Inflammation, Vol. 16, No. 5, 1992
NEUROPEPTIDES IN PULMONARY EDEMA FLUID OF ADULT RESPIRATORY DISTRESS SYNDROME ROGER MICHAEL
F. E S P I R I T U , A. MATTHAY,
JEAN-FRANCOIS and EDWARD
PITTET, J. G O E T Z L
Departments of Medicine and Microbiology-Immunology, and Cardiovascular Research Institute University of California, San Francisco, California
Abstract--A role for peptidergic nerves in the adult respiratory distress syndrome (ARDS) was examined by radioimmunochemically quantifying neuropeptides in pulmonary edema (PE) fluids from seven patients with ARDS and six patients with PE from congestive heart failure (CHF). The PE fluid mean concentrations of substance P (SP) and gastrin-releasing peptide (GRP) were significantly higher in ARDS (0.59 _ 0.29 SD and 0.10 _-_-0.03 nM, respectively, P < 0.001 for both) than in CHF (0.19 + 0.08 and 0.04 _ 0.01), whereas no difference was detected between the mean levels of vasoacti ve intestinal peptide (VIP) and calcitonin gene-related peptide (CGRP) in the two forms of PE. Mean alveolar fluid concentration of SP was 8.7 nM (range 2.1-20.5 nM, N = 4) in sheep with acute lung injury from intravenous Pseudomonas aeruginosa, but was undetectable in sheep with balloon-induced high left atrial pressure simulating CHF (N = 2) or control sheep (N = 2). Pulmonary lymphatic clearance of SP, which reflected the rate of generation of SP in the lungs, attained a maximum of 25-95 pmol/h in sheep given P. aeruginosa intravenously, but was detected in only one of four control sheep at a lower level. Some pulmonary neuropeptides thus are released locally by acute lung injury and may contribute to endothelial and/or epithelial abnormalities underlying the altered capillary-alveolar permeability in ARDS.
INTRODUCTION
T h e a d u l t r e s p i r a t o r y cListress s y n d r o m e ( A R D S ) is a c o m p l e x p u l m o n a r y r e s p o n s e to d i v e r s e f o r m s o f s e v e r e l u n g o r s y s t e m i c injury. A R D S is c h a r a c t e r i z e d b y a diffuse i n c r e a s e i n e n d o t h e l i a l a n d e p i t h e l i a l p e r m e a b i l i t y , w h i c h results in m o v e m e n t o f p r o t e i n - r i c h fluid into t h e a l v e o l i a n d the c o n s e q u e n t m a j o r a b n o r m a l i t i e s in p u l m o n a r y m e c h a n i c s a n d gas e x c h a n g e (1, 2). I n f l a m m a t i o n is a f r e q u e n t c o m p o n e n t o f A R D S , as m a n i f e s t e d b y n e u t r o p h i l influx, 509 0360-3997/92/1000-0509506.50/0 9 1992 Plenum Publishing Corporation
510
Espiritu et al.
local generation of immunologic cytokines, and elevated pulmonary edema (PE) fluid concentrations of numerous inflammatory mediators (1, 2). A role for neuropeptides in ARDS was expected based on the density of pulmonary peptidergic neurons and the capacity of many of the agents causing ARDS to release neuropeptides from these nerve endings (3). Quantification of human PE fluid neuropeptides that have potent vascular, smooth muscle, and/or epithelial effects now has revealed significant elevations in the concentrations of substance P (SP) and gastrin-releasing peptide (GRP) in patients with ARDS, in contrast to those with cardiogenic PE. Studies in a sheep model, where ARDS-like acute lung injury is induced by intravenous administration of Pseudomonas aeruginosa, demonstrated elevated concentrations of SP both in the alveolar fluid and in the lymphatic effluent from pulmonary interstitial tissues.
MATERIALS AND METHODS Selection and Evaluation of Patients. Thirteen patients with severe acute respiratory failure, requiring endotracheal intubation and mechanical ventilation in the intensive care unit of the University of California Medical Center in San Francisco during 1990-1991, were included in the study and evaluated as described (1, 2). Seven patients fulfilled the criteria for ARDS, including diffuse bilateral pulmonary infiltrates by radiography, normal central venous pressure, alveolar-arterial oxygen gradient exceeding 300 m m Hg, and a ratio of simultaneous protein concentrations in PE fluid and plasma that exceeded 0.6 (Table 1). Six patients were classified as having cardiogenic PE from congestive heart failure (CHF), based on radiographic evidence of PE, elevated central venous pressure, abnormal left ventricular function by two-dimensional echocardiography, and a ratio of simultaneous protein concentrations in PE fluid and plasma that was 0.6 or less (Table 2) (I, 2). Studies of Acute Lung Injury in a Sheep Model for ARDS. Sheep were anesthetized and ventilated, and vascular and lymphatic duct cannulae were inserted in order to assess pulmonary and systemic hemodynamic characteristics, as well as pulmonary lymph flow (4, 5). Of the eight sheep investigated, four were given a total of 109 immunotype 1 Pseudomonas aeruginosa per hour
Table 1. Clinical and Physiological Characteristics of Patients with CHF
Case
Age/sex
Diagnosis"
Central venous pressure (ram Hg) b
1 2 3 4 5 6
60/F 61/F 41/M 56/M 48/M 65/M
MI MI Lymphoma AV prosthesis Cardiomyopathy MI
18 21 15 15 16 18
aAV = aortic valve; MI = myocardial infarction. bThe normal range is 2-10 m m Hg.
