Elevated Histamine and Tryptase Levels in Smokers’ Bronchoalveolar Lavage Fluid

Elevated Histamine and Tryptase Levels in Smokers' Bronchoalveolar Lavage Fluid* Do Lung Mast cells Contribute to Smokers' Emphysema? Rosemary Kalenderian, M.D.; Linga Raju, M.D., F.C.C.P.; William Roth, Ph.D.; Lawrence B. Schwartz, M.D.; Barry Gruber, M.D.; and Aaron Janofj; Ph.D.

Human lung mast cells have been reported recently to contain small amounts of the elastolytic protease present in the neutrophil and implicated in the pathogenesis of alveolar wall destruction in emphysema. Since mast cells are numerous within alveolar walls, release of inHammatory mediators (and possibly elastase) by cigarette smoking could contribute to alveolar injury in this disease. We therefore examined bronchoalveolar lavage (BAL) Huid for the mast

cell granule constituents histamine and tryptase. The results, while not conclusive, supported the possibility that cigarette smoking increases secretion of histamine releasing activity by alveolar macrophages with subsequent degranulation of local mast cells. Mast cell discharge of inHammatory mediators (including neutrophil chemotactic factors and perhaps the elastolytic protease) could then participate in the destruction of alveolar walls.

he location and number of mast cells in the human T lung has recently been re-evaluated by several

noted extracellular enzyme in association with elastin fibers despite the paucity of neutrophils in the emphysematous areas, leading to the speculation that the source of the elastase was migrating neutrophils. Degranulation of the local mast cell population of the alveolar interstitium might provide yet another source for the elastolytic enzyme observed on adjacent elastin fibers, although the reports of elastase content of the mast cell granule require confirmation. Perhaps more important is the release of neutrophil chemotactic activities and LTC49 from local mast cells which could recruit more neutrophils to add to the burden of destructive enzyme. Thus, we were interested in determining if levels of the mast cell mediators histamine and tryptase were elevated in bronchoalveolar lavage fluids (BAL) of smokers, since that would be suggestive of increased mast cell numbers or mast cell degranulation in their airways. 10 In addition, since human lung macrophages derived from thoracotomy specimens from patients with significant smoking histories and lung cancer have been reported to secrete a histamine releasing activity (HRA) in vitro,l1,12 we were interested in examining whether there might be an increased production of HRA by smokers' alveolar macrophages when compared to nonsmokers.

investigators, providing evidence of a significant population lining the alveoli within microns of the alveolar lumen. Utilizing either electron microscopy or specific immunohistochemical staining after proper preservation of lung tissues, mast cells have been found occupying 2 to 4 percent of the alveolar surface. 1-3 This observation, coupled with the recent reports of localization of neutrophil elastase in the granules of human lung mast cells' and its release via an IgE-mediated stimulus," led us to postulate that the mast cell might participate in smoking-induced lung disease. Currently the most widely accepted hypothesis for the destructive changes seen in emphysema involves an imbalance between the elastolytic proteases derived from alveolar macrophages and neutrophils and the inhibitors present in alveolar secretions," Compelling evidence for the role of neutrophil elastase includes production of emphysema in animal models with intrapulmonary instillation of enzyme purified from human neutrophils,? and localization of neutrophil elastase in the alveolar interstitium of emphysematous human lung tissue." In the latter study the authors *From the Departments of Pathology and Medicine, State University of New York, Stony Brook; the Pulmonary Disease Section, Nassau County Medical Center, East Meadow NY; and the Department of Medicine, Medical College of Virginia, Virginia Commonwealth University Richmond. This study was supported by USPHS grant HL-14262, NIH grant AI-20487, and the Veterans Administration. Presented in part at the Annual Meeting, American Thoracic Society New Orleans, May 11, 1987. Manuscript received October 5; revision accepted December 21. Reprint requests: Dr. Kale nderian, SUNY at Stony Brook, Stony

