INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Idiopathic diffuse hyperplasia of pulmonary neuroendocrine cells (IDHPNC) is a lung disease that is typically diagnosed in women in their fifth or sixth decade (1), but it has been observed in all age groups and both sexes. It is manifested clinically by a nonproductive cough and dyspnea, usually of long duration (> 10 yr). IDHPNC may be misdiagnosed as asthma or bronchitis for years. It is characterized pathologically by diffuse hyperplasia of neuroendocrine (NE) cells involving the distal bronchi and bronchioles (1). There is a spectrum of lesions ranging from diffuse hyperplasia of NE cells to numerous neuroepithelial bodies, prominent carcinoid tumorlets, and even typical carcinoid tumors (1). Often, lesions are seen obliterating the lumen of small bronchioles. In addition, abundant peribronchial connective tissue may compromise airways. The hyperplastic NE cells contain a variety of neuropeptides, including bombesinlike peptides (BLP) (1). The basis of IDHPNC is not known.

NE cells are prominent in fetal lung, where BLP act as important paracrine growth factors (2, 3). Normally, NE cells decrease in prominence after birth and only rare NE cells are seen in the lung by the first decade of life (2). In a subset of asymptomatic smokers (4) and in patients with several adult smoking-related lung diseases such as eosinophilic granuloma (5) and respiratory bronchiolitis intersitial lung disease (6), NE cells containing BLP and other neuropeptides are greatly increased in number. Increased BLP in these patients may contribute to the pathogenesis of these disorders. It is well established that BLP are autocrine growth factors for both small cell lung cancer (SCLC) (7) and non-SCLC (8). Furthermore, BLP are mitogenic for bronchial epithelial cells (9, 10) and fibroblasts (5, 11).

Neutral endopeptidase (NEP, CD10, CALLA, E.C. 3.4.24.11) is a cell-surface peptidase (12) that hydrolyses multiple small bioactive peptides such as BLP (13, 14). Because NEP hydrolyses peptides and also competes with their receptors, NEP can regulate peptide availability at the cell surface. Recently, targeted disruption of the NEP locus in mice resulted in enhanced lethality to endotoxin shock with pronounced gene-dosage effect (15) demonstrating a very important role for NEP in the modulation of septic shock. Also, NEP null mice had a 20% lower blood pressure than did control mice, with widespread basal plasma extravasation in postcapillary venular endothelia, which was reversed by recombinant NEP (16). Interestingly, NEP is inactivated in many SCLC (13, 17) and non-SCLC cell lines and tumors (17, 18) as well as in the BAL of patients with lung cancer (19). Inactivated NEP at the cell surface could promote increased activity of autocrine neuropeptide growth factors in lung cancer. In addition, recombinant NEP decreases cancer cell growth in vivo and in vitro (13, 20). In view of these data, we hypothesized that NEP was also inactivated in IDHPNC, perhaps because of a genetic alteration (21, 22) prolonging the actions of BLP, with resultant NE cell hyperplasia and epithelial and fibroblast proliferation. Contrary to our hypothesis NEP in IDHPNC is robustly expressed and is not mutated. The high expression of NEP in IDHPNC may perhaps occur in response to high levels of neuropeptides.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population

The study population consisted of 13 subjects: three with IDHPNC, five with idiopathic pulmonary fibrosis (IPF), five with hypersensitivity pneumonitis (HSP). In addition, four histologically normal lungs were studied as controls (three transplant donor lungs and one lung from a resection for lung cancer). Frozen tissue or paraffin blocks were not available on all study patients. All patients with IDHPNC, IPF, and HSP were enrolled in an interstitial lung disease (ILD) study supported by a Specialized Centers of Research Program from the National Institutes of Health. Informed consent was obtained from each patient enrolled in the ILD study, and the study was approved by the Institutional Human Subjects Review Committee.

Idiopathic diffuse hyperplasia of pulmonary neuroendocrine cells. Three patients with IDHPNC confirmed by open lung biopsy were studied. They were all women (mean age, 55 yr; range, 49 to 62 yr). Two patients were never smokers and one was an ex-smoker. IDHPNC was diagnosed on the basis of a compatible history and biopsy in the absence of any other identifiable cause of obstructive airway disease (1).