Protein concentration Edema (g/dl)
Plasma (g/dl)
Ratio
2.1 2.8 1.7 2.7 2.9 2.7
4.6 6.4 4.3 4.5 5.3 5.3
0.46 0.44 0.40 0.60 0.55 0.51
Neuropeptides in Adult Respiratory Distress Syndrome
511
Table 2. Clinical and Physiological Characteristics of Patients with ARDS
Case
Age/sex
Diagne sis
Central venous pressure (mm Hg)
1 2 3 4 5 6 7
63/M 65/F 24/M 36/F 33/M 59/F 39/M
Peritonitis Septicemia Aspiration Aspiration Bone marrow transplant Liver transplant Septicemia
8 4 10 8 10 10 9
Protein concentration Edema (g/dl)
Plasma (g/dl)
Ratio
2.7 4.1 3.2 5.2 6.7 5.6 4.1
3.9 5.0 4.9 5.1 5.4 6.0 4.5
0.69 0.82 0.65 1.02 1.24 0.93 0.91
intravenously for 8 h to induce an increase in pulmonary endothelial permeability analogous to that observed in the early phase of human acute lung injury (4). Four of the sheep served as controls, of which two had a left atrial balloon inflated to increase atrial pressure to 15 cm of H20 and the remaining two were evaluated without specific perturbation. Three milliliters per kilogram of an isosmolar solution of 5 g/100 ~ of ovine albumin were instilled with a fiberoptic bronchoscope into one lower lobe prior to ad:~ainistration of P. aeruginosa or elevation of left atrial pressure, in order to provide adequate alvec,lar fluid for replicate measurements of SP (5). After hemodynamic variables and pulmonary lymph flow had been stable for 2 h, intravenous P. aeruginosa was administered (N = 4) and left atrial pressure was increased with the balloon (N = 2). Lymph was collected in 1-h samples and one 2-3 ml aspirate of alveolar fluid was obtained at the end of every study. Each fluid was centrifuged at 10,000g for 15 rain at 4~ to remove cells and debris, portions were taken for measurement of protein concentration (6), and multiple aliquots of each were immediately frozen and stored at - 7 0 ~ Determination of Concottrations of Neuropeptides. Replicate aliquots of human PE fluid and sheep lymph and alveolar tluid were supplemented with the peptidase inhibitors phenylmethylsulfonyl fluoride (Sigma Chemical Co., St. Louis, Missouri) and DL-thiorphan (Peninsula Laboratories, Inc., San Carlos, Califcrnia) at respective final concentrations of 1 mM and 10/zg/ml and incubated for 15 min at room temperature prior to radioimmunoassays. Fluids of 0.1-0.5 ml were acidified with 0.1 M acetic acid to pH 3.5 and applied separately to Sep-Pak columns (Waters Associates, Milford, Massachusetts), which had been prewashed with 6 ml of methanol and then twice with 5 mi of 0.1 M acetic acid. The columns were washed twice with 5 ml of 0.1 M acetic acid and once with 5 ml of 30% methanol in 0.1 M acetic acid, prior to elution of the neuropeptides with 2 ml of methanol-0.1 M acetic acid (100:0.1 ml) (7). The eluates were dried in a Speed-Vac centrifuge (Savant Instruments, Inc.) and dissolved in 1 ml of radioimmunoassay (RIA) buffer, consisting of 0.1 M sodium phosphate-buffered 0.05 M NaC1 with 0.1% Triton X-100 (v/v) and 0.1 g/100 ml recrystallized bovine serum albumin (pH 7.4). Radioimmunoassays were conducted without knowledge of patient characteristics. Replicate 5- and 50-gl aliquots of each sample were diluted to 0.1 ml with RIA buffer and incubated with O. 1 mi of buffer containing 12,000-15,000 cpm of [L25t]peptide and 0.1 ml of a dilution of rabbit anti-peptide antibody for 2 h at room temperature and 16 h at 4~ The ~25I-labeled SP, VIP, and CGRP were obtained from Amersham Corp., Arlington Heights, Illinois, and had similar specific activities of 2000 Ci/mmol. The corresponding nonradiolabeled peptides were synthesized and purified as described (8), and used to prepare standard curves for each neuropeptide in the range of 0.001-0.3 pmol/tube. Suspensions of preformed complexes of goat anti-rabbit IgG (Pel-Freeze
512
Espiritu et ai.
Biologicals, Inc., Rogers, Arkansas) and normal rabbit IgG in antibody excess were prepared as described (7, 8), 0.2 ml was added to each tube, and the incubation continued for 2 h at room temperature. Then 0.7 ml of cold RIA buffer was added to each tube, which was mixed and centrifuged at 4000g for 30 min at 4~ The supematant fluids were discarded and the radioactivity in the pellets quantified in a gamma counter. A similar set of reagents for the GRP RIA was purchased from Peninsula Laboratories, Inc., San Carlos, Califomia. The mean quantities of SP, VIP, CGRP, and GRP that displaced 50% of the radiolabeled peptides from immune complexes in standard curves were 0.015, 0.010, 0.008, and 0.012 pmol. The SP in some specimens, which had sufficiently high concentrations to permit purification, was identified by high-performance liquid chromatography (HPLC) and specific RIA of the SP in portions of the eluate. The material eluted from a Sep-Pak column was redissolved in 100/~1 of 0.07% (v/v) trifluoroacetic acid (TFA) in water and injected onto a 4.6 • 250-ram column of 300-,~ pore, 7-#m diameter octylsilane (Alltech Associates, Inc., Deerfield, Illinois), which was developed in a Beckman Instruments system with dual 100A metered pumps at a flow rate of 0.8 ml/min for 10 min with 0.07 % TFA, 10 min with a linear gradient to 21.6 % acetonitrile in 0.063 % TFA, 30 rain with a second linear gradient to 50.4% acetonitrile in 0.055% TFA, and 20 min at the latter condition. The optical density of the elnate was monitored continuously at 215,260, and 280 nm with a 1040A diode-array detector (Hewlett-Packard, Inc., Palo Alto, California). The eluate was collected in 0.8-ml portions; several different volumes of each were transferred to separate tubes, dried in vacuo, and reconstituted in 100/~1 of RIA buffer; and the SP was quantified by RIA.
