by lymphocytes of asthmatics (3-5), prompted us to examine the effect of IL-2 on airway responsiveness. Wewished to establish whether alterations in airway responsivenesscould be caused by IL-2 in normal animals, and if so whether there was any interaction between the inherent airway responsiveness of the animals and the effects of IL-2. The aim of the present report was to study the effects of an intravenous administration of a nonlethal dose of IL-2 on lung mechanics and on airway responsiveness. We report the effects of an intravenous infusion of 750,000 units/kg of IL-2 to two strains of rats differing in airway responsiveness to methacholine (6), on ventilation, airway responsiveness, and lung histology. Methods Preparation of Animals and Measurement of Pulmonary Mechanics Twelvepathogen-free male Lewisand 13male Fisher 344 rats (Charles River, Canada; Mississauga, Ont) 6 to 8 wk of age were anesthetized intraperitoneally with sodium pentobarbital (50 mg/kg) or urethane (1 g/kg). Blind orotracheal intubation was then performed using a 6-cm length of PE-240 polyethylene catheter. A heating pad was used to maintain 380
SUMMARY We studied the acute effects of Interleukln-2 (IL-2), the prlnclpallymphoklne responsible for lymphocyte proliferation, on lung mechanics and airway responsiveness to methacholine (MCh) In rats. lewis (n = 12) and Fisher 344 (n = 13) rats were anesthetized and Intubated, and Intravenous and Intra-arterial lines were Inserted. IL-2 (750,000U/kg) was Infused Intravenously over 2 to 4 min Into seven lewis and seven Fisher rats, and vehicle alone was administered to five lewis and six Fisher rats. Blood pressure, heart rate, respiratory frequency (f), tidal volume (VT), minute ventilation (VE), and lung resistance (RL) were measured before and every 5 min for 45 min after the Infusion of IL-2. Lung compliance was measured before and 30 min after IL-2. Bronchial provocation testing with MCh was performed 45 min after the Infusion of IL-2. Subsequently, the animals were exsanguinated, and the lungs were removed for histologic examination. Infused IL-2 did not alter heart rate or blood pressure. VT, f, VE, and RL Increased significantly by 15 min (p < 0.05), but they returned to baseline by 45 min. Lung compliance decreased significantly In both rat strains. IL-2 Increased airway responsiveness only In lewis rats; the concentration of MCh that caused a 0.003) In IL-2-treated and control rats, doubling of RL (EC2ooRL) was 0.6 mg/ml and 4.3 mg/ml (p respectively. The airway responsiveness did not change significantly In Fisher rats; EC200RL was 0.09). Histologic examination 0.13and 0.35 mg/ml for IL-2-treated and control rats, respectively (p showed significant separation of the epithelium from the basement membrane and bronchovascular edema In both rat strains. The release of IL-2 may contribute to the altered airway responsiveness that follows antigen presentation and activation of airway lymphocytes.
=
=
AM REV RESPIR DIS 1991; 143:380-385
constant body temperature. The external jugular vein and internal carotid arteries were cannulated with PE-50 polyethylene tubing for infusion of drugs and measurement of blood pressure, respectively. Intubated and catheterized animals were kept in the decubitus position to measure lung compliance. Transpulmonary pressure (Ptp) was measured using a water-filled catheter placed in the lower third of the esophagus connected to one port of a differential pressure transducer (268 BC; Sanborn Co., Waltham, MA). The other port was connected to the endotracheal tube during measurement of pressure-volume curves of lungs or during tidal breathing to a Plexiglas'[ box used to measure tidal airflow (see below). The esophageal catheter consisted of a polyethylene tube (PE240, 20 em long) with a terminal tip (6 em) of a smaller bore tube (PE-I60). The time constant of the catheter-transducer system was 14 ms, which results in < 10070 loss of amplitude at 5 Hz. To measure the pressure-volume curves of the lung, the lungs were inflated three times with air from FRC to a Ptp of 30 cm H 2 0 using a calibrated syringe. After the third inflation, the lungs were deflated by 0.5 ml decrements until the animal resumed spontaneous breathing or the FRC was reached. These measures were made three times. Lung
compliance was calculated as the slope of the pressure-volume curve between a Ptp of 5 and 10 em H 2 0 and expressed as ml/cm H 2 0 .
