© 2002 American Thoracic Society Clara Cell Secretory Protein Deficiency Alters Clara Cell Secretory Apparatus and the Protein Composition of Airway Lining FluidDepartment of Environmental and Occupational Health, University of Pittsburgh, Pittsburgh, Pennsylvania; Department of Anatomy and Cell Biology, University of Bergen, Bergen, Norway; Department of Human Physiology, Flinders University of South Australia, Adelaide, Australia; and Department of Cell Biology and Anatomy, University of California at Davis, Davis, California Address correspondence to: Dr. Barry R. Stripp, Department of Environmental and Occupational Health, University of Pittsburgh, FORBL Rm 314, 3343 Forbes Avenue, Pittsburgh, PA 15260. E-mail: brs2{at}pitt.edu
Clara cells represent the predominant secretory cell within distal conducting airways of mammals and exhibit functional alterations with chronic lung disease. We previously demonstrated that Clara cell secretory protein (CCSP) deficiency results in enhanced susceptibility to environmental agents. The present study was undertaken to define changes in Clara cell secretory function associated with CCSP deficiency in knockout mice. Comparative morphometry of Clara cell ultrastructure revealed dramatic alterations in secretory apparatus between wild-type (WT) and CCSP knockout (CCSP-/-) mice. Secretory granules, which occupy greater than 2% of Clara cell cytoplasmic volume in WT mice, were completely absent among Clara cells of CCSP-/- mice. Moreover, Clara cells of CCSP-/- mice exhibited a > 95% reduction in rough endoplasmic reticulum and alterations to Golgi apparatus, relative to WT controls. Ultrastructural perturbations to Clara cells were associated with altered protein composition of airway lining fluid as revealed by two-dimensional gel analysis of bronchoalveolar lavage proteins, but were not associated with altered abundance or secretion of CC26, another Clara cell secretory protein. We conclude that CCSP is required for the appearance of Clara cell secretory granules and that functional changes to Clara cells that result from CCSP deficiency lead to alterations in the composition of epithelial lining fluid.
Abbreviations: Clara cell secretory protein, CCSP • CCSP knockout, CCSP-/- • chromogranin A, CGA • calcitonin gene related peptide, CGRP • chronic obstructive pulmonary disease, COPD • cytochrome P450-2F2, CYP2F2 • rough endoplasmic reticulum, rER • smooth endoplasmic reticulum, sER • wild type, WT
Prominent ultrastructural characteristics of nonciliated bronchiolar (Clara) cells include an abundance of smooth endoplasmic reticulum and secretory granules (1). Nonciliated cells with similar ultrastructural characteristics have since been described within bronchioles of all mammalian species investigated (2, 3). Even though earlier ultrastructural analysis of Clara cells clearly indicated that they synthesized proteinaceous material to be secreted from the apical surface into the airway lumen, the nature of Clara cell secretions and the roles played by these secretory proteins in airway homeostasis are only recently being appreciated (4). Degranulation and consequent secretion of granule contents can be induced by inhaled pollutants such as ozone (5), agonists of ß adrenergic or cholinergic agents (6–8), and a variety of microorganisms or their products, including bacterial endotoxin (9) and paramyxoviruses (10, 11). The functional roles of Clara cell secretions that are induced by exposure to various environmental agents are poorly understood. However, mice deficient for the most abundant secreted product of Clara cells, Clara cell secretory protein (CCSP), exhibit altered susceptibility to inhaled and systemic pollutants, aeroantigens, and microorganisms (12–17). Moreover, studies using CCSP as a biomarker of airway changes associated with human lung disease indicate that Clara cell secretion is affected by a variety of chronic disease states such as asthma, cigarette smoking, and chronic obstructive pulmonary disease (18–22). These studies indicate that modulation of Clara cell secretory function may play an important role in airway homeostasis and disease pathogenesis. Clara cell