RESULTS Radioimmunoassays for a range of neuropeptides, which have been implicated as vascular and inflammatory mediators in other human diseases (9, 10), were used to quantify the neuropeptides in solid-phase extracts of PE fluids (Figure 1). The concentrations of SP and GRP in the PE fluids from patients with ARDS were significantly higher than those of the patients with CHF, whereas there was no difference for either VIP or CGRP. In order to confirm the identity of the SP in the PE fluids, extracts from three of the patients with ARDS and several samples of 0.01-0.1 pmol of synthetic SP were subjected to reverse-phase HPLC and the portions of eluate known to contain authentic SP were pooled and dried in vacuo prior to quantification of the SP immunoreactivity. When corrected for the recovery of synthetic SP, a range of 81-95% of the SP detected in extracts of PE fluid was found to coelute in HPLC with the authentic peptide. There was no specific correlation between the concentration of SP or GRP in the PE fluids and the level of any of the other neuropeptides or the ratio of protein concentrations in PE fluid and plasma. Preliminary studies of the species specificity of radioimmunoassays for human neuropeptides showed that SP, but not GRP, was accurately quantified in extracts of PE and other fluids of sheep. Therefore, SP was used to study the recruitment of peptidergic neurons by P. aeruginosa-induced acute lung injury in sheep. At 8 h after the introduction of P. aeruginosa intravenously, the mean concentration of immunoreactive SP in alveolar fluid was 8.7 nM (range 2.1-
Neuropeptidesin Adult RespiratoryDistress Syndrome 0.9-
513
z~
0.7-
i
0.6-
i~
0.5-
_g
i 0.4o "5
0.3-
l
iii 0.2-
A
~
gi GRP
SP
m VIP
CGRP
Fig. 1. Concentrations of neur, gpeptides in pulmonary edema fluid of patients with ARDS and CHF. Each bar and bracket depicts the mean + SD of the concentrations of a neuropeptide in the edema fluid obtained from seven ARDS ([]) and six CHF (t3) patients. The statistical significance of the difference between the mean concentrations of any neuropeptide in the ARDS and CHF groups was determined by a standard two-sample t test and shown by the symbol A for P < 0.001.
20.5 nM, N = 3). In contrast, no SP was detectable in alveolar fluid from control sheep with left atrial hypertension (N = 2) or no intervention (N = 2). HPLC analysis of extracl:s of PE fluid from two of the sheep with P. aeruginosainduced PE showed tha! 78% and 83% of the immunoreactive SP was attributable to authentic SP. After intravenous P. aeruginosa, lymph flow increased substantially without reduction in lymph protein concentration, whereas the increase in lymph flow following induction of left atrial hypertension was associated with a decrease in lymph protein concentration, as had been observed (4, 5). The quantity of
Espiritu et al.
514 100
9O ~.
80
PseudomonasI Control
7o c). 60
60
~
50
~0
40-
._~ 30-
-~
20-
10 II A
O~
~
0
i
100
-~
i
i
200 300 Time, min
~
i
400
9
500
Fig. 2. Time course of appearance of SP in the pulmonary lymphatic effluent from a sheep with Pseudomonas-induced pulmonary edema (---u--) and a concurrent control sheep ( - - 0 - - ) . Each point represents the product of the mean of duplicate measurements of immunoreactive SP concentration in the pool of lymph, collected between the time of that point and the preceding time, and the volume of lymph in the pool.
SP found in pulmonary lymph was expressed in units of clearance of picomoles per hour (Figures 2 and 3). In the P. aeruginosa-treated sheep, SP appeared in the effluent lymph immediately, and the amount increased progressively over the 8 h of the study (Figure 2). In contrast, very little SP was detected in the pulmonary lymph of control sheep. In four different sheep that received P. aeruginosa, SP was detected in lymph immediately after the challenge and rose to a peak level at 6 or 8 h. Of the two control sheep that were not challenged and the two in which left atrial pressure was elevated to levels typical of CHF, SP was detected in the effluent lymph only in one of the latter pair with model CHF and at far lower concentrations than in the acute lung injury model of ARDS (Figure 3). HPLC analysis of extracts of lymph from two of the sheep that received P. aeruginosa led to recovery of 73 % and 80% of the immunoreactive SP in the extract as authentic SP. DISCUSSION The elevated concentrations of SP and GRP in PE fluids from patients with ARDS of diverse causes (Table 2), compared to those from patients with CHF (Table 1), indicate a direct association of the neuropeptides with increased pul-
Neuropeptides in Adult Respiratory Distress Syndrome 100
80-
515
••
Pseudomonas
1
J~
"6 E
60-
r-]
High left atrial pressure
o2 oO
"6
=o 40o
.o_
~Q. E
20-
0
60
180
360
480
Time, rain
Fig. 3. Time course of appearance of SP in the pulmonary lymphatic effluent from four sheep with Pseudomonas-induced pulmonary edema and one control sheep. The values depicted by each bar were derived from the lymph concentration of SP and the volume of lymph, as described in the legend for Figure 2. Of the two unperturbed control sheep and the two exposed to high left atrial pressures, SP was only detected in the pulmonary lymphatic effluent of one (~3).