Measurement of Heart Rate and Blood Pressure A saline-filled PE 50 catheter was placed in the carotid artery and connected to one port of a differential pressure transducer (268 BC; Sanborn). Values for heart rate and mean blood pressure were determined from the pressure tracing. Mean blood pressure (BP) was calculated from systolic and diastolic BP using
(Received in original form April 6, 1990 and in revised form July 16, 1990) 1 From the Meakins Christie Laboratories and the Department of Pathology, Royal Victoria Hospital, McGill University, and the Pulmonary Service,St-Luc Hospital, Universityof Montreal, Montreal, Quebec, Canada. 2 Supported by Grant MA-I0637 from the Medical Research Council of Canada. J Presented in part at the Annual Meeting of-the American Physiological Society, New Orleans, March 1989. 4 Correspondence and requests for reprints should be addressed to Dr. P. M. Renzi, MeakinsChristie Laboratories, 3626 St-Urbain Street, Montreal, Quebec, Canada H2X 2P2.
381
INTERLEUKIN-2 AND WNG MECHANICS AND RESPONSIVENESS
the formula: Mean BP = diastolic BP + (systolic - diastolic BP)/3. Measurement of Pulmonary Resistance Pulmonary resistance was measured during spontaneous tidal breathing with the animals in the left lateral decubitus position as previously described (7). Flow was measured by placing the tip of the tracheal tube inside a small Plexiglas box (265 ml in volume). A pneumotachograph (model No. 0, Fleisch, Lausanne, Switzerland) coupled to a differential transducer (MP-45, ± 2 cm H 20; Validyne Corp., Northridge, CAl was attached to the other end of the box to measure airflow, and volume was obtained by electrical integration of the flow signal using a respiratory integrator (HP8815A; Hewlett-Packard, Waltham, MA). Tidal volume (VT), frequency of breathing (0, and minute ventilation (VE) were also calculated. All signals were amplified and recorded on an eight-channel strip chart recorder (HP 7758 B; Hewlett-Packard). Pulmonary resistance (RL) was obtained by the electrical subtraction method of Mead and Whittenberger (8). Endotracheal tube resistance was 0.11 em H 20/ml/s at the flow of 25 ml/s and was almost constant over a range 0 f flows from zero to 25 ml/s. Tube resistance was subtracted from all values of RL. Study Protocol Human recombinant IL-2 (750,000 U/kg) or a control vehicle (containing 50 mM acetic acid and 8 ng/ml Escherichia coli endotoxin) was infused intravenously through a constant flow pump (Harvard infusion pump series 940; Harvard Apparatus Co., South Natick, MA) at a rate of 0.13 ml/min. The control vehicle was prepared to be comparable to the vehicle for IL-2, which contained 50 mM acetic acid and less than 8 ng/ml endotoxin. In preliminary experiments (n = 2), 1.1 million U/kg of IL-2 in 0.4 ml was infused intravenously. This led to a substantial increase in VE, RL, and airway responsiveness. In order to avoid the complicating influence of increased VE and RL on the measurement of airway responsiveness to methacholine (MCh), we chose a dose of 750,000 U/kg of IL-2 and waited till measurements of VE and RL had returned to baseline before performing a MCh challenge. Seven Lewis and seven Fisher rats received IL-2, and five Lewis and six Fisher rats receivedthe vehicle. Lung compliance was measured at baseline and at 30 min after the intravenous infusion. Heart rate, BP, f, VT, VE, and RL were measured at baseline at the end of the intravenous infusion (time 0), every 5 min for 30 min, and at 45 min. A MCh challenge was performed at 45 min after which the animals were killed by exsanguination. Methacholine Challenge Aerosols were generated with a Hudson nebulizer containing 3 ml of solution and driven by a compressed air source at an airflow of 7.5 L/min. The nebulizer output was 0.18