Brook, New York 11794

METHODS

Subjects

Healthy smoking and nonsmoking volunteers between the ages of 18 and 45 were recruited from university center personnel. None had a history of asthma or other pulmonary disease or was taking medication. Each subject gave informed written consent for the study which had been approved by the Human Subjects Research CHEST / 94 / 1 / JUL~

1988

119

Committee of SUNY-Stony Brook and of Nassau County Medical Center. The smokers participating in the studies evaluating histamine and tryptase content (seven men and five women) had a mean age ± SE of 29 ± 2 years and mean ± SE smoking history of 19 ± 4 pack-years. The nonsmokers (six men and four women) had a mean age of 25 ± 1 years. The smokers participating in the second part of the study (eight men and six women) had a mean of 16±2 packyear smoking history and a mean age of 29 ± 1 years, while that of the nonsmokers (six men and five women) was 26 ± 1 years of age. Smokers abstained from smoking for eight hours prior to bronchoscopy No attempt was made to correlate results between the studies on mediators in BAL supernatants and those on secretion of HRA by macrophages.

BAL After premedication with atropine sulfate, 0.6 mg intramuscularly BAL was performed with 2 percent lidocaine used topically for local anesthesia. The bronchoscope was wedged in an upper lobe subsegment. All BAL fluids analyzed for histamine and tryptase were obtained using five, 20 ml aliquots of 0.9 percent saline solution, discarding the return of the first aliquot. This was done to minimize the possible contribution of histamine of bacterial origin, as the first aliquot is said to represent more bronchial rather than alveolar' airway secretions. 13-16 The fluid recovered was filtered through sterile gauze and centrifuged at 200 g for ten minutes to obtain cell-free supernatants which were stored at -700C. In the later studies to obtain alveolar macrophages for culture, 200 ml of 0.9 percent saline solution in 50 ml aliquots was used, including the first aliquot, to increase cell yield.

Macrophage Culture The cells obtained from BAL were washed twice and then resuspended in either RPMI 1640 supplemented with 5 mM glutamine and 25 mM N-2-hydroxyethylpiperazine-N' -2-ethanesul£onic actd (HEPES) or a buffer consisting of 110 mM NaCl, 5 mM KC1, 25 mM piperazine-N, N' -bis (2-ethane-sulfonic acid) (PIPES), 30 mglL human serum albumin, and 1 y/L glucose at pH 7.4. In the initial third of these macrophage culture experiments, cells were cultured in microtiter plates at an average concentration of 3.3 X 1()6/ml,l1 or 1.5 X 1()6/cm2• Later experiments utilized 35 or 60 mm culture dishes at concentrations of 3.0 X 1()6/ml or 0.3 X 1()6/cml , on average. All cultures included 100 U/ml penicillin and 100 ....g/ml streptomycin in the media. Those BAL cells examined by Hemaeolor staining of cytospin preparations were >85 percent macrophages as has been previously reported.A" and will be referred to as alveolar macrophages (AM) in the text. Cell viability of some of these cultures was examined by trypan blue dye exclusion, and was 89 ± 17 percent viable preculture, and 74 ± 14 percent after 24 hours in culture (mean ± SD~ In the initial experiments, 150 ....1 aliquots of the AM suspension were cultured in microtiter plates; 24 hours later, 100 ....1 dextran sedimented peripheral blood leukocytes were directly added to the microtiter wells for the basophil release assay In later experiments, the AM were cultured in sterile 35 or 60 mm plates; after 24 hours, the supernatants were centrifuged at 400xg and 150 ....1 aliquots of the cell-free supernatants used for the basophil release assay

Basophil Release Assay Either 150 ....1 of AM and their conditioned media or 150 ....1 of the cell-free conditioned media was added to 100 ....1of the basophil preparation. The latter was prepared by dextran sedimentation of venous blood by a modification" of the procedure of Lichtenstein and Osler. Ie After incubation at 3rc for 45 minutes in the presence of 5 mM CaC12 and 1 mM MgC12, the basophil release reaction was stopped by centrifugation (4OOxg, ten minutes, at 4°C) and removal of the supernatant. These supernatants were assayed for