Idiopathic pulmonary fibrosis. Five patients with IPF were evaluated. All subjects were evaluated clinically, radiographically, and with lung function testing. The diagnosis was established according to previously described clinical and pathologic criteria (23). A confirmatory lung biopsy was performed in all subjects. Patients with connective tissue disease, drug or chemical exposure, or other possible etiologies of interstitial lung disease such as bronchiolitis obliterans organizing pneumonia, diffuse alveolar damage, lymphocytic interstitial pneumonia, or unclassifiable forms of chronic interstitial pneumonia were excluded from this study group. There were four men and one woman (mean age, 53 yr; range, 47 to 61 yr). Two patients were current smokers, two were ex-smokers, and one was a never-smoker.

Hypersensitivity pneumonitis. We studied five patients with HSP confirmed by lung biopsy. The diagnosis of HSP was made when the patient had a clinical history strongly suggesting exposure to agents that might result in a hypersensitivity reaction and chest radiographic and pulmonary physiologic findings compatible with HSP. All had histologic lesions consistent with HSP, specifically, peribronchiolar lymphocyte accumulation and poorly formed granulomas, with or without bronchiolitis obliterans. There were two men and three women (mean age, 58.4 yr; range, 41 to 69 yr). Two were ex-smokers; two were never smokers, and one was a current smoker. All subjects were evaluated clinically, radiographically, and with lung function testing.

Normal-appearing lung tissue. Lung tissue for immunohistochemistry was obtained from three lung transplant donors whose lungs were found to be unsuitable for transplantation, and from the histologically normal lung of one patient who had a pneumonectomy for lung cancer. Three were known to be smokers; the smoking status of the fourth was not known.

Histopathologic Evaluation

Lung tissue samples from the patients with interstitial lung disease were obtained by open or thoracoscopic lung biopsy and were prepared in an identical fashion. The biopsies were taken from two different lobes, generally the upper and lower lobes of the same lung, with avoidance of the tip of the lingula or middle lobe. In all instances, the surgeons attempted to sample portions of the lung that were obviously affected but not severely scarred. All tissue was collected, fixed, and stained in the same manner. The tissue was fixed immediately in 10% neutral buffered formalin and embedded in paraffin within 24 h. Subsequently, sections of tissue 4 µm thick were cut onto glass slides. Multiple sections were taken from each site and stained with Masson pentachrome and hematoxylin-eosin. Tissue on some patients was also either frozen immediately in liquid nitrogen or embedded with optimum cutting temperature (OCT) and frozen immediately at 70° C.

Immunohistochemistry

Immunohistochemistry was performed as previously described (17). Tissue sections 4 µm thick were cut from formalin-fixed, paraffin- embedded blocks of tumor and normal lung and placed on charged, precleaned slides (Probe On Plus; Fisher Scientific, Pittsburgh, PA). A mouse IgG₁ monoclonal anti-NEP ascites, J5 (a generous gift from Dr. Margaret Shipp, Dana Farber Cancer Center, Boston, MA) at a dilution of 1:500 was placed on the sections and incubated for 9 min at 37° C (13). Also, selected specimens were incubated with an antibombesin monoclonal antibody (Boehringer-Mannheim, Indianapolis, IN) at a concentration of 50 ng/ml. To examine the specificity of the primary antibody, controls included the substitution of the primary antibody with either normal mouse serum (Biogenex, San Ramon, CA) or the addition of an irrelevant IgG₁ (control) monoclonal antibody, MOPC 31 (Sigma, St. Louis, MO) at a dilution of 1:500. J5 Slides were graded into two categories: negative, no staining; positive, staining. We noted only intracellular staining with the J5 anti-NEP monoclonal antibody and were concerned initially that NPE was being retained intracellularly. Therefore, select specimens were placed in ChemMate buffer (BioTek, Santa Barbara, CA) (20 ml of buffer in 180 ml of distilled H₂O) and microwaved for 5 min twice in an 800-watt microwave oven set to high power to expose additional epitopes. Select specimens were also stained with 3,3'-diaminobenzidine (DAB) and commercially available J5 (Sigma) for photography purposes (13).

Cell Culture

Fibroblasts from a patient with IDHPNC were grown from explant and harvested at the fourth to tenth passage in RPMI with 10% fetal calf serum cell culture medium and subjected to RT-PCR and sequencing. Cell lines were grown in 5% CO₂ at 37° C with 100% humidity.

NEP Activity and ELISA Assay

Cell and tissue lysates were evaluated for NEP enzymatic activity and protein using a coupled chromogenic assay in a 96-well ELISA plate as previously described (17).

Immunoblot

Immunoblot was performed as previously described (17). Equal amounts of protein were added to each lane. NEP staining intensity on immunoblot was graded as either strong, detectable, or undetectable by a blinded observer.