monary endothelial-epithelial permeability (Figure 1). This possibility is supported by evidence that the nanomolar levels of SP and GRP detected in the ARDS PE fluids are capable of evoking maximal smooth muscle, epithelial, glandular, and vascular responses (3, 9, 10). Involvement of neuropeptides in the responses to acute lung injury is further supported by the detection of similarly elevated levels of SP in the alveolar fluid of sheep with an ARDS-like capillary-alveolar leak ,;yndrome, elicited by intravenous P. aeruginosa. The time course of appearance of SP in effluent pulmonary lymph of sheep manifesting the Pseudomona,;-induced ARDS-like syndrome, which reflects the production of SP in the lttngs, paralleled the evolution of the increase in lung endothelial permeability (Figures 2 and 3). The elevated levels of SP and other bronchoconstrictive neuropeptides in the pulmonary alveolar space also may contribute to alterations in airway mechanics, which are observed in some instances of acute lung injury (1). The close temporal relationship between the appearance of vasoactive neuropeptides in PE fluids and the development of altered endothelial-epithelial permeability, however, is not sufficient evidence to establish a primary pathogenetic role for the neuropeptides. SP and some other neuropeptides are released
516
Espiritu et al.
from peptidergic nerve endings by histamine, other low molecular weight mediators, some cytokines, and polypeptide nerve growth factors (9, 10), one or more o f which m a y be the primary pathogenetic principles leading concurrently to the permeability defect and the release o f neuropeptides and other secondary mediators. Similarly, several neuropeptides are capable o f stimulating mast cells and mononuclear leukocytes to release a wide range o f mediators and cytokines (9, 10), o f which one or more may directly evoke the permeability defect. The potential importance o f involvement o f more than one neuropeptide derives from their frequent functional interactions, which include additive or synergistic activities, different but overlapping ranges o f effects, and recruitment or occasional inhibition o f proteases and other metabolic pathways central to the tissue survival o f neuropeptides. The concentrations o f neuropeptides detected in PE fluids in patients with A R D S , as well as in the alveolar fluid and lymph in experimental models for A R D S , are sufficient to elicit many o f the characteristic pathophysiological responses to acute lung injury. However, the exact pathogenetic roles of the neuropeptides in A R D S , as in reactive airways disease (3), still remain to be defined by introducing sufficiently specific inhibitors o f release and antagonists o f action at different times in the course o f disease. Acknowledgments--Supported in part by Pulmonary Vascular SCOR grant HL 19155 (M.A.M.)
from the National Institutes of Health,
REFERENCES
1. WEINER-KRONISH, J. P., M. A. GROPPER, and M. A. MATTHAY.1990. The adult respiratory
2.
3. 4. 5.
6.
7.
distress syndrome: Definition and prognosis, pathogenesis and treatment. Br. J. Anaesthesia 65:107-129. MATTHAY,M. A., and J. P. WIENER-KRONISH.1990. Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am. Rev. Respir. Dis. 142:1250-1257. BARNES,P. J., J. N. BARANIUK,and M. G. BELVISI. 1991. Neuropeptides in the respiratory tract. Am. Rev. Respir. Dis. 144:1187-1198, 1391-1399. BRIGHAM,K. L., W. C. WOOLVERTON,L. H. BLAKE,and N. C. STAUB.1974. Increased sheep lung vascular permeability caused by Pseudomonas bacteremia. J. Clin. Invest. 54:792-804. WEINER-KRoNISH,J. P., K. H. ALBERTINE,and M. A. MATTHAY.1991. Differential responses of the endothelial and epithelial barriers of the lung in sheep to Escherichia coli endotoxin. J. Clin. Invest. 88:864-875. RODKEY, F. L. 1965. Direct spectrophotometric determination of albumin in human serum. Clin. Chem. 11:478-487. WALKER,K. B., M. H. SERWONSKA,F. H. VALONE,W. S. HARKONEN,O. L. FRICK,K. H. SCRIVEN,W. D. RATNOFF,J. G. BROWNING,D. G. PAYAN,and E. J. GOETZL.1988. Distinctive patterns of release of neuroendocrine peptides after nasal challenge of allergic subjects with ryegrass antigen. J. Clin. lmmunol. 8:108-113,
Neuropeptides in Adult Respiratory Distress Syndrome
517
8. GOETZL,E. J., S. P. SREEDHARAN,and C. W. TURCK. 1988. Structurally distinctive vasoactive intestinal peptides from rat basophilic leukemia cells. J. Biol. Chem. 263:9083-9086. 9. GOETZL, E. J., S. P. SREEDHARAN, and W. S. HARKONEN. 1988. Pathogenetic roles of neuroimmunological mediators. Immunol. Allergy Clin. North Am. 8:183-200. 10. GOETZL, E. J., D. C. AI3ELMAN, and S. P. SREEDHARAN. 1990. Neuroimmunology. Adv. Immunol. 48:161-190.