120

histamine content. Total cellular histamine of the 100 ....1 of the basophil preparation was determined on aliquots treated with 2 percent 'Iriton-X 100. The histamine content of macrophage supernatants alone was subtracted, as was spontaneous release from basophils incubated with buffer alone; both values were <5 percent of the total cellular histamine. For calculation of histamine release as a percentage of total cellular histamine, spontaneous release was subtracted from the numerator and denominator. Macrophage supernatants from each individual were tested in triplicate, and each of these triplicate samples was assayed in duplicate for histamine content. The basophils throughout these experiments were obtained from a single healthy nonsmoking donor. To normalize for the varying AM concentrations utilized for culture, the percentage of total cellular histamine released by a macrophage supernatant was expressed per 5 x 1()6 macrophages per milliliter. When more than one culture was obtained from one subjects AM, the percentage of histamine released by the AM conditioned media was reported as the average of the values obtained for individual culture supernatants tested.

Histamine Determination of BAL Supernatants Cell-free BAL supernatants were boiled for ten minutes, lyophilized, and reconstituted at 35 X their original concentration. Histamine content was assayed using the radioenzyme assay of Beaven et al.- Preparation of a typical standard curve utilizing histamine dissolved in normal saline solution that had been concentrated 35 X by lyophilization produced no deviation from the expected values. Histamine content was assayed in duplicate on two or three separate occasions. The values for BAL supernatant histamine content were normalized to the albumin content of the samples. The albumin content was determined by radial immunodiffusion on LC-Partigen plates with Behring antisera as standard.

'Iryptase Content of BAL Supernatants Cell-free BAL supernatants were concentrated 40 X by ultrafiltration and then assayed for tryptase utilizing a sandwich enzymelinked immunosorbent assay as previously described. 21 These values were also normalized to the albumin content of the BAL sample.

Statistical Analysis Differences between means were analyzed by either the approximate Students t-test (one-tailed) or the Mann-Whitney U-test (onetailed). Correlation between histamine and tryptase content of BAL was evaluated utilizing the product-moment correlation coefficient. 22 RESULTS

Mediator Content of BAL Histamine levels in BAL are typically below the level of detection of the standard double isotope radioenzyme assay as measured by other investigators and by us. Concentration of the samples 35 x by lyophilization permitted accurate detection of as little as 60 pg/ml (x 35 = 2 ng/ml, the lower limit of the assay). The range of values for smokers was s60 pg/rnl to 4 ng/ml with a mean ± SE of 500 ± 300 pgl ml. The histamine content of the nonsmokers' BAL was 115 ± 25 pg/ml (mean ± SE). The results normalized to albumin content are presented in Figure 1A (left). The mean BAL fluid histamine content of the 12 smokers was greater than that of the ten nonsmokers when analyzed by the approximate Students t-test (p
A

8

30

900

•

800

•

25

700 c

'e

:)

c

-e

20

.0

•

C'

E

2'

Z

•

~

~

.0

o

~ 500

" E

15

l&J

en i:

600

~

'0

"'-

•

10

w 400

en ~ ~

300

a:

~

200 5

•

.,•

-1:-!-2.4t.04

o

NONSMOKERS

I•

100

0

SMOKERS

•

I.

30 ·6 :t S.1 -·If-••

NONSMOKER

-01

_

SMOKERS

FIGURE 1A (left~ BAL Buid histamine levels in healthy smoking (n = 12) and nonsmoking (n = 10) adults. Numbers given are mean ± SEM of group, significantly different at p
tase levels of the smokers was also higher than that of the nonsmokers (Fig IB [right]), but this difference was of borderline statistical significance (p = 0.06, approximate Students t-test). This information is important because, while histamine is found in the granules of both basophils and mast cells, tryptase is a specific marker for the latter. 23 Thus, in those smokers' BAL samples with elevated tryptase levels, there is clear evidence of mast cell involvement. Despite the lack of strict statistical significance of the difference between the mean tryptase levels, there is a statistically significant correlation between the histamine and tryptase content of the BAL fluids (Fig 2, r=O.765, p
Alveolar macrophages obtained from healthy individuals via BAL spontaneously produced a histamine releasing factor in vitro. The percentage of release of total basophil histamine normalized to 5 X 1()6 macrophages per milliliter for the 14 smokers and 11 nonsmokers is presented in Figure 3. Five of the 14 smokers had macrophages which produced HRA in sufficient quantity to stimulate >5 percent histamine release, whereas only two of the 11 nonsmokers did so. Moreover, the average release for the five smokers was 39 percent, vs 12 percent for the two positive