ml/min. The nebulizer was connected to one side port of the Plexiglas box; during aerosolization, airflow was diverted through a second side port. To assess airway responsiveness to MCh, rats inhaled aerosols of normal saline and progressively doubling concentrations of MCh for 30 s; MCh concentrations ranged from 0.063 to 16mg/ml in normal saline. RL was measured before and after inhalation of saline and after each of the concentrations of MCh. We measured the peak value of RLafter each administration. An interval of approximately 3 min elapsed between each concentration of MCh. Airway responsiveness was calculated as the concentration of MCh required to double RL (EC 2ooRL) and was determined by linear interpolation between the two concentrations (in logarithms) bounding the point at which RLreached 200070 of control. We used the value of RL after saline as the control because saline did not significantly alter RL. Assessment of Pathologic Changes in the Lungs The left lung was fixed for 48 h in 10070 formalin at a pressure of 25 em H 20. Slices from paraffin-embedded sections of the left lung were stained with hematoxylin-eosin. The bronchi, veins, arteries, and alveoli of each animal were assessed for pathologic changes by two observers blinded to the group status of the specimens. The extent of edema, cellular infiltration, and epithelial detachment were assessedwith a scoring system.A scoreof zero was given if none of the specific structures examined had the changes looked for: 1+ if > 0 but ~ 25070, 2 + if > 25 but ~ 50070, and 3 + if > 50070 of the specific structures had the changes. There was good interobserver reproducibility of the scoring (r = 0.917; Spearman's rank test). The wet/dry weight ratio was obtained by weighing the right lung before and after drying for 48 h at 56° C. Statistical Analysis Group results are expressed as mean ± 1 SEM, except for values of EC 2ooRL, which are reported as geometric means. Statistical analyses involving EC 20 0 RL were performed with the log-transformed data. The significance of differences between test (IL-2) and control groups was evaluated using unpaired t tests except for pathologic scores, which were evaluated by the Mann-Whitney
U test. Significance was considered to be established for values of p < 0.05. Chemicals Human recombinant IL-2 was a gift of the Glaxo Institute of Molecular Biology (Geneva, Switzerland). Methacholine chloride and urethane were purchased from Sigrna Chemical Co. (St. Louis, MO). Pentobarbital was purchased from Canada Packers (Mississauga, Ont).
Results
Baseline Characteristics Seven Lewis rats (239 ± 17 g) received IL-2, and five control rats (225 ± 20 g) received the vehicle. Seven Fisher rats (189 ± 5 g) received IL-2, and six control rats (177 ± 7 g) received the vehicle. One Fisher rat died 20 min after the infusion of IL-2 and was eliminated from analysis. Baseline RL prior to the infusion of IL-2 or vehicle (preinfusion) and 45 min later (postinfusion) are presented in table 1. No significant difference was found between RL of IL-2- and vehicle-treated Lewis rats. Although Fisher rats treated with IL-2 tended to have higher RL than did vehicle-treated rats prior to MCh challenge, this difference was not statistically significant (0.14 ± 0.01 versus 0.08 ± 0.02, p = 0.07). Effect of IL-2 on Physiologic Parameters The administration of IL-2 did not significantly affect heart rate or BP. The heart rate was 331 ± 17 and 341 ± 12 beats/min, and the mean BP was 108 ± 8 and 112 ± 11 mm Hg at baseline and 15 min after IL-2 infusion, respectively. No differences were found in baseline VE, f, and VT in rats treated with IL-2 or with vehicle (table 2). IL-2 increased VE in both rat strains (figure lA). VE peaked immediately after the infusion in Fisher rats and at 15 min in Lewis rats. At 45 min, no differences were found in VE between the groups. At 45 min, VE was 147 ± 17and 137 ± 28 ml/min in Lewis rats receiving IL-2 and in those receiving ve-
TABLE 1 EFFECT OF IL-2 ON LUNG RESISTANCE AND COMPLIANCE* Compliance (ml/em H2O)
Resistance (em H20/mils) Preinfusion Lewis IL-2 Lewis vehicle Fisher IL-2 Fisher vehicle
0.1 0.09 0.12 0.11
± 0.02 ± 0.02
± 0.03 ± 0.02
* Results are presented as mean ± SEM.
