Expression of the Potent Inflammatory Cytokines, GM-CSF, IL6, and IL8, in Bronchial Epithelial Cells of Asthmatic Patients

Nat! Acad Sci 1985; 82:4633-37 8 Holtzman MJ. Species speci6city of lipoxygenase and cyclooxygenase activities expressed in pulmonary airway epithelial cells. Adv Prostaglandin Thromhoxane Leukotriene Res 1987; 17:17779 9 Holtzman MJ, Hansborough JR, Rosen GO, Turk J. Uptake, release and novel species-dependent oxygenation of arachidonic acid in human and animal airway epithelial cells. Biochim Biophys Acta 1988; 963:401-13 10 Davies RJ, Toumbis MG, Sapsford RJ, et aI. Functional and metabolic assessment of human bronchial epithelial cells in vitro. J Allergy Clin Immunoll990; 85:155 11 Rosen EM, Goldberg 10. Protein factors which regulate cell motility in vitro. Cell Develop Bioi 1989; 25:1079-87 12 Ruoslahti E, Engvall E, Hayman EH. Fibronectin: current concepts of its structure and functions. Coli Relat Res 1981; 1:95-128 13 Shoji S, Ertl RF, Linder J, Romberger OJ, Rennard SI. Bronchial epithelial ceUs produce chemotactic activity for bronchial epithelial ceUs: possible role for 6bronectin in airway repair. Am Rev Respir Dis 1990; 141:218-25 14 Aas K. Heterogeneity of bronchial asthma. Allergy 1981; 36: 3-10

Expression of the Potent Inflammatory Cytokines, GM-CSF; IL6, and 1L8, in Bronchial Epithelial Cells of Asthmatic Patients· Sabrina Mattoli, M.D., Ph.D.;t* Maurizio Marini, M.D.;* and Angelo Fasoli, M.D. t

C

ultured human bronchial epithelial cells produce cytokines with potent proinflammatory properties upon exposure to immunologic and nonimmunologic stimuli or during recovery from injury!-7 particularly GM-CSF and IL6. Because of this ability to release factors which affect the migration of granulocytes and lymphocytes and promote their local activation,2,3,6-9 airway epithelial cells can be viewed as cells that signal the inflammatory and immune system that penetration or disruption of the airway mucosa has occurred, and epithelial cell-derived cytokines may play an important role in the genesis and persistence of bronchial inflammation in some pathologic conditions, like asthma. Indeed, we have previously demonstrated that the airway secretions from patients with symptomatic asthma contain increased amounts ofGM-CSF and IL6, and these cytokines have been localized in the cytoplasm ofnonciliated bronchial epithelial cells by immunospecific labeling. 10 In addition, we have provided direct evidence of an upregulation of the expression of GM-CSF gene and protein in the bronchial epithelial cells of these patients. 6,7 The GM-CSF released from asthmatic bronchial epithelial cells could prolong the survival of eosinophils, make them hypodense, and promote *From the tSection of Lung Cellular and Molecular Biology, Department of Internal Medicine, University of Milan; and *Diagnostic Center for Respiratory Diseases, Milan, Italy. This study was supported by the Minister of University and Research, Italy. Reprint requests: Dr. Mattoli, Via C. Battisti 5, Tenli, Italy 05100

the release of mediators from those cells,6,7 thereby reproducing in vitro phenomena that are known to occur in asthma and that contribute to the pathogenesis of airway hyperresponsiveness. 11 In the present study, we have further evaluated the expression of cytokines with inflammatory properties in bronchial epithelial cells of asthmatic patients and tested the ability of corticosteroids and nedocromil sodium to downregulate their production. Six asthmatic patients (4 women, mean age ± SD: 31.2±9.8 years) and 6 healthy volunteers (3 women, mean age ± SD: 43.0 ± 15.5 years) completed the study. All subjects were nonsmokers and two in each group showed positive responses to skin prick testing with 12 common allergen extracts 12 (Table 1). The asthmatics were receiving treatment with inhaled bronchodilators alone, on demand, but their symptoms were not controlled, as they were elicited by nonspecific irritating stimuli, or required high doses of bronchodilators daily, or disturbed sleep. No patient had received corticosteroids in the last month. They had pulmonary function measurements and bronchoprovocation tests with methacholine 13 to define their nonspecific bronchial hyperresponsiveness. Their mean baseline FEV. (± SD) was 83.3 ± 14.5% of predicted value) and their geometric mean PCmM was 0.187 mglml (range from 0.04 to 0.844 mwml) (Table 1), the latter indicating that airway reactivity was moderately to severely increased in those patients. Control subjects had no past history of asthma and no evidence of bronchial hyperresponsiveness (Table 1). Asthmatics and control subjects underwent fiberoptic bronchoscopy according to a standard protocol,lO and 8 to 10 endobronchial biopsies were performed through the bron-

Table I-Subject Characteristics and Cytoldne Releaae from Bronchial Epithelial Cells Over the Firat 48 H of Incubation Immunoreactive Cytokines,* nglmlll()& cellsl48 h Subject No.