1000

900

n=22

•

r =0.765

800

c:

-e

p< 0.01

700

:)

.Q

0

C'

600

E

"

~ ...... iLl

(/)

~ no >a:: ~

500

400 300 200

100

5

10

15

20

HISTAMINE (ng/mg albumin) 2. Correlation of histamine and tryptase content of BAL Buid of healthy smoking and nonsmoking adults (n = 22, r = 0.765, FIGURE

p
121

100

72

w

en

« w

LL1 20 0:: W

Z

ss

15

en i:

~

W

10

~

W

a, 5

-------

OL.......l~_

... NONSMOKERS

SMOKERS

FIGURE 3. Histamine releasing activity produced by smokers' (n = 14) and nonsmokers' (n = 11)alveolar macrophages. Each bar represents HRA production from macrophages derived from one subject, expressed as percentage of release of total cellular histamine from an aliquot of basophils derived from a single donor. Data have been normalized to number of macrophages cultured (see text). Number above bar is that of the number of macrophage cultures tested for HRA from a single subject when that number was greater than one. Analysis by Mann-Whitney V-test gives p=O.072.

nonsmokers. Analysis of these data by the MannWhitney V-test gives a p=O.072 (after correction for tied variates). Although not conclusive, the data are suggestive of increased production of a basophil (and thus, perhaps mast cell) secretagogue by unstimulated smokers' alveolar macrophages. DISCUSSION

These results show that the level of histamine and tryptase detectable in the BAL of nonsmokers is uniformly low whereas there is a subpopulation of smokers with increased levels of these mast cell mediators in their BAL fluids (Fig 1). This is not likely to be due to physiologic release during processing of BAL, as the recovered fluid is at 4° to 25°C and devoid of calcium. However, ex situ artifactual release, by trauma to the mast cells during BAL, cannot be ruled out. Other investigators have attempted to assess that possibility by reaspirating BAL through the bronchoscope and have documented the stability of its histamine content. 24 Thus, BAL fluid histamine may truly reflect release of this mediator into airway secretions in vivo. In this context, one should also note that the tryptase values in BAL were, on average, 20-fold greater than the histamine values (Fig 2), despite the presence of roughly equal amounts of these mediators in mast cell granules (1-15 pg histamine per mast cell vs 12 pg tryptase per mast cell)," This may reflect the 122

greater diffusibility of the smaller histamine molecule (MW = III vs tryptase MW = 144,(00), leading to lower recovery of the former in BAL. By contrast, if the mediators were only released during ex situ BAL processing, their recoveries would have been comparable. Even if the histamine release occurs during BAL processing, our data would still reflect an increased number of mast cells in the airways of some smokers, retrievable by BAL. These mast cells might be stimulated to discharge their inflammatory mediators by cigarette smoke, either directly or indirectly It is interesting to note that in an electron microscopy study contrasting the mast cell frequency of terminal and respiratory bronchioles, a similar wide range of values in excess of the nonsmokers was found in the smokers' tissues, although the differences did not achieve statistical significance. 25 We did not evaluate the pulmonary function of these subjects. In future studies, it might be of interest to attempt such a correlation to see if those with high mediator levels have either reduced expiratory How rates or a decline of pulmonary function with continued smoking over time. Our laboratory is also interested in the alveolar macrophage as a potential mediator of the changes occurring in lung mast cells in smokers. Alveolar macrophages produce a wide variety of mediators capable of modulating other inflammatory cells involved in lung disease, and are themselves, readily accessible to inhaled cigarette smoke and to the alveolar mast cell. In addition, there is evidence for increased production of potent inflammatory mediators by alveolar macrophages exposed to cigarette smoke. Specifically neutrophil chemotactic factor is elaborated spontaneously by smokers' alveolar macrophages in vitro, while nonsmokers' alveolar macrophages require exposure to cigarette smoke solution. 17 In a similar fashion, production of IL-l by murine alveolar macrophages in vitro was increased by prior exposure to cigarette smoke and that of murine peritoneal macrophages by exposure to the tobacco smoke component benzo(a)pyrene. 26 •27 This is notable since human monocyte IL-I recently has been reported as inducing histamine release from human basophils.P Recently a factor produced by lung macrophages from patients with lung cancer and heavy smoking histories was found capable of stimulating histamine release from basophils and lung mast cells. ll , 12 We evaluated the possibility that this factor may also be produced by macrophages of healthy smokers, and perhaps not by nonsmokers. OUf results document the production of a histamine-releasing factor by unstimulated macrophages from healthy individuals (both smokers and nonsmokers), and suggest there may be an increased production of HRA by macrophages of smokers, although this is not conclusively Histamineand TryptaseLevelsin Smokers'SAL Fluid (KaJendarlan et 81)