Postinfusion 0.1 0.09 0.14 0.08
± ± ± ±
0.02 0.03 0.01 0.02
Preinfusion 0.51 0.50 0.34 0.29
± ± ± ±
0.08 0.08 0.05 0.05
Postinfusion
Value
0.04 0.04 0.04 0.05
0.04 0.35 0.03 0.38
0.37 0.44 0.25 0.32
± ± ± ±
RENZI, DU, SAPIENZA, WANG, AND MARTIN
382
hicle, respectively, and was 113 ± 10and 140 ± 16mllmin in Fisher rats receiving IL-2 and in those receiving vehicle, respectively. IL-2 increased f in both Lewis and Fisher rats (figure IB); f peaked at 10 min in Lewis rats and immediately after the infusion in Fisher rats, and then returned towards baseline over the subsequent 45 min in both. IL-2 in-
TABLE 2 BASELINE PHYSIOLOGIC PARAMETERS OF LEWIS AND FISHER RATS
Ve Lewis control Lewis IL-2 Fisher control Fisher IL-2
175 129 139 100
± ± ± ±
VT
f (breaths/min)
(ml/min)
40 12 15 13
97 85 99 85
± ± ± ±
5 7 10 5
(m/)
1.8 1.5 1.4 1.2
± ± ± ±
0.5 0.2 0.2 0.2
A 200
LEWIS
200 180
z
Z
0
~wll10
0 160 ~~
~
Ld 140
~~140
>lD 120 III~ ~ of 100
... 0 1 0 0
80
~80 :I
c(w
.J~
~~
zi
FISHER 344
180
ZW
~~120
creased VT in both strains (figure IC), and it followed a similar time course as that of f. Results for peak measurements of VE, f, and VT were significantly higher in the IL-2-treated group than in the vehicle-treated group.
Effect of IL -2 on Lung Resistance and Complian ce The adm inistration of IL-2 increased RL in both Lewis and Fisher rats (figure 2). The peak in RLoccurred immediately after the administration of IL-2 in Fisher rats and at 30 min in Lewis rats. Lung resistance returned to baseline by 45 min. Lung compliance was measured prior to infusing IL-2 or vehicle (preinfusion) and 30 min later (postinfusion) . IL-2 significantly decreased lung compliance in Lewis and Fisher rats (table 1). In contrast, the lung compliance of animals treated with vehicle did not change significantly.
wu.
"='.".
110 ·10
0
10
20
30
40
60 ·10
50
0
10
20
TI ME
30
40
50
TIME
Effect ofIL-2 on Airway Responsiveness to Methacholine IL-2 significantly increased airway responsiveness to MCh in Lewis rats (figure 3). The geometric means of the
B 150
>
FISHER 344
15 0
LEWIS
'80
140 > U Z 13 0 w "=' ~ O W 120 w~ a: .J ... W 110
140
U
Z 130 w "=,Ow w~ 120 a:.J ... w ~~ 110 ZlD -u. 100 ~o C."w- 90 a: III 80
160
w u
z
~
~~ ZlD
-XO u. 10 0 ~ .". 90 Wa: 80 1lI
70 ·10
,,, W
s
~ ~
' 20
t
' 00
~
~
80
60
70 0
20
10
30
40
-1 0
50
20
10
0
30
40
.1 0
50
TIME
TIME
' 80
C LEWIS
160
150
150
140
140
w u
_ 130 WW ::I ~ "='-' 120 .Jw o~ 110 >lD .Ju. CO 100
~
W
130
~
~
...e -~
fiS HE R 344
Ww
FISHER 344 z
~ ~ 120 .J W
g ~eo ;;i
~ 100
...e .".
90
110
-
'"
~ ~
120
3 t
100
80
90
60
. 10
80
80
~
'0
20
30
'0
TI ME
70
70 ·10
0
10
20 TIME
30
40
50
- 10
0
10
20
30
40
50
TIME
Fig. 1. Effect of an intravenous infusion of 750,000 U/kg of IL-2 on minute ventilation, respiratory frequency, and tidal volume : Time 0 is the end of the infus ion of IL·2 . Minute ventilation (A), breathing frequency (B), and tidal volume (C) are expressed as a percentage of baseline ± SEM. Open squares = Lewis animals receiving IL-2 (n = 7). Closed squares = Lewis animals rece iving vehicle (n = 5). Dotted squares = Fisher animals receiving IL·2 (n = 6). Diamonds = Fisher animals receiving vehicle (n = 6).