RNA Extraction and Reverse Transcription

Cells and lung tissue were homogenized as previously described (16). Total cellular RNA was isolated by the method of Chomczynski and Sacchi (24). A 1-µg aliquot of total cellular RNA was reverse-transcribed to cDNA as previously described (17).

PCR Amplification

An aliquot of the RT product was subjected to two sequential rounds of (nested) PCR as previously described (17). The PCR products generated by these reactions spanned by 238-2437 of the NEP cDNA (17). The following two primer pairs (see Table 1) were used for the first round of PCR: 1S and 2AS, and 3S and 4AS, as previously described (17). In the second round, three sets of primer pairs were used: 1S and 1AS, 3S and 3AS, and 4S and 4AS (see Table 1). Second round PCR products were visualized on ethidum-bromide-stained 1% agarose gels. PCR was also carried out for 40 cycles using primers for the "housekeeping gene" HPRT (HPRT1S and HPRT1AS) (see Table 1) as a quality control for mRNA of each cell line. Negative controls included samples without template and without primers. No PCR product was obtained with these primers when genomic DNA was used as template, demonstrating that primers were on separate exons. The size of the amplification product was checked against a 1-kb ladder (Gibco BRL, Gaithersbeurg, MD). In addition, human NEP cDNA (a generous gift from Dr. Margaret Shipp, Dana Farber, Boston, MA) was used as a positive control for the PCR reaction.

DNA Sequence Analysis

Selected normally sized NEP PCR products from patients with IDHPNC were subcloned into the PGEM-T vector (Promega, Madison, WI) (17). Template DNA was isolated from plasmid samples (Qiagen, Chatsworth, CA). DNA sequencing was performed using a Prism Ready Reaction Dye Deoxy Terminator Cycle Sequencing Kit with FS enzyme (Perkin Elmer, Branchburg, NJ) analyzed on an ABI Model 373 A DNA sequencer (Perkin Elmer).

Statistical Analysis

All values were expressed as mean ± SEM. Parametric data were analyzed using a two-tailed t test. Nonparametric data were analyzed using Wilcoxon's signed rank test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NEP Immunoreactivity Is Increased in IDHPNC as Compared with IPF and Normal Lung Tissue

NPE immunoreactivity with the J5 monoclonal antibody at 1/500 was detectable in alveolar and bronchial epithelial cells, and neuroendocrine cells from three of three (100%) patients with IDHPNC, but present in only one of five (20%) patients with IPF in the alveolar epithelial cells and none of the normal subjects (Table 2). Intracellular NEP staining was evident in an area consistent with the location of the Golgi apparatus. After antigen retrieval, NEP immunostaining was also observed at the cell surface and in the cytoplasm of alveolar and bronchial epithelial cells. NEP monoclonal antibody immunoreactivity was partially quenched by preincubation with recombinant NEP. Of note, NEP immunoreactivity was observed in pulmonary neuroendocrine cells. When consecutive slides were stained with an antibombesin antibody, staining was colocalized to neuroendocrine cells (Figure 1).

View larger version (111K): [in this window] [in a new window]   View larger version (128K): [in this window] [in a new window]   View larger version (118K): [in this window] [in a new window]   Figure 1.   Sections of an airway from open lung biopsy of a patient with IDHPNC. The airway demonstrates hyperplasia of pulmonary neuroendocrine cells and increased surrounding connective tissue. Panel A: immunostained with J5 anti-NEP monoclonal antibody. Panel B: immunostained with bombesin monoclonal antibody. Panel C : immunostained with omission of primary antibody (negative control).

NEP Activity Is Greater in Patients with IDHPNC as Compared with IPF and HSP Controls

We examined three patients with IDHPNC and eight patients with IPF or HSP (Tables 3 and 4). Of the 8 patients with IPF or HSP, two were current smokers, four were ex-smokers, and two were never smokers. NEP activity was significantly greater in control patients with IDHPNC than in those with IPF or HSP (Table 4; Figure 2). In IDHPNC, NEP activity was 15.48 + 5.06 ng NEP/mg protein, whereas in IPF or HSP controls, it was 3.03 + 0.95 ng NEP/mg protein (p < 0.04, Wilcoxon's signed rank test). NEP activity did not correlate with the length of storage at 70° C (16).

View larger version (5K): [in this window] [in a new window]   Figure 2.   Scatterplot of NEP activity results in patients with IDHPNC versus those with IPF and HSP. (IDHPNC = idiopathic diffuse hyperplasia of pulmonary neuroendocrine cells, IPF = idiopathic pulmonary fibrosis, HSP = hypersensitivity pneumonitis.)