secretory protein is a 16-kD homodimeric secretory protein expressed abundantly by nonciliated airway epithelial (Clara) cells of the lung (23–25). CCSP, known by a number of other names including urinary protein 1 (UP1), uteroglobin (UG), PCB-binding protein (PCB-BP), and Clara cell 10 kD protein (CC10), has been described as a "multifunctional protein" (26, 42), and among these many postulated functions is most frequently quoted as being a potent regulator of the inflammatory response (27–29). Studies aimed at further delineating functional roles for CCSP have included the establishment of mice homozygous for a null allele of the CCSP gene (12, 30). However, the demonstration of ultrastructural alterations to Clara cells of CCSP knockout (CCSP-/-) mice initially described by Stripp and colleagues (12, 31) and of systemic defects among CCSP-/- mice developed by Zhang and coworkers (30), suggest that the phenotype of CCSP-/- mice may be more complex that CCSP deficiency per se. Consistent with this theory is the demonstration that CCSP deficiency in mice is accompanied by altered expression of other genes within the lung (32). Altered regulation of a novel Clara cell–specific gene encoding a secreted protein suggested the potential for functional alterations to Clara cells and their secreted products, whereas altered expression of immunoglobulin A suggested that CCSP deficiency may be associated with more global alterations in lung homeostasis with either direct or indirect effects on lung immunoregulation. The present study was undertaken to investigate cellular changes within the conducting airway epithelium of CCSP-/- mice to improve our understanding of the roles played by CCSP in Clara cell secretory function and regulation of airway lining fluid composition.
Animals Wild-type (WT) strain 129 (Taconic, Germantown, NY) and CCSP-/- strain 129 mice used in this study were maintained as specific pathogen free, in-house colonies and were allowed food and water ad libitum. Representative animals from the colony were screened quarterly for the absence of pathogens using a comprehensive 16-agent serologic panel (Microbiological Associates, Rockville, MD). Male mice between the ages of 2 and 4 mo were used for all experiments. Mice used for morphologic or biochemical analysis were deeply anesthetized with pentobarbital (100 mg/kg intraperitoneally) and exsanguinated, and the lungs processed according to specific methods outlined below.
Ultrastructural Analysis
Confocal Microscopy
Western Blot Analysis of Lung Tissue and Lavage
Two-dimensional Gel Analysis of BAL Protein Composition
Clara Cells from WT and CCSP-/- Mice Show No Gross Morphologic Alterations The overall configuration of Clara cells lining terminal bronchioles of CC10-/- mice did not differ markedly from that of Clara cells from WT mice (compare Figure 1A with Figure 1B). Apices of cells projected into the airway lumen and the base was attached to the basal lamina. The nucleus was centrally placed in the cytoplasm on the basal side of the apical projection.
Ultrastructural Features of Clara Cells from WT and CCSP-/- Mice The predominant organelle in the cytoplasm of Clara cells from both genotypes of mice was the smooth endoplasmic reticulum (sER), which occupied  20% of the cell cytoplasm in Clara cells of both WT and CCSP-/- mice (Figures 2A and 3)
. Despite the similarities in abundance there were two aspects of the sER that differed markedly between WT and CCSP-/- mice. First, electron-dense material that was abundant within the lumen of sER of Clara cells from WT mice was not as evident within the lumen of sER of Clara cells from CCSP-/- mice. Second, Clara cells of CCSP-/- mice were unique in that they possessed large concentric whorls of endoplasmic reticulum located within the apical portion of the cell (Figure 4)
. Membrane-bound inclusions which appear to be surrounded by at least one layer of endoplasmic reticulum were found at the center of these whorls and contained material of varying electron density and compaction (Figure 4). Some of these inclusions have the characteristics of mitochondria based on the presence of cristae-like membrane inclusion.