Fig. 2. Effect of an intravenous infusion of 750,000 Ulkg of IL·2 on lung resistance: Time 0 is the end of the infusion of IL·2 or vehicle. Lung resistance is expressed as percent of baseline ± SEM. Open squa res = IL-2· treated Lewis animals (n = 7). Closed squares vehicle-treated Lewis animals (n = 5). Dotted squares = IL·2·treated Fisher animals (n = 6). Diamond vehicle-treated Fishe r anima ls (n = 6).
=
INTERLEUKIN·2 AND WNG MECHANICS AND RESPONSIVENESS
383
16
o
6
+
4
Fig. 3. Effect of an intravenous infusion of 750,000Ulkg of IL-2 on airwayresponsiveness to methacholine: Results are expressed as EC2••RL(concentration of methacholine causinga doublingin lung resistance). The horizontal bars represent the geometric means.
2 II: ...J
g ....
U W
z: _ E
.5
Cl
E
0 0
o o
.2 5
....-
0
o o
....0-
S
. 12
-00
.0 6 BAS
EC 200 RL were 0.6 ± 0.2 and 4.3 ± 1.4 in animals receiving IL-2 and in those receiving vehicle, respectively (p = 0.003). In Fisher rats, IL-2 also tended to increase airway responsiveness to MCh. The geometric means of the EC 20 0 RL were 0.13 ± 0.03 and 0.35 ± 0.25 in the animals receiving IL-2 and in those receiving vehicle, respectively (p = 0.09).
Effect of IL-2 on Histopathologic Findings The major histopathologic changes were epithelial detachment and peribronchial and perivascular edema (figure 4). There was little, if any, cellular infiltration of parenchyma, airways, or vessels. Similarly, alveolar edema was not prominent in either vehicle- or IL-2-treated groups. The scores for epithelial detachment and edema of the bronchi, arteries, and veins and airway responsiveness to MCh are presented for Lewis rats treated with 750,000 U/kg IL-2 or with vehicle in table 3. IL-2 significantly increased epithelial detachment in both Lewis and Fisher rats. The average epithelial detachment scores were 1.6 ± 0.3 and 1.0 ± 0.1 for the animals treated with IL-2 and for those treated with vehicle, respectively (p < 0.05). Correlations were performed between the log EC 200 RL and epithelial detachment scores of Lewis and Fisher rats. A significant correlation was found between the epithelial detachment score and airway responsiveness for Lewis rats (r = 0.837, p < 0.05). However, no significant correlation was found for Fisher rats (r = 0.424, p > 0.05). The intravenous administration of IL-2 to Fisher rats increased bronchovascular edema. The average bronchovascular edema scores were 1.7 ± 0.2 versus 1.2 ± 0.1 for the animals treated with IL-2 and for those treated with vehicle, respectively (p < 0.05). However, there was no difference in edema between IL-2-
0
0
o LEW IL-2
8
LEWVEH
F344 IL-2
F344 VEH
and vehicle-treated Lewis rats; the average edema scores were 1.6 ± 0.2 and 1.3 ± 0.2 for the Lewis animals treated with IL-2 and for those treated with vehicle, respectively.