NEUROPEPTIDES IN PULMONARY EDEMA FLUID OF ADULT RESPIRATORY DISTRESS SYNDROME ROGER MICHAEL
F. E S P I R I T U , A. MATTHAY,
JEAN-FRANCOIS and EDWARD
PITTET, J. G O E T Z L
Departments of Medicine and Microbiology-Immunology, and Cardiovascular Research Institute University of California, San Francisco, California
Abstract--A role for peptidergic nerves in the adult respiratory distress syndrome (ARDS) was examined by radioimmunochemically quantifying neuropeptides in pulmonary edema (PE) fluids from seven patients with ARDS and six patients with PE from congestive heart failure (CHF). The PE fluid mean concentrations of substance P (SP) and gastrin-releasing peptide (GRP) were significantly higher in ARDS (0.59 _ 0.29 SD and 0.10 _-_-0.03 nM, respectively, P < 0.001 for both) than in CHF (0.19 + 0.08 and 0.04 _ 0.01), whereas no difference was detected between the mean levels of vasoacti ve intestinal peptide (VIP) and calcitonin gene-related peptide (CGRP) in the two forms of PE. Mean alveolar fluid concentration of SP was 8.7 nM (range 2.1-20.5 nM, N = 4) in sheep with acute lung injury from intravenous Pseudomonas aeruginosa, but was undetectable in sheep with balloon-induced high left atrial pressure simulating CHF (N = 2) or control sheep (N = 2). Pulmonary lymphatic clearance of SP, which reflected the rate of generation of SP in the lungs, attained a maximum of 25-95 pmol/h in sheep given P. aeruginosa intravenously, but was detected in only one of four control sheep at a lower level. Some pulmonary neuropeptides thus are released locally by acute lung injury and may contribute to endothelial and/or epithelial abnormalities underlying the altered capillary-alveolar permeability in ARDS.
INTRODUCTION
T h e a d u l t r e s p i r a t o r y cListress s y n d r o m e ( A R D S ) is a c o m p l e x p u l m o n a r y r e s p o n s e to d i v e r s e f o r m s o f s e v e r e l u n g o r s y s t e m i c injury. A R D S is c h a r a c t e r i z e d b y a diffuse i n c r e a s e i n e n d o t h e l i a l a n d e p i t h e l i a l p e r m e a b i l i t y , w h i c h results in m o v e m e n t o f p r o t e i n - r i c h fluid into t h e a l v e o l i a n d the c o n s e q u e n t m a j o r a b n o r m a l i t i e s in p u l m o n a r y m e c h a n i c s a n d gas e x c h a n g e (1, 2). I n f l a m m a t i o n is a f r e q u e n t c o m p o n e n t o f A R D S , as m a n i f e s t e d b y n e u t r o p h i l influx, 509 0360-3997/92/1000-0509506.50/0 9 1992 Plenum Publishing Corporation
510
Espiritu et al.
local generation of immunologic cytokines, and elevated pulmonary edema (PE) fluid concentrations of numerous inflammatory mediators (1, 2). A role for neuropeptides in ARDS was expected based on the density of pulmonary peptidergic neurons and the capacity of many of the agents causing ARDS to release neuropeptides from these nerve endings (3). Quantification of human PE fluid neuropeptides that have potent vascular, smooth muscle, and/or epithelial effects now has revealed significant elevations in the concentrations of substance P (SP) and gastrin-releasing peptide (GRP) in patients with ARDS, in contrast to those with cardiogenic PE. Studies in a sheep model, where ARDS-like acute lung injury is induced by intravenous administration of Pseudomonas aeruginosa, demonstrated elevated concentrations of SP both in the alveolar fluid and in the lymphatic effluent from pulmonary interstitial tissues.
MATERIALS AND METHODS Selection and Evaluation of Patients. Thirteen patients with severe acute respiratory failure, requiring endotracheal intubation and mechanical ventilation in the intensive care unit of the University of California Medical Center in San Francisco during 1990-1991, were included in the study and evaluated as described (1, 2). Seven patients fulfilled the criteria for ARDS, including diffuse bilateral pulmonary infiltrates by radiography, normal central venous pressure, alveolar-arterial oxygen gradient exceeding 300 m m Hg, and a ratio of simultaneous protein concentrations in PE fluid and plasma that exceeded 0.6 (Table 1). Six patients were classified as having cardiogenic PE from congestive heart failure (CHF), based on radiographic evidence of PE, elevated central venous pressure, abnormal left ventricular function by two-dimensional echocardiography, and a ratio of simultaneous protein concentrations in PE fluid and plasma that was 0.6 or less (Table 2) (I, 2). Studies of Acute Lung Injury in a Sheep Model for ARDS. Sheep were anesthetized and ventilated, and vascular and lymphatic duct cannulae were inserted in order to assess pulmonary and systemic hemodynamic characteristics, as well as pulmonary lymph flow (4, 5). Of the eight sheep investigated, four were given a total of 109 immunotype 1 Pseudomonas aeruginosa per hour
Table 1. Clinical and Physiological Characteristics of Patients with CHF
Case
Age/sex
Diagnosis"
Central venous pressure (ram Hg) b
1 2 3 4 5 6
60/F 61/F 41/M 56/M 48/M 65/M
MI MI Lymphoma AV prosthesis Cardiomyopathy MI
18 21 15 15 16 18
aAV = aortic valve; MI = myocardial infarction. bThe normal range is 2-10 m m Hg.