FEV., %pred

Asthmatics 1 68.0 2 86.4 3 102.3 4 93.5 5 84.8 6 65.0 Control subjects 7 96.4 8 91.6 9 98.5 10 105.3 11 101.0 12 106.3

Atop~

+/-

0.03 0.22 1.12 0.39 0.03 0.49 >32 >32 >32 >32 >32 >32

+

+

+ +

GM-CSF

11.,6

ILB

2.50 0.85 0.61 1.52 1.91 1.28

1.85 0.81 0.54 0.96 1.12 0.41

2.65 1.60 <0.35 0.87 2.11 1.86

<0.04 0.15 <0.04 0.36 <0.04 <0.04

<0.16 <0.16 <0.16 <0.16 <0.16 <0.16

<0.35 <0.35 <0.35 <0.35 <0.35 <0.35

*Triplicate samples ofsupernatants ofbronchial epithelial cells from each donor were assayed by EUSA and the results were meaned. Sensitivity of GM-CSF assay, 0.04 ng/ml; sensitivity of 11.,6 assay, 0.16 ng/ml; sensitivity of ILB assay, 0.35 ng/ml. Intraassay and interassay variability lower than 10% (coefficient of variation). CHEST I 101 I 3 I MARCH, 1992 I Supplement

27S

choscope in randomly selected lobar and segmental bronchi ofthe lungs.·· 7 Appropriate informed consent was obtained from all studied subjects. Bronchial epithelial cells were isolated from the biopsy specimens as previously described"· 7 Total RNA was extracted from the cells by the acid guanidinium thiocyanatephenol-chloroform extraction procedure, and mRNA was purified by chromatography on oligothymidylic acid cellulose. Messenger RNA from samples of 10' cells was reverse transcribed into eDNA and the expression of transcripts for an array of cytokines (GM-CSF, G-CSF, ILIa, ILI~, IL6, and ILB) was evaluated by message amplification phenotyping (MAPPing) using PCR and specific amplimer sets (Clontech Laboratories, Inc, Palo Alto, Cal). This method allows detection of cytokine transcripts in a few cells by yielding 10' to IO'-fold amplification of DNA. The results of cytokine MAPPing by PCR are reported in Figure 1. Freshly isolated bronchial epithelial cells of asthmatic patients invariably expressed appreciable levels of GM-CSF and IL6 transcripts, and the cells from 5 patients also expressed ILB transcripts. The epithelial cells of 2 control subjects expressed low levels of GM-CSF mRNA, and both the donors of those cells were atopic (Fig 1). Freshly isolated normal epithelial cells did not express the mRNA for any other cytokine tested. To evaluate the ability of bronchial epithelial cells from asthmatic patients to release GM-CSF, IL6, and ILB, and to test the inhibitory effect of corticosteroids and nedocromil, replicate samples of epithelial cells were plated at high density (10" cells/well) in collagen-treated 24-well culture plates and incubated for 4 days in IMDM medium without growth factors and fetal calf serum, in presence or in absence of different concentrations of hydrocortisone or nedocromil. In hormone-free medium, the cells attach to the culture surface but do not grow. The supernatants were harvested at 48-h intervals and replaced by fresh medium. The contents of immunoreactive GM-CSF, IL6, and ILB were measured by ELISA.'·7... Fresh IMDM was used to dilute standards and as blank. During the first 48 h of incubation, bronchial epithelial cells from asthmatic donors released high amounts of immunoreactive GM-CSF, IL6, and ILB (Table 1). Only the supernatants of epithelial cells isolated from the 2 atopic control subjects contained measurable amounts ofGM-CSFlike protein. Apparently, normal epithelial cells did not produce any IL6 or ILB (Table 1). The cytokine release from asthmatic bronchial epithelial cells spontaneously decreased during the last 2 days of incubation: the contents of immunoreactive GM-CSF and IL6 in the supernatants declined from (mean ± SD) 1.45 ± 0.69 to 0.57 ± 0.49 nglmVIO" cells (n = 6, p
r--A--, r--N----, S 1 2 3 4 5 6 7 8 9101112 C 1353 1078 872 603

310 234

f;3 ACTIN

.

, :·· ..... ----- ........t

-

FIGURE 1. MAPPing of bronchial epithelial cells from asthmatic patients (lanes 1-6) and from healthy donors (lanes 7-12) using the amplimer sets for GM-CSF, IL6, 11.8 and J3-actin (the latter as <:ontrol for samplinW. Agarose gel electrophoresis and ethidium bromide stain. The amplified DNA bands were visualized with an UV transilluminator. The PCR products (cytokine MAPPing amplimers) appear as single hands of the appropriate size. S indicates lXl74/HaeIII digest as size marker; C, preamplified DNA as PCR <:ontroI.