Effect of IL-2 on Lung Wet/Dry Weight Ratio IL-2 caused an increase in the lung wet/dry weight ratio in Fisher rats (figure 5); the ratio was 5.1 ± 0.1 and 4.3 ± 0.1 for IL-2- and vehicle-treated animals, respectively (p < 0.005). No difference was found between the lung wet/dry weight ratio of Lewis animals treated with IL-2 (5.1 ± 0.2) or with vehicle (5.2 ± 0.2). In an additional five Lewis rats, we measured wet-to-dry weight ratios after animals received vehicle intravenously and were challenged with aerosolized saline. The wet/dry weight ratio of these animals was 4.4 ± 0.1, not different from that of vehicletreated Fisher rats but significantly lower than the wet/dry weight ratio of Lewis rats challenged with IL-2. Discussion
We investigated the effects of an acute intravenous administration of IL-2 on lung mechanics, airway responsiveness, lung histology, and edema in rats. We studied two highly inbred rat strains, the Lewis rat, which has a normal response to MCh , and the Fisher 344 rat, which is relatively hyperresponsive to MCh (6). IL-2 increased VT, f, VE, and RL transiently and decreased lung compliance in both rat strains. IL-2 caused pathologic changes of which epithelial detachment and bronchovascular edema were the most striking. Airway responsiveness increased in both strains, although not significantly in Fisher rats. The intravenous infusion of IL-2 causes tachycardia, hypotension, hypoxemia, and respiratory distress in humans
(9, 10), which has been attributed to an increased microvascular permeability. Although nonspecific, the immediate changes in respiratory parameters after the infusion of 750,000 U/kg of IL-2 to rats are consistent with the development of edema. IL-2 transiently increased f, VT and VE. These changes occurred rapidly and peaked around 15 min in Lewis rats. The peak changes occurred slightly earlier in Fisher 344 rats. IL-2 also caused variable increases in RL. However, aside from lung compliance, lung mechanical changes returned to baseline by 45 min. IL-2 did not significantly affect blood pressure or heart rate. The reversibility of physiologic changes, at the dose employed here, is consistent with the findings of Matory and coworkers (11) that the intravenous injection of as much as 106 U/kg every 2 days is well tolerated by rats. IL-2 increased airway responsiveness to MCh in Lewisrats but not significantly in Fisher rats. This difference may be due to genetic differences in susceptibility of these two highly inbred strains to IL-2. However, there was a strong trend for responsiveness to be increased by IL-2 in the Fisher rats, which may have been obscured by the greater variability in airway responsiveness of vehicle-treated Fisher animals. Despite these apparent differences between rat strains, IL-2 caused similar pathologic changes in the lungs of both strains. There was significant bronchial epithelial detachment. This effect was not caused by the fixation procedure since only one control animal had epithelial detachment in more than 25lifo of the airways examined. In contrast, epithelial detachment was found in more than 25l1fo of the airways examined in seven of the 13 lungs of IL-2ctreated animals. The significant correlations between epithelial detachment scores and the EC 2 00RL for the Lewisrats suggest the possibility that epithelium damage contributed to the altered airway responsiveness . Epithelial damage is a postulated cause of airway hyperresponsiveness in asthma (12) because it may decrease the release of epithelium-derived relaxant factors (13) or may permit more MCh to reach the smooth muscle. Interestingly, epithelial desquamation has been described in asthmatic airways (14), and although it has been ascribed mostly to the toxic effects of eosinophil products (15), it is tempting to speculate that the release of IL-2 from airway lymphocytes activated by inhaled antigen (16) or during viral
RENZI, DU, SAPIENZA, WANG, AND MARTIN
384
•
.
,
Fig. 4. Effect of an intravenous infusion of 750,000 Ulkg of IL·2 on histolog ic find ings in a representative animal : Note edema of the airway wall and vein (top panel) and epithel ial detachment (bottom panel).
infections (17) might also contribute to these changes. The bronchovascular edema caused by IL-2 is potentially an important cause of the altered airway responsiveness. Indeed, bronchial responsivenesshas been reported to be increased in subjects with pulmonary edema caused by left ventricular failure (18,19). Thickening of the airway wall secondary to the bronchovascular edema would be expected to amplify the effects of smooth muscle shortening on RL (20). If peribronchial edema were to reduce the mechanical interdependence between the airways and lung parenchyma, and in consequence lessen the impedance to shortening of the smooth muscle, this could also account for changes in airway responsiveness (21). Despite the above arguments, there are several reasons to suspect that bronchovascular edema is not the cause of altered responsiveness . IL-2 increased the bronchovascular edema score and wet/dry weight ratio of the lungs of Fisher rats, and responsiveness was not significantlyaltered. Surprisingly, there was an equivalent degree of edema in both groups of Lewis rats. However, we believe