shown. This factor may be responsible for increased mast cell discharge in the alveolar interstitium, resulting in the finding of increased mast cell mediator content in BAL fluid of smokers. It should be pointed out, once again, that the differences in BAL histamine and tryptase that we observed in smokers vs nonsmokers may reflect increases in the numbers of mast cells present in the peripheral lung, rather than differences in rates of mast cell mediator secretion. In either case, the result is a greater propensity for mast cell-mediated injury in the smoker. The release of neutrophil chemotactic activities and perhaps neutrophil elastase from mast cells of smokers could then augment the alveolar effacement seen in emphysema. While this hypothesis remains unproven, our data at least suggest a role for mast cells in the disruption of alveolar integrity in the smoker, a concept that has not been previously advanced, except in the most general terms. 29,30 ACKNOWLEDGMENT: The writers thank Dr. David Alling, Special Assistant for Biometry National Institute of Allergy and Infectious Diseases, Bethesda, for assistance with the statistical analysis.

12

13 14 15 16

17 18 19

20

REFERENCES

1 Fox B, Bull TB, Guz A. Mast cells in the human alveolar wall: an electronmicroscopic stud~ J Clin Patholl981; 34:1333-42 2 Craig SS, De Blois G, Schwartz LB. Mast cells in human keloid, small intestine, and lung by an immunoperoxidase technique using a murine monoclonal antibody against tryptase. Am J Patholl986; 124:427-35 3 Warton A, Papadimitriou JM, Goldie RG, Paterson W An ultrastructural study of mast cells in the alveolar wall of normal and asthmatic lung. Aust J Exp BioI Med Sci 1986; 64:435-44 4 Damiano"" Schulman ES, Kucich U, Weinbaum G. Elastase in mast cells. Am Rev Respir Dis 1987; 135:A90 5 Meier HL, Heck L~ Schulman ES, MacGlashan Jr OW Purified human mast cells and basophils release human elastase and cathepsin G by an 19E-mediated mechanism. Int Archs Allergy Appl Immunoll985; 77:179-83 6 Janoff A. Elastases and emphysema. Am Rev Respir Dis 1985; 132:417-33 7 Janoff A, Sloan B, Weinbaum G, Damiano"" Sandhaus R, Elias J, et ale Experimental emphysema induced with purified human neutrophil elastase: tissue localization of the instilled protease. Am Rev Respir Dis 1977; 115:461-78 8 Damiano Tsang At Kucich U, Abrams WR t Rosenbloom J, Kimbel ~ et ale Immunolocalization of elastase in human emphysematous lungs. J Clio Invest 1986; 78:482-93 New 9 Schwartz LB. The mast cell. In: Kaplan A~ ed. Alle~ York: Churchill Livingstone, 1985:53-92 10 Schwartz LB, Metcalfe DO, Miller JS, Earl H, Sullivan '[ Uyptase levels as an indicator of mast-cell activation in systemic anaphylaxis and mastocytosis. N Engl J Med 1987; 316:1622-26 11 Schulman ES, Liu MC, Proud D~ MacGlashan Jr D~ Lichtenstein LM, Plaut M. Human lung macrophages induce