NEP by ELISA Is Significantly Greater in IDHPNC than in IPF or HSP

NEP, as assessed by an ELISA assay, was significantly greater in homogenized lungs from patients with IDHPNC, 12.86 + 3.43 ng NEP/mg protein, than in control patients with IPF and HSP, 1.87 + 0.46 (p < 0.01 Mann-Whitney U Statistic) (Table 4). There was no relationship between NEP levels and smoking status.

NEP Is Present in IDHPNC by Immunoblot

We examined NEP protein expression by immunoblot analysis in two patients with IDHPNC (Table 4). NEP was more strongly expressed in patients with IDHPNC than in those with IPF or HSP. NEP as observed on Western blot was the expected molecular weight of 95 kD (Figure 3). Immunoreactivity of the polyclonal antibody was quenched by incubation of the antibody with recombinant NEP.

View larger version (46K): [in this window] [in a new window]   Figure 3.   Western blot of patients with IDHPNC versus those with IPF and HSP. Equal amounts of protein were loaded in each lane. NEP molecular weight was the expected size (95 kD). For definitions of abbreviations, see Figure 2. The blot photograph was physically rearranged to order patients by diseases.

NEP mRNA Is Expressed in Patients with IDHPNC and Is the Expected Size and Free of Mutation

Nested RT-PCR was carried out on lung tissue total cellular RNA on three patients with IDHPNC and compared with two patients with IPF and with two normal subjects. Fibroblast RNA from a patient with IDHPNC was also studied. RNA quality was verified by RT-PCR (unnested) amplification of the housekeeping gene HPRT. The expected size NEP PCR products were seen in three of three patients with IDHPNC, two of two IPF patients, and two of two normal lung. DNA sequence analysis of these PCR products from two patients with IDHPNC was consistent with the sequence of NEP. NEP mRNA was expressed in the fibroblasts from a patient with IDHPNC. There was no evidence of mutation of alternative splicing in the region of the NEP cDNA from bp 238-2437, except that one clone of five for one patient tested had a 13-bp deletion from bp 426-439.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients with IDHPNC have NE cell hyperplasia and overproduce BLP and other neuropeptides. We hypothesized that low or absent NEP expression would be observed in IDHPNC, as is frequently found in lung tumors with NE differentiation (12, 17). In this study, we observed that NEP was active and highly expressed as assessed by multiple assays in patients with IDHPNC, as compared with patients with other interstitial lung diseases and normal subjects. Therefore, IDHPNC is unlikely to be the result of a defect in NEP expression. The etiology of IDHPNC is unknown. Neuroendocrine hyperplasia of the lung has been observed in 76% of patients with carcinoid tumors (25). p53 null/null and Rb null/wt knockout mice (26) also demonstrate bronchial epithelial neuroendocrine cell hyperplasia, and abnormalities of these genes may play a role in IDHPNC.

NEP was highly expressed in IDHPNC as compared with the ILD control groups. It is possible that increased NEP in IDHPNC is due to induction feedback by BLP, other neuropeptides, and/or cytokines induced by neuropeptides. In a previous study, cytokines such as tumor necrosis factor and interleukin-1 upregulated NEP expression in fibroblasts (27). Cigarette smoking and/or steroid usage could have potentially created differences in the NEP levels between the study and the control groups. Cigarette smoke inactivated NEP activity in the airway of guinea pigs (28), but we did not identify a relationship between smoking and NEP levels in these patients. In other studies, steroids have been demonstrated to upregulate NPE expression (29, 30). Because the IDHPNC patients had not been prescribed steroids, steroids do not account for the high NEP levels measured.

It has generally been accepted that NPE is reduced in lung inflammation. This contention has been derived from the results of several animal studies of lung inflammation (31) and studies of neutrophils (34). Our data demonstrate that NEP is abundantly expressed in IDHPNC, a disorder characterized by neuroendocrine cell hyperplasia and airway obstruction. Furthermore, we speculate that upregulation of NEP may be a useful adaptive response, acting perhaps to limit the detrimental effects of neuropeptides in the human lung. High levels of cell surface NEP could have a protective or counterbalancing effect in response to the high levels of BLP and other neuropeptides synthesized by hyperplastic NE cells. We speculate that this may contribute to the long history and low death rate observed in IDHPNC as compared with other fibrotic lung diseases such as IPF.