that the administration of higher doses of MCh to vehicle-treated Lewis rats may have affected the histologic assessment of edema . Three vehicle-treated Lewisanimals received aerosols in which the concentration of MCh exceeded 4 mg/ml and had the highest edema scores, whereas the two Lewis rats that received the least MCh had the lowest edema scores (table 3). Our impression that MCh caused bronchovascular edema is strengthened by the finding that the wet/dry weight ratio was 4.4 ± 0.1 in five Lewisrats injected with vehicleand challenged with saline , significantly lower than that in vehicle-treated Lewis rats challenged with MCh. To our knowledge, bronchial provocation with MCh has not been reported to cause bronchovascular edema or changes in the lung wet/dry weight ratio. We do not know whether it is the MCh alone or the combination of the infusion of vehicle (which contained 8 ng/ml of Escherichia coli endotoxin) with MCh that affected our measurements of edema. Conversely, the lowest edema scores in the IL-2-treated animals werefound in the animals receiving the highest dosage of MCh . (2 mg/ml), which suggests that it was IL-2 that accounted for edema in these animals rather than MCh . Even though the bronchial epithelial detachment and bronchovascular edema
INTERLEUKIN-2 AND WNG MECHANICS AND RESPONSIVENESS
385
TABLE 3 0
EFFECT OF IL-2 ON THE HISTOLOGY SCORE' OF LEWIS RATS Edema Score Experiment IL-2 IL-2 IL-2 IL-2 IL-2 IL-2 IL-2 Vehicle Veh icle Veh icle Vehicle Vehicle
Bronchi 0 0 2 2 1 1 1 1 1 1 1 0
Artery
Vein
1 1 2 2 2 3 2 2 2 3 1 1
1 1 2 2 2 3 2 1 2 2 1 1
Alveoli 0 1 1 0 1 0 0
Epithelium Detachment 0 0 1 2 2 3 3
1 0 1 0 0
i= oct a: EC2 •• RL (mg/mf)
10.2 4.21 4.1 1.7 1.4
• 0 = no changes on the slide examined; 1 = > 0 to " 25% of the structures examined had the changes looked for; 2 '" 25% to " 50% of the structures had the changes; 3 = > 50% of the structures examined had the changes looked for.
may have played a part in the increased airway responsiveness to MCh, we cannot exclude the possibility that IL-2 may have acted more directly on the bronchial smooth muscle itself or by release of a bronchoactive mediator. The intravenous infusion of IL-2 leads to the rapid production of thromboxane A 2 , which has been implicated as the cause of acute pulmonary edema in animals (22-24). A plausible explanation is that the synthesis of thromboxane A 2 , a potent bronchoconstrictor, may have increased airway responsiveness . Thromboxane A 2 has been shown to effect the increase in airway responsiveness in ozone-induced lung injury in the dog (25). In summary, we have shown that the acute intravenous infusion of IL-2 transiently increases breathing frequency, tidal volume, minute ventilation, and lung resistance. IL-2 can increase airway responsiveness to MCh. Increased airway responsiveness prior to treatment with IL-2 does not predispose to the induction of further hyperresponsiveness. Although the mechanism of increased airway responsiveness is unknown, it may be related to epithelial detachment, bronchovascular edema, or thromboxane A2 generation. Acknowledgment The writers would like to thank Miss Zeina Hammoud for the preparation of the manuscript.
=
References I. Lane HC, Masur H, Gelmann EP, Faucci AS. Therapeutic approaches to patients with AIDS. Cancer Res 1985; 45:46745-65. 2. Rosenberg SA, Lotze MT, Muel LM, et al. Observations on the systemic administration of autologous Iymphokine activated killer cells and recombinant interleukin-2 to patients with metastatic cancer. N Engl J Med 1985; 313:1485-92. 3. Yoshizawa I, Noma T, Kawano Y, Yata J. Allergen-specificinduction of interleukin-2 responsiveness in lymphocytes from children with asthma. I. Antigen specificity and initial events of the induction. J AllergyClin Immunol1989; 84:246-55. 4. Corrigan CJ, Hartnell A, Kay AB. T lymphocyteactivation in acute severeasthma. Lancet 1988; 1:1129-32. 5. Noma T, Yoshizawa I, Maeda K, Ichikawa K, Baba M, Yata J . Allergen-specific induction of interleukin-2 responsiveness in lymphocytes from children with asthma. II. Regulation of IL-2 responsiveness by supernatants of norrnal Iymphocytes, J Allergy Clin Immunol 1989; 84:255-62. 6. Martin JG, BellofioreS, Guttmann RD. Strainrelated differencesin airway reactivityamong highly inbred rats (abstract). Am Rev Respir Dis 1987; 135:A473 7. Bellofiore S, Martin JG. Antigen challenge of sensitized rats increases airway responsiveness to methacholine. J Appl Physiol 1988; 65:1642-6. 8. Mead J , Whittenberger J. Physical properties of human lung measured during spontaneous respiration . J Appl Physiol 1953; 5:770-96. 9. Glauser FL, DeBlois G, Bechard D, FowlerAA, Merchant R, Fairman RP. Cardiopulmonary toxicity of adoptive immunotherapy. Am J Med Sci 1988; 296:406-12. 10. Mier JW, Aronson FR, Numerof RP, Vachino G, Atkins MB. Toxicity of immunotherapy with interleukin-2 and Iymphokine-activated killer cells. Pathol Immunopathol Res 1988; 7:459-76. II. Matory YL, Chang AE, Lipford EH III, et al. Toxicity of recombinant human interleukin-2 in rats folowing intravenous infusion. J BioI Response Mod 1985; 4:377-90. 12. WannerA, Abraham WM, Douglas JS, Drazen JM, Richerson HB, Sri Ram J . NHLBI workshop summary. Models of airway hyperresponsiveness.