vv

21 22 23 24

25 26

27 28

29

30

histamine release from basophils and mast cells. Am Rev Respir Dis 1985; 131:230-35 Liu MC, Proud 0, Lichtenstein LM, MacGlashan Jr D~ Schleimer ~ Adkinson Jr NF, et ale Human lung macrophagederived histamine-releasing activity is due to IgE-dependent factors. J Immunol 1986; 136:2588-95 Reynolds HE Bronchoalveolar lavage. Am Rev Respir Dis 1987; 135:250-63 Ghafouri MA, Rasmussen JK, Sears K, Clayton M, Ertl R, Robbias RAt et ale Use of sequential bronchoalveolar lavage to enrich for bronchial V8 alveolar material. Clio Res 1985; 33:464 Bryant DH, Pui A. Histamine content of sputum from patients with asthma and chronic bronchitis. Clio Allerg 1982; 12:19-27 Sheinman BD, Devalia JL, Crook SJ, 1abaqchali St Davies RJ. Histamine synthesis by Haemophilus inftuenzae; not all histamine in bronchial secretions comes from mast cells. J Allergy Clio Immunoll986; 77:208 Hunninghake G~ Crystal RG. Cigarette smoking and lung destruction. Am Rev Respir Dis 1983; 128:833-38 Kaplan A~ Haak-Frendscho M, Fauci A, Dinarello C t Halbert E. A histamine-releasing factor from activated human mononuclear cells. J Immunoll985; 135:2027-32 Lichtenstein LM, Osler AG. Studies on the mechanisms of hypersensitivity phenomena: IX. Histamine release from human leukocytes by ragweed pollen antigens. J Exp Med 1964; 120:507-30 Beaven MA, Jacobsen A, Horakova Z. Modification of an enzymatic isotopic assay of histamine and its application to measurement of histamine in tissues, serum, and urine. Clin Chim Acta 1972; 37:91-103 Wenzel S, Irani AA, Sanders MA, Bradford TR, Schwartz LB. Immunoassay of tryptase from human mast cells. J Immunol Meth 1986; 86:139-42 Sokal RR, Rohlf FJ. Biometry 2nd ed. San Francisco: W H. Freeman and Co, 1981 Castells MC, Irani AA, Schwartz LB. Evaluation of human peripheral blood leukocytes for mast cell tryptase. J Immunol 1987 138:2814-19 Casale TB, Wood 0, Richerson HB, Thlpp S, Metzger Jt Zavale 0, et ale Elevated bronchoalveolar lavage fluid histamine levels in allergic asthmatics are associated with methacholine bronchial hyperresponsiveness. J Clio Invest 1987; 79:1197-1203 Lamb 0, Lumsden A. Intra-epithelial mast cells in human airway epithelium: evidence for smoking-induced changes in their frequency Thorax 1982; 37:334-42 Doukas MA, GariOla CG, Cohen DA, Gass C, Gallicchio VSt Messino MJ, et ale Tobacco smoke inhalation induces interleukin-1 (ILl) elaboration by alveolar macrophages. Exp Hematol 1987; 15:528 Lyte M, Bick E Modulation of interleukin-1 production by macrophages following benzo(a)pyrene exposure. Int J Immunopharmacoll986; 8:377-81 Subramanian N, Bray MA. Interleuldn 1 releases histamine from human basophils and mast cells in vitro. J Immunol 1987; 138:271-75 Becker CG, Dubin 1: Wiedemann HE Hypersensitivity to tobacco antigen. Proc Nat Acad Sci USA 1976; 73:1712-16 Lehrer SB, Barbandi F, 18ylor J~ Salvaggio JE. Tobacco smoke "sensitivity"-is there an immunologic basis? J Allergy Clin Immunol 1984; 73:240-45

CHEST I 94 I 1 I JUL~

1888

123