Protein concentration Edema (g/dl)
Plasma (g/dl)
Ratio
2.1 2.8 1.7 2.7 2.9 2.7
4.6 6.4 4.3 4.5 5.3 5.3
0.46 0.44 0.40 0.60 0.55 0.51
Neuropeptides in Adult Respiratory Distress Syndrome
511
Table 2. Clinical and Physiological Characteristics of Patients with ARDS
Case
Age/sex
Diagne sis
Central venous pressure (mm Hg)
1 2 3 4 5 6 7
63/M 65/F 24/M 36/F 33/M 59/F 39/M
Peritonitis Septicemia Aspiration Aspiration Bone marrow transplant Liver transplant Septicemia
8 4 10 8 10 10 9
Protein concentration Edema (g/dl)
Plasma (g/dl)
Ratio
2.7 4.1 3.2 5.2 6.7 5.6 4.1
3.9 5.0 4.9 5.1 5.4 6.0 4.5
0.69 0.82 0.65 1.02 1.24 0.93 0.91
intravenously for 8 h to induce an increase in pulmonary endothelial permeability analogous to that observed in the early phase of human acute lung injury (4). Four of the sheep served as controls, of which two had a left atrial balloon inflated to increase atrial pressure to 15 cm of H20 and the remaining two were evaluated without specific perturbation. Three milliliters per kilogram of an isosmolar solution of 5 g/100 ~ of ovine albumin were instilled with a fiberoptic bronchoscope into one lower lobe prior to ad:~ainistration of P. aeruginosa or elevation of left atrial pressure, in order to provide adequate alvec,lar fluid for replicate measurements of SP (5). After hemodynamic variables and pulmonary lymph flow had been stable for 2 h, intravenous P. aeruginosa was administered (N = 4) and left atrial pressure was increased with the balloon (N = 2). Lymph was collected in 1-h samples and one 2-3 ml aspirate of alveolar fluid was obtained at the end of every study. Each fluid was centrifuged at 10,000g for 15 rain at 4~ to remove cells and debris, portions were taken for measurement of protein concentration (6), and multiple aliquots of each were immediately frozen and stored at - 7 0 ~ Determination of Concottrations of Neuropeptides. Replicate aliquots of human PE fluid and sheep lymph and alveolar tluid were supplemented with the peptidase inhibitors phenylmethylsulfonyl fluoride (Sigma Chemical Co., St. Louis, Missouri) and DL-thiorphan (Peninsula Laboratories, Inc., San Carlos, Califcrnia) at respective final concentrations of 1 mM and 10/zg/ml and incubated for 15 min at room temperature prior to radioimmunoassays. Fluids of 0.1-0.5 ml were acidified with 0.1 M acetic acid to pH 3.5 and applied separately to Sep-Pak columns (Waters Associates, Milford, Massachusetts), which had been prewashed with 6 ml of methanol and then twice with 5 mi of 0.1 M acetic acid. The columns were washed twice with 5 ml of 0.1 M acetic acid and once with 5 ml of 30% methanol in 0.1 M acetic acid, prior to elution of the neuropeptides with 2 ml of methanol-0.1 M acetic acid (100:0.1 ml) (7). The eluates were dried in a Speed-Vac centrifuge (Savant Instruments, Inc.) and dissolved in 1 ml of radioimmunoassay (RIA) buffer, consisting of 0.1 M sodium phosphate-buffered 0.05 M NaC1 with 0.1% Triton X-100 (v/v) and 0.1 g/100 ml recrystallized bovine serum albumin (pH 7.4). Radioimmunoassays were conducted without knowledge of patient characteristics. Replicate 5- and 50-gl aliquots of each sample were diluted to 0.1 ml with RIA buffer and incubated with O. 1 mi of buffer containing 12,000-15,000 cpm of [L25t]peptide and 0.1 ml of a dilution of rabbit anti-peptide antibody for 2 h at room temperature and 16 h at 4~ The ~25I-labeled SP, VIP, and CGRP were obtained from Amersham Corp., Arlington Heights, Illinois, and had similar specific activities of 2000 Ci/mmol. The corresponding nonradiolabeled peptides were synthesized and purified as described (8), and used to prepare standard curves for each neuropeptide in the range of 0.001-0.3 pmol/tube. Suspensions of preformed complexes of goat anti-rabbit IgG (Pel-Freeze
512
Espiritu et ai.
Biologicals, Inc., Rogers, Arkansas) and normal rabbit IgG in antibody excess were prepared as described (7, 8), 0.2 ml was added to each tube, and the incubation continued for 2 h at room temperature. Then 0.7 ml of cold RIA buffer was added to each tube, which was mixed and centrifuged at 4000g for 30 min at 4~ The supematant fluids were discarded and the radioactivity in the pellets quantified in a gamma counter. A similar set of reagents for the GRP RIA was purchased from Peninsula Laboratories, Inc., San Carlos, Califomia. The mean quantities of SP, VIP, CGRP, and GRP that displaced 50% of the radiolabeled peptides from immune complexes in standard curves were 0.015, 0.010, 0.008, and 0.012 pmol. The SP in some specimens, which had sufficiently high concentrations to permit purification, was identified by high-performance liquid chromatography (HPLC) and specific RIA of the SP in portions of the eluate. The material eluted from a Sep-Pak column was redissolved in 100/~1 of 0.07% (v/v) trifluoroacetic acid (TFA) in water and injected onto a 4.6 • 250-ram column of 300-,~ pore, 7-#m diameter octylsilane (Alltech Associates, Inc., Deerfield, Illinois), which was developed in a Beckman Instruments system with dual 100A metered pumps at a flow rate of 0.8 ml/min for 10 min with 0.07 % TFA, 10 min with a linear gradient to 21.6 % acetonitrile in 0.063 % TFA, 30 rain with a second linear gradient to 50.4% acetonitrile in 0.055% TFA, and 20 min at the latter condition. The optical density of the elnate was monitored continuously at 215,260, and 280 nm with a 1040A diode-array detector (Hewlett-Packard, Inc., Palo Alto, California). The eluate was collected in 0.8-ml portions; several different volumes of each were transferred to separate tubes, dried in vacuo, and reconstituted in 100/~1 of RIA buffer; and the SP was quantified by RIA.