In Clara cells of WT mice, rough endoplasmic reticulum (rER) occupied < 8% of cell volume and was located in long strands immediately adjacent to the nucleus in the basal portion of the cell. In contrast, rER occupied less than 0.5% of cytoplasmic volume within Clara cells of CCSP-/- mice (Figure 2B). The Golgi apparatus, which was generally located in a perinuclear region, occupied a small proportion of the cytoplasmic volume in Clara cells of WT mice and was composed of long membrane enclosed profiles surrounded by small vesicles (Figures 2B and 5) . In Clara cells of CCSP-/- mice, the Golgi apparatus occupied approximately twice the cytoplasmic volume of that observed in Clara cells of WT mice and contained a greater array of vesicles (Figure 5). Secretory granules occupied 2% of the cytoplasmic volume of Clara cells in WT mice. They had roughly circular profiles with a uniformly electron-dense matrix surrounded by a single membrane, and were usually located near the apical membrane or in the cytoplasm in close relationship to the Golgi apparatus (Figure 1). In CCSP-/- mice, secretory granules were absent from the cytoplasm of Clara cells (Figure 2B). Mitochondria were abundant throughout the cytoplasm of WT and CCSP-/- mice and occupied approximately the same proportion of cytoplasmic volume (Figures 1 and 2A). In both WT and CCSP-/- mice, the configuration of the mitochondria varied from large circular profiles surrounded by two membranes with a small number of internal cristae to narrow oblong profiles with large numbers of internal cristae (Figures 1 and 3). The nuclei in Clara cells of both genotypes had an oblong profile, with a small amount of heterochromatin associated with the nuclear envelope and a relatively prominent nucleolus (Figure 1). Taken together, these results demonstrate significant perturbations in Clara cell ultrastructure associated with CCSP deficiency that principally affect organelles of the secretory compartment.
Localization and Secretion of Clara Cell 26 kD Protein The cellular abundance and distribution of the Clara cell–specific secretory protein CC26, a nonselenium glutathione peroxidase (35), was compared in lung tissue of WT and CCSP-/- mice. Dual immunofluorescence analysis of CCSP and CC26 expression in bronchioles of WT mice demonstrated coexpression of these proteins within individual Clara cells (Figure 6) . However, the intracellular distribution appeared to be distinct. CC26 immunoreactive protein was present at higher concentrations within the central cytoplasm, whereas CCSP was localized more to the peripheral cytoplasm of the Clara cell. This finding was supported by optimal dilution analysis, which demonstrated that CC26 was concentrated in the apical cytoplasm basal to the luminal projection of WT Clara cells, whereas CCSP was equally distributed in the apical projection and the apical cytoplasm basal to it. The general distribution and overall cellular abundance of CC26 was similar at all airway locations of WT and CCSP-/- mice. CCSP was undetectable in CCSP-/- mice by immunofluorescence or immunoperoxidase.
CC26 abundance within lung tissue and airway fluid of WT and CCSP-/- mice was determined by Western blot analysis (Figure 7) . Consistent with the immunofluorescence and immunoperoxidase analysis, CC26 levels were similar within lung tissue homogenate and airway fluid of WT and CCSP-/- mice. In wild-type mice, the ratio of BAL:tissue content for CCSP and CC26 differed markedly. CC26 was only weakly detected in BAL, with higher levels present within lung tissue. In contrast, CCSP was present at higher levels per unit BAL protein than per unit lung tissue protein. These data, together with data showing differences in the intracellular distribution of CCSP and CC26, suggest that both intracellular routing and mechanisms of secretion are different for these two proteins. Similarities in the distribution and abundance of CC26 within lung tissue and airway lining fluid of WT and CCSP-/- mice suggest that this aspect of Clara cell protein synthesis and secretion is not compromised by CCSP deficiency.