at concentrations up to 10- 4 M, did not reduce the production of this cytokine (mean content of immunoreactive GMCSF ± SD in the supernatants of cells treated with 10- 4 M nedocromil 1.27 ± 0.65 nglmVIO" cells/48 h vs 1.45 ± 0.69 nglmVIO" cells/48 h in supernatants of untreated cells, n = 6, p>O.05). Our results provide direct evidence that the synthesis and release of some cytokines with inflammatory properties are upregulated in bronchial epithelium of patients with symptomatic asthma and that corticosteroids, which are the most effective drugs in inducing remission ofasthmatic symptoms, also abolish this upregulation. The GM-CSF can induce the expression of adhesion molecules on endothelial and epithelial cells and primes granulocytes to release increased amounts of mediators upon stimulation, thereby promoting the infiltration of airway mucosa with inflammatory cells and their local activation."" The IL6 induces T-cell activation and proliferation,'·2.9 and ILB is chemotactic for neutrophils and T-cells. H Since bronchial epithelium represents the first cell barrier encountered by agents known to induce asthma and airway inflammation," we believe that the local effect of these epithelial cell-derived factors can explain many of the pathologic alterations observed in the bronchial mucosa of asthmatic patients, including eosinophil and T-cell activation,'O and can make the inflammatory response in asthma different from that of any other lung disease. 34th Annual Thomas L Petty Aspen Lung Conference

ACKNOWLEDGMENT: The authors acknowledge the expert assistance of Dr. J. Hollemborg for PCR and Mrs. E. Vittori for bronchoprovocation tests and skin prick tests.

REFERENCES

2

3

4 5

6

Mattoli S, Miante S, Calabro F, Mezzetti M, Allegra L. Human bronchial epithelial cells exposed to isocyanates potentiate the activation and proliferation ofT-cells induced by antigen receptor triggering through the release of ILl and IL6. In: Johansson SGD, ed. Cellular communication in allergic asthma. Uppsala: Pharmacia Allergy Foundation, 1990: 25-35 Mattoli S, Miante S, Calabro FE, Mezzetti M, Fasoli A, Allegra L. Bronchial epithelial cells exposed to isocyanates potentiate activation and proliferation of T cells. Am J Physiol (Lung Cell Mol Physiol) 1990; 259:L32o-27 Mezzetti M, Soloperto M, Fasoli A, Mattoli S. Human bronchial epithelial cells modulate CD3 and mitogen induced DNA synthesis in T-cells but function poorly as antigen presenting cells as compared to pulmonary macrophages. J Allergy Clin Immunoll991; 87:930-38 Mattoli S, Mezzetti M, Mattoso VL, Fasoli A, Allegra L. ILl and LPS upregulate the production of IL6 and GM-CSF in human bronchial epithelial cells. Eur Respir J 1990; 3(suppl):255 Marini M, Soloperto M, Mezzetti M, Fasoli A, Mattoli S. Interleukin 1 binds to specific receptors on human bronchial epithelial cells and upregulates granulocyte-macrophage colonystimulating factor synthesis and release. Am J Respir CeO Mol Bioi 1991; 4:519-24 Mattoli S, Marini M, De Franchis R, Fasoli A. GM-CSF synthesis and release are upregulated in asthmatic bronchial epithelium and contribute to eosinophil activation. Am Rev Respir Dis 1991; 143(suppl):I44

7 Soloperto M, Mattoso VL, Fasoli A, Mattoli S. A bronchial epithelial cell-derived factor in asthma which promotes eosinophil activation and survival as GM-CSF. Am J Physiol (Lung Cell Mol Physiol) 1991; 260:L530-38 8 Lopez AF, Williamson OJ, Gamble JR, Begley CG, Harlan JM, Klebanoff SJ, et al. Recombinant human granulocyte-macrophage colony-stimulating factor stimulates in vitro mature human neutrophil and eosinophil function, surface receptor expression, and survival. J Clin Invest 1986; 78:1220-28 9 Holsti MH, Raulet DH. IL-6 and IL-l synergize to stimulate IL-2 production and proliferation of peripheral T cells. J Immunoll989; 143:2514-19 10 Mattoli S, Mattoso VL, Soloperto M, Allegra L, Fasoli A. Cellular and biochemical characteristics of bronchoalveolar lavage fluid in symptomatic nonallergic asthma. J Allergy Clin Immunoll991; 87:794-802 11 Barnes PJ. New concepts in the pathogenesis of bronchial hyperresponsiveness and asthma. J Allergy Clin Immunoll989; 83:1013-26 12 Corbo GM, Foresi A, Morandini S, Valente S, Mattoli S, Ciappi G. Probit analysis applied to the allergen dose-response curve: a method for epidemiologic surve~ J Allergy Clin Immunol 1988; 81:41-47 13 Mattoli S, Rosati G, Mormile F, Ciappi G. The immediate and short-term effects of corticosteroids on cholinergic hyperreaetivity and pulmonary function in subjects with well-controlled asthma. J Allergy Clin Immunoll985; 76:214-22 14 Larsen CG, Anderson AD, Appella E, Oppenheim JJ, Matsushima K. The neutrophil-activating protein (NAP-I) is also chemotactic for T-Iymphocytes. Science 1989; 243: 1464-67 15 Allegra L, Fabbri LM, Picotti G, Mattoli S. Bronchial epithelium and asthma: a review Eur Respir J 1989; 6:S460-68

CHEST I 101 I 3 I MARCH, 1992 I Supplement

29S