RESULTS Radioimmunoassays for a range of neuropeptides, which have been implicated as vascular and inflammatory mediators in other human diseases (9, 10), were used to quantify the neuropeptides in solid-phase extracts of PE fluids (Figure 1). The concentrations of SP and GRP in the PE fluids from patients with ARDS were significantly higher than those of the patients with CHF, whereas there was no difference for either VIP or CGRP. In order to confirm the identity of the SP in the PE fluids, extracts from three of the patients with ARDS and several samples of 0.01-0.1 pmol of synthetic SP were subjected to reverse-phase HPLC and the portions of eluate known to contain authentic SP were pooled and dried in vacuo prior to quantification of the SP immunoreactivity. When corrected for the recovery of synthetic SP, a range of 81-95% of the SP detected in extracts of PE fluid was found to coelute in HPLC with the authentic peptide. There was no specific correlation between the concentration of SP or GRP in the PE fluids and the level of any of the other neuropeptides or the ratio of protein concentrations in PE fluid and plasma. Preliminary studies of the species specificity of radioimmunoassays for human neuropeptides showed that SP, but not GRP, was accurately quantified in extracts of PE and other fluids of sheep. Therefore, SP was used to study the recruitment of peptidergic neurons by P. aeruginosa-induced acute lung injury in sheep. At 8 h after the introduction of P. aeruginosa intravenously, the mean concentration of immunoreactive SP in alveolar fluid was 8.7 nM (range 2.1-
Neuropeptidesin Adult RespiratoryDistress Syndrome 0.9-
513
z~
0.7-
i
0.6-
i~
0.5-
_g
i 0.4o "5
0.3-
l
iii 0.2-
A
~
gi GRP
SP
m VIP
CGRP
Fig. 1. Concentrations of neur, gpeptides in pulmonary edema fluid of patients with ARDS and CHF. Each bar and bracket depicts the mean + SD of the concentrations of a neuropeptide in the edema fluid obtained from seven ARDS ([]) and six CHF (t3) patients. The statistical significance of the difference between the mean concentrations of any neuropeptide in the ARDS and CHF groups was determined by a standard two-sample t test and shown by the symbol A for P < 0.001.
20.5 nM, N = 3). In contrast, no SP was detectable in alveolar fluid from control sheep with left atrial hypertension (N = 2) or no intervention (N = 2). HPLC analysis of extracl:s of PE fluid from two of the sheep with P. aeruginosainduced PE showed tha! 78% and 83% of the immunoreactive SP was attributable to authentic SP. After intravenous P. aeruginosa, lymph flow increased substantially without reduction in lymph protein concentration, whereas the increase in lymph flow following induction of left atrial hypertension was associated with a decrease in lymph protein concentration, as had been observed (4, 5). The quantity of
Espiritu et al.
514 100
9O ~.
80
PseudomonasI Control
7o c). 60
60
~
50
~0
40-
._~ 30-
-~
20-
10 II A
O~
~
0
i
100
-~
i
i
200 300 Time, min
~
i
400
9
500
Fig. 2. Time course of appearance of SP in the pulmonary lymphatic effluent from a sheep with Pseudomonas-induced pulmonary edema (---u--) and a concurrent control sheep ( - - 0 - - ) . Each point represents the product of the mean of duplicate measurements of immunoreactive SP concentration in the pool of lymph, collected between the time of that point and the preceding time, and the volume of lymph in the pool.
SP found in pulmonary lymph was expressed in units of clearance of picomoles per hour (Figures 2 and 3). In the P. aeruginosa-treated sheep, SP appeared in the effluent lymph immediately, and the amount increased progressively over the 8 h of the study (Figure 2). In contrast, very little SP was detected in the pulmonary lymph of control sheep. In four different sheep that received P. aeruginosa, SP was detected in lymph immediately after the challenge and rose to a peak level at 6 or 8 h. Of the two control sheep that were not challenged and the two in which left atrial pressure was elevated to levels typical of CHF, SP was detected in the effluent lymph only in one of the latter pair with model CHF and at far lower concentrations than in the acute lung injury model of ARDS (Figure 3). HPLC analysis of extracts of lymph from two of the sheep that received P. aeruginosa led to recovery of 73 % and 80% of the immunoreactive SP in the extract as authentic SP. DISCUSSION The elevated concentrations of SP and GRP in PE fluids from patients with ARDS of diverse causes (Table 2), compared to those from patients with CHF (Table 1), indicate a direct association of the neuropeptides with increased pul-
Neuropeptides in Adult Respiratory Distress Syndrome 100
80-
515
••
Pseudomonas
1
J~
"6 E
60-
r-]
High left atrial pressure
o2 oO
"6
=o 40o
.o_
~Q. E
20-
0
60
180
360
480
Time, rain
Fig. 3. Time course of appearance of SP in the pulmonary lymphatic effluent from four sheep with Pseudomonas-induced pulmonary edema and one control sheep. The values depicted by each bar were derived from the lymph concentration of SP and the volume of lymph, as described in the legend for Figure 2. Of the two unperturbed control sheep and the two exposed to high left atrial pressures, SP was only detected in the pulmonary lymphatic effluent of one (~3).