0
0
o
J:
e
w
1.04 1.0 0.79 0.5 0.29 0.29 0.05
6
t-
;;: >a: c
-----8
-It0 0 0
0
-80
0
-f0
j:: w
;;: LEW IL
LEW C
F344 IL F344 C
Fig. 5. Effect of an intravenous infusion of 750,000 Ulkg of IL·2 on lung weVdryweight ratio: Lewis (LEW) or Fisher (F344) rats received an infusion of IL-2 (IL) or vehicle (C). LEW IL, n = 7; LEW C, n = 5; F344 IL, n = 6; F344 C, n = 6. Closed squares represent overlapping values. The weVdry weight ratio wassignificantly lower in F344 C rats than in F344 IL rats. Both groups of Lewis animals were similar. Asterisk indicates p < 0.05 versus IL.
Am Rev Respir Dis 1990; 141:253-57. 13. Vanhoutte PM . Epithelium-derived relaxing factor(s) and bronchial reactivity. J Allergy Clin Immunol 1989; 83:855-61. 14. Laitinen LA, Heino M, Laitinen A, Kava T, Haahtela T. Damage of the airway epithelium and bronchial reactivity in patients with asthma. Am Rev Respir Dis 1985; 13l:599-606. 15. Barnes PJ. Effect of corticosteroids on airway hyperresponsiveness. Am Rev Respir Dis 1990; 141:570-6. 16. Cantrell DA, Smith KA. Transient expression of interleukin-2 receptors. Consequences for T cell growth. J Exp Med 1983; 158:1895-911. 17. Kasain M, Biron CA . The role of interleukin2 in T cell activation and proliferation in vivo (abstract). FASEB J 1989; 3:A478. 18. Cabanes LR, WeberSN, Matran R, et al. Bronchial hyperresponsiveness to methacholine in patients with impaired left ventricular function . N Engl J Med 1989; 320:1317-22. 19. Pison C, Malo JL, Rouleau JL, Chalaoui J, Ghezzo H, Malo J. Bronchial hyperresponsiveness to inhaled methacholine in subjects with chronic left heart failure at a time of exacerbation and after increasingdiuretic therapy. Chest 1989; 96:230-5. 20. Moreno RH, Hogg JC , Pare PD. Mechanisms of airway narrowing . Am Rev Respir Dis 1986; 133:1I71-BO. 21. Ding OJ, Martin JG, Macklem PT. Effects of lung 'volume on maximal methacholine-induced bronchoconstriction in normal humans. J Appl Physiol 1987; 62:1324-30. 22. Ferro TJ, Johnson A, Everitt J, Malik AB. IL-2 induces pulmonary edema and vasoconstriction independent of circulating lymphocytes. J Immunol 1989; 142:1916-21. 23. Klausner JM, Morel N, Paterson IS, et al. The rapid induction by interleukin-2 of pulmonary microvascular permeability. Ann Surg 1989; 209:1I9-28. 24. Klausner JM, Paterson IS, Morel NML, et al. Role of thromboxane in interleukin-2-induced lung injury in sheep. Cancer Res 1989; 49:3542-9. 25. Aizawa H, Chung KF, Leikauf GD, et al. Significance of thromboxane generation in ozoneinduced airway hyperresponsivenessin dogs. J Appl Physiol 1985; 59:1918-23.