Clara Cell Secretory Protein (CC16): Features as a Peripheral Lung Biomarker F. BROECKAERT, A. CLIPPE, B. KNOOPS, C. HERMANS, AND A. BERNARDa Industrial Toxicology and Occupational Medicine Unit, Faculty of Medicine, School of Public Health, Catholic University of Louvain, B-1200 Brussels, Belgium
ABSTRACT: Clara cell protein (CC16 or CC10) is a 15.8-kDa protein secreted all along the tracheobronchial tree and especially in the terminal bronchioles where Clara cells are localized. Even though the exact in vivo function of CC16 remains to be clarified, evidence is accumulating that CC16 plays an important protective role in the respiratory tract against oxidative stress and inflammatory response. CC16, however, presents also a major interest as a peripheral lung marker for assessing the cellular integrity or the permeability of the lung epithelium. The serum concentrations of CC16 are decreased in subjects with chronic lung damage caused by tobacco smoke and other air pollutants as a consequence of the destruction of Clara cells. By contrast, serum CC16 increases in acute or chronic lung disorders characterized by an increased airways permeability. The sensitivity of serum CC16 to an increased leakiness of the lung allows for the detection of defects of the epithelial barrier at ozone levels below current air-quality guidelines. Although the clinical significance of these early epithelial changes detected by serum CC16 remains to be determined, these results clearly show that the assay in serum of lung secretory proteins such as CC16 represents a new noninvasive approach to evaluate the integrity of the respiratory tract.
INTRODUCTION The central activity of Clara cells, one of the most multifunctional epithelial cell type of the mammalian lung, appears to be devoted to the protection of the respiratory tract against toxic inhaled agents. Indeed, Clara cells have been shown to repair damaged epithelium, to detoxify xenobiotics, and to secrete proteins with important biological activities such as surfactant-associated proteins, leukocyte-protease inhibitors, and the 15.8-kDa Clara cell protein (CC16).1,2 CC16 is the major protein secreted by Clara cells and one of the main secretory proteins of the lung. The protein occurs in very high concentrations in the epithelial lining fluid where it appears to play an antioxidant/inflammatory role, notably by modulating the production and/ or activity of phospholipase-A2, interferon-γ, and tumor necrosis factor-α (for a review, see reference 3). From a diagnostic point of view, CC16 is at the origin of a novel approach for assessing the integrity of the respiratory epithelium, based on the assay in serum of aAuthor for correspondence: A. Bernard, Industrial Toxicology and Occupational Medicine Unit, Faculty of Medicine, School of Public Health, Catholic University of Louvain, 30.54 Clos Chapelle-aux-Champs, B-1200 Brussels, Belgium.
[email protected]
68
BROECKAERT et al.: CC16
69
lung-specific proteins.2–6 Most biomarkers of lung inflammation or injury described so far are constituents of the epithelial lining fluid sampled by bronchoalveolar lavage (BAL) techniques. These biomarkers consist of plasma-derived proteins or of molecules produced locally by the epithelial and inflammatory cells such as cytokines, enzymes, or growth factors. Although of great potential for evaluating the extent of inflammation or tissue damage, markers in BAL are not applicable for monitoring populations exposed to air pollutants in the environment or workplace. They also cannot be used for assessing health risks in children and patients with compromised cardiopulmonary function who are the most vulnerable to air pollution. Recent clinical and experimental studies have shown that the determination of CC16 in serum is a sensitive biomarker to detect increased leakiness of the lung epithelial barrier and/or to evaluate the integrity of Clara cells. The aim of this paper is to summarize recent advances concerning the application of CC16 as a lung peripheral marker.