monary endothelial-epithelial permeability (Figure 1). This possibility is supported by evidence that the nanomolar levels of SP and GRP detected in the ARDS PE fluids are capable of evoking maximal smooth muscle, epithelial, glandular, and vascular responses (3, 9, 10). Involvement of neuropeptides in the responses to acute lung injury is further supported by the detection of similarly elevated levels of SP in the alveolar fluid of sheep with an ARDS-like capillary-alveolar leak ,;yndrome, elicited by intravenous P. aeruginosa. The time course of appearance of SP in effluent pulmonary lymph of sheep manifesting the Pseudomona,;-induced ARDS-like syndrome, which reflects the production of SP in the lttngs, paralleled the evolution of the increase in lung endothelial permeability (Figures 2 and 3). The elevated levels of SP and other bronchoconstrictive neuropeptides in the pulmonary alveolar space also may contribute to alterations in airway mechanics, which are observed in some instances of acute lung injury (1). The close temporal relationship between the appearance of vasoactive neuropeptides in PE fluids and the development of altered endothelial-epithelial permeability, however, is not sufficient evidence to establish a primary pathogenetic role for the neuropeptides. SP and some other neuropeptides are released
516
Espiritu et al.
from peptidergic nerve endings by histamine, other low molecular weight mediators, some cytokines, and polypeptide nerve growth factors (9, 10), one or more o f which m a y be the primary pathogenetic principles leading concurrently to the permeability defect and the release o f neuropeptides and other secondary mediators. Similarly, several neuropeptides are capable o f stimulating mast cells and mononuclear leukocytes to release a wide range o f mediators and cytokines (9, 10), o f which one or more may directly evoke the permeability defect. The potential importance o f involvement o f more than one neuropeptide derives from their frequent functional interactions, which include additive or synergistic activities, different but overlapping ranges o f effects, and recruitment or occasional inhibition o f proteases and other metabolic pathways central to the tissue survival o f neuropeptides. The concentrations o f neuropeptides detected in PE fluids in patients with A R D S , as well as in the alveolar fluid and lymph in experimental models for A R D S , are sufficient to elicit many o f the characteristic pathophysiological responses to acute lung injury. However, the exact pathogenetic roles of the neuropeptides in A R D S , as in reactive airways disease (3), still remain to be defined by introducing sufficiently specific inhibitors o f release and antagonists o f action at different times in the course o f disease. Acknowledgments--Supported in part by Pulmonary Vascular SCOR grant HL 19155 (M.A.M.)
from the National Institutes of Health,
REFERENCES
1. WEINER-KRONISH, J. P., M. A. GROPPER, and M. A. MATTHAY.1990. The adult respiratory
2.
3. 4. 5.
6.
7.
distress syndrome: Definition and prognosis, pathogenesis and treatment. Br. J. Anaesthesia 65:107-129. MATTHAY,M. A., and J. P. WIENER-KRONISH.1990. Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am. Rev. Respir. Dis. 142:1250-1257. BARNES,P. J., J. N. BARANIUK,and M. G. BELVISI. 1991. Neuropeptides in the respiratory tract. Am. Rev. Respir. Dis. 144:1187-1198, 1391-1399. BRIGHAM,K. L., W. C. WOOLVERTON,L. H. BLAKE,and N. C. STAUB.1974. Increased sheep lung vascular permeability caused by Pseudomonas bacteremia. J. Clin. Invest. 54:792-804. WEINER-KRoNISH,J. P., K. H. ALBERTINE,and M. A. MATTHAY.1991. Differential responses of the endothelial and epithelial barriers of the lung in sheep to Escherichia coli endotoxin. J. Clin. Invest. 88:864-875. RODKEY, F. L. 1965. Direct spectrophotometric determination of albumin in human serum. Clin. Chem. 11:478-487. WALKER,K. B., M. H. SERWONSKA,F. H. VALONE,W. S. HARKONEN,O. L. FRICK,K. H. SCRIVEN,W. D. RATNOFF,J. G. BROWNING,D. G. PAYAN,and E. J. GOETZL.1988. Distinctive patterns of release of neuroendocrine peptides after nasal challenge of allergic subjects with ryegrass antigen. J. Clin. lmmunol. 8:108-113,
Neuropeptides in Adult Respiratory Distress Syndrome
517
8. GOETZL,E. J., S. P. SREEDHARAN,and C. W. TURCK. 1988. Structurally distinctive vasoactive intestinal peptides from rat basophilic leukemia cells. J. Biol. Chem. 263:9083-9086. 9. GOETZL, E. J., S. P. SREEDHARAN, and W. S. HARKONEN. 1988. Pathogenetic roles of neuroimmunological mediators. Immunol. Allergy Clin. North Am. 8:183-200. 10. GOETZL, E. J., D. C. AI3ELMAN, and S. P. SREEDHARAN. 1990. Neuroimmunology. Adv. Immunol. 48:161-190.