CCSP-/- Mice Exhibit Altered Airway Lining Fluid Protein Composition To determine if CCSP deficiency and the associated changes in Clara cell secretory apparatus resulted in altered protein content of airway lining fluid, total BAL protein from either WT or CCSP-/- mice was separated by two-dimensional gel electrophoresis. A representative image of silver-stained 2D profiles of pooled BAL proteins from groups of three WT and three CCSP-/- mice is shown in Figure 8 . A large number of protein species can be discerned. Closed arrows indicate reference protein spots common to BAL from WT and CCSP-/- mice. Open arrows denote protein species exhibiting differential abundance between BAL from WT and CCSP-/- mice. These profiles demonstrate that only a limited number of proteins resolved in 2D gels were more abundant within airway fluid of WT mice relative to CCSP-/- mice, and that these proteins are generally of higher molecular weight. In contrast, a larger number of protein species show increased abundance within BAL of CCSP-/- mice and these were generally of lower molecular weight. These differences in 2D protein profiles of BAL samples from pooled representatives of WT and CCSP-/- backgrounds further support the notion that CCSP deficiency leads to changes in Clara cell function that influence normal airway homeostasis.
In the present study we define ultrastructural changes to Clara cells in vivo that result from establishment of a homozygous null allele of the gene encoding CCSP. Clara cells of CCSP-deficient mice lack secretory granules that are characteristic of this cell type in WT mice and all other mammalian species. Moreover, a dramatic reduction in the cytoplasmic fraction occupied by granular endoplasmic reticulum indicates the potential for more global changes in the biosynthesis of secreted proteins by Clara cells. Despite ultrastructural changes to the Clara cell secretory apparatus, CCSP-/- mice exhibit normal expression and distribution of CC26, another Clara cell–specific secretory protein. However, differences between WT and CCSP-/- mice in the protein composition of airway lining fluid suggest that alterations to Clara cells that result from CCSP deficiency have the potential to alter normal functions of the airway lining fluid. The demonstration of ultrastructural and biochemical changes to airways of CCSP-/- mice suggest that these alterations, coupled with CCSP deficiency, lead to compromised lung function that culminates in the previously described lung phenotype, most notably pollutant susceptibility and reduced host pathogen defense. Very little is known of Clara cell secretory granule biogenesis that may account for the current findings. Interestingly, biochemical properties of CCSP, which include an acidic PI and calcium binding, are similar to those of another family of secreted proteins produced by neuronal and neuroendocrine cells, the chromogranins. Among these, chromogranin A (CGA) has been shown to regulate dense core granule biogenesis in vitro (36). Like CGA, CCSP is a major constituent of the secreted protein of Clara cells, accounting for up to 40% of the total secreted product (23). Also, consistent with our observations of a lack of Clara cell secretory granules and altered airway lining fluid protein content in CCSP-/- mice, Kim and coworkers demonstrated that downregulation of CGA expression in PC12 cells results in a loss of dense core granules and reduced expression of other secretory granule proteins (36). As such, our finding of defects in Clara cells of CCSP-/- mice may reflect roles for CCSP in regulating secretory granule biogenesis. Despite the paucity of information regarding biogenesis and packaging of secreted proteins from Clara cells, considerably more work has been performed investigating mechanisms of Clara cell secretion. A number of studies have demonstrated that secretion of granule contents by Clara cells is a highly regulated process, activated in vivo by either adrenergic or cholinergic agents (37). Physical stimuli have also been shown to trigger Clara cell degranulation through mechanisms that are independent of signaling through the ß-adrenergic system (38). Less well characterized are the mechanisms regulating Clara cell degranulation after Sendai virus infection and ozone exposure (5, 10, 11). Degranulation of Clara cells after ozone exposure precedes their entry into the cell cycle, implying that epithelial cell injury and subsequent signaling events contribute to the synchronization of protective responses, such as degranulation, and activation of epithelial regeneration. As such, the lack of secretory granules and associated alterations in the composition of airway lining fluid may compromise the natural protective functions of the airway epithelium, leading to increased ozone susceptibility among CCSP-/- mice. Despite the clear demonstration of pathways for regulated Clara cell secretion and our finding that Clara cells of CCSP-/- mice lack secretory granules, expression of CC26, another Clara cell–specific protein, was unaffected in lungs of CCSP-/- mice. This indicates that either general protein synthesis and routing are