FEATURES AND METABOLISM OF HUMAN CC16 The Clara cell protein family encodes a group of approximately 20 homologous proteins including the human CC16. This protein, first identified in urine of patients with renal failure7 and purified later from lung lavage,8 is the counterpart of rabbit uteroglobin (UG), a secretory protein of the lung (both sexes) and the uterus (during pregnancy) of lagomorphs (rabbits, hares).9 Both proteins are homodimeric, consisting of two amino acid subunits oriented antiparallel and connected by two disulfide bonds. Although unreduced CC16 shows an apparent molecular weight of approximately 10 kDa on gel electrophoresis, its molecular mass, determined by electrospray ionization/spectrometry is exactly 15,840 Da.10 In humans, the organ- and tissue-specific expression of the CC16 gene has been studied by Northern blot analysis using specific probes for CC16. As shown in FIG URE 1, CC16 is mainly expressed in the lung, the trachea, and the fetal lung. Earlier studies, however, have shown that CC16 is not produced uniformly in the respiratory tract, with the protein being predominantly found in airways with an increasing density from the trachea and bronchi to the terminal bronchioles.11,12 An extrapulmonary synthesis has also been evidenced in the prostate, the endometrium, and the kidney, but the levels of expression in these organs are on average 20 times lower than in the lung. This much higher pulmonary expression of CC16 combined with the fact that the lung offers the largest exchange surface area with blood explains why serum CC16 derives almost exclusively from the respiratory tract.2 This tissue distribution of CC16 is fully consistent with the pattern of CC16 concentrations in biological fluids. The highest concentrations of CC16 are found in pulmonary fluids such as the epithelial lining fluid (ELF), the bronchoalveolar lavage fluid (BALF), and the sputum. CC16 also is present in high concentrations in human amniotic fluid, where it reaches a mean value of about 1 mg/L before delivery.13 The protein occurs also in extrapulmonary fluids such as serum and urine, but in much smaller concentrations (10–20 and 5–10 µg/L, respectively). In human BALF, the concentration of CC16 is on average between 1 and 2 mg/L, which represents 2–3% of the total soluble protein content of BALF. Since the lavage tech-
70
ANNALS NEW YORK ACADEMY OF SCIENCES
FIGURE 1. Master Blot analysis of poly(A)+ RNA from approximately 50 human organs and tissues (Human RNA Master Blot, Clontech, San Diego, CA). For hybridization, cDNA 32P-radiolabeled probes specific for CC16 and β-actin (control probe) were used on the same Northern blot. From reference 2.
BROECKAERT et al.: CC16
71
nique results in an approximately 100-fold dilution of the ELF, CC16 concentrations at the surface of airways can be estimated at about 100 mg/L. However, the concentrations of CC16 in BALF of healthy subjects show a great dispersion that cannot be explained solely by the variable dilution of BALF samples, suggesting the existence of interindividual variations in the pulmonary synthesis/secretion of CC16. In serum from healthy nonsmokers, CC16 concentrations are on average between 10 and 15 µg/L, that is, a value about 10,000 times lower than in ELF. This huge concentration gradient probably provides the driving force for the movement of CC16 from the lung into the blood.2 The integrity of the bronchoalveolar/capillary membranes is an important determinant of this equilibrium as indicated by situations in which this barrier is compromised (e.g., acute lung injury, sarcoidosis). Like other lowmolecular-weight proteins, CC16 is rapidly cleared from plasma by glomerular filtration with a half-life estimated at about 2 to 3 hours. As a corollary, serum CC16 rises as the glomerular filtration rate (GFR) declines.14 The slight elevation of CC16 in serum with aging15 is presumably due to the age-related decrease of the GFR. CC16 in serum, however, is independent of the level of lipids, body mass index, and sex and does not show nocturnal variations.16 Once filtered through the glomeruli, CC16 is almost completely reabsorbed by the proximal tubules where it is catabolized, explaining why urine contains normally only small amounts of CC16. In males, however, from puberty on, CC16 is secreted by the prostate directly into the urogenital tract, which increases the urinary concentrations of CC16 by a factor of 10 up to 100 depending on the mode of urine collection (only the first mL are highly contaminated due to the washout of the urogenital tract). A defect in the tubule reabsorptive capacity results in a markedly increased urinary excretion of CC16, which correlates with that of other low-molecular-weight proteins characteristic of a tubular proteinuria.17 In rodents, CC16 follows the same metabolic pathway as in humans, leaking from the respiratory tract into serum and increasing in urine when proximal tubular function fails.18 While the exact mechanism by which lung CC16 reaches the bloodstream remains to be clarified, various observations support a transport by convection and/or passive diffusion. Recent pharmacokinetic studies in rats have shown that, following intratracheal administration, recombinant CC16 rapidly reaches the blood, from where it is eliminated by renal filtration with a half-life of approximately 12 hours.19 There is now strong evidence from in vitro studies and experiments on transgenic mice that CC16 has an important protective role in the respiratory tract against processes of oxidative stress, inflammation, and ultimately fibrosis. It is thus very likely that a depletion of CC16 due to Clara cell damage or intravascular leakage of the protein across the breached epithelium may play a critical role in the development of lung inflammation and injury.3
CC16 IN ACUTE LUNG INJURY Because of their high biotransformation activity, Clara cells are very sensitive to lung toxicants undergoing a metabolic activation. Classic examples of compounds activated by Clara cells are the toxins, 4-ipomeanol and naphthalene, which cause
72
ANNALS NEW YORK ACADEMY OF SCIENCES
TABLE 1. Serum concentrations of Clara cell protein (CC16) in humans and rodents with short-term exposure to ozone Ozone level (ppm) Cyclists Male Female Mice C57Bl/6
C3H Rats Sprague-Dawley
Duration of CC16 controls CC16 exposed exposure (h) (µg/L) (µg/L)
p
Ref.
* *
23 23
0.076 ± 0.010 idem
2 2
11.2 ± 0.8 11.1 ± 0.6
12.3 ± 0.9 11.9 ± 1.3
0.08 ± 0.01 idem 0.11 ± 0.02 1.82 ± 0.12 0.11 ± 0.02 1.82 ± 0.12
4 8 24 3 24 3
59 ± 3 idem 62 ± 7 56 ± 4 36 ± 5 40 ± 4
75 ± 5 88 ± 3 122 ± 5 201 ± 21 74 ± 4 93 ± 8
* 3, 23 ** 3, 23 * 3, 6 *** np * 3, 6 * np
0.3 0.6 1.0
3 3 3
16 ± 3 idem idem
19 ± 3 33 ± 9 159 ± 22
ns * **
24 24 24
NOTE: np, not published; statistical significance: *p < 0.05; **p < 0.01; ***p < 0.001; all values are expressed as mean ± SE.
highly selective damage to Clara cells. Exposure of rodents to such compounds results in a rapid destruction of Clara cells, with an ensuing decrease in the abundance of CC16 mRNA and protein levels in BALF.20,21 The acute changes induced by 4ipomeanol are associated with a parallel elevation of CC16 in serum and of albumin in BALF due to increased permeation of these proteins across the damaged respiratory epithelium.20 Parallel increases of CC16 in serum and of albumin in BALF have been found in rats during LPS-induced lung inflammation.22 The high sensitivity of serum CC16 to lung epithelium injury was demonstrated in recent studies in rodents and cyclists exposed to low concentrations of ozone (O3) (TABLE 1). In a study on cyclists, for instance, the determination of CC16 in serum before and after a 2-h ride during episodes of photochemical smog revealed that ambient O3 produces a dose-dependent increase of serum CC16 at exposure levels below current air-quality guidelines (with O3 levels on average between 0.033 and 0.103 ppm23). This finding was reproduced in mice exposed to an O3 level corresponding to the new national U.S. air-quality standard (0.08 ppm). These observations demonstrate that the assay of serum CC16 allows for the detection of very subtle defects in the lung epithelial barrier permeability. However, at higher O3 levels, the increase of serum CC16 was found to correlate with the extent of lung injury as assessed by the levels of albumin, LDH, and inflammatory cells in BALF. 24 Changes in serum CC16 induced by O3 are paralleled by an elevation of the urinary excretion of CC16 resulting from an overloading of the tubular reabsorption process.24 Although O3 at high exposure levels causes a significant reduction of CC16 concentrations in BALF, CC16 mRNA levels in lung tissue were not altered, suggesting that O3 could affect the process of CC16 secretion at a posttranscriptional
BROECKAERT et al.: CC16
73
level. This hypothesis is supported by the study of Dodge et al.,12 which showed that CC16 accumulates in Clara cell granules of rats exposed for 20 months to 1 ppm O3. In humans, a transient elevation of serum CC16 has been found in firefighters after only 20 minutes of smoke inhalation.16 Of interest, this increase was found in the absence of any functional signs of lung impairment (spirometric tests), confirming the high sensitivity of CC16 as a marker of acute lung injury. CC16 IN CHRONIC LUNG INJURY The determination of serum CC16 can also be used to evaluate the chronic effects of air pollutants on the respiratory epithelium. The concentration of CC16 in serum is decreased in a dose-dependent way by tobacco smoking, with the concentration of the protein decreasing on average by 15% for each 10-pack-year.15 This effect, which has been consistently found in several independent studies,2,25 is a reflection of a concomitant reduction of CC16 in lung lavage due to the progressive loss of Clara cells. A significant reduction of serum CC16 has also been found in workers exposed to crystalline silica26 or foundry dust (unpublished data). In most studies, these changes were found in asymptomatic subjects with normal or only slightly impaired lung function tests. Changes of CC16 concentrations in BALF and/or serum have also been studied in different human lung diseases.2 A polymorphism in the 5′-untranslated region of exon 1 of the CC16 gene has recently been associated with an increased risk of asthma in childhood, with homozygous and heterozygous subjects demonstrating a 6.9fold and 4.2-fold increased risk of developing asthma, respectively.27 Of interest, the mutation is associated with a significant decrease of the serum concentration of CC16.28 This observation is in agreement with a previous work showing that the mean CC16 concentration in BALF of asthmatics was significantly decreased as compared with control healthy subjects.29 In serum, asthmatic patients with a long duration of the disease (≥10 years) had significantly lower CC16 levels than those with a short duration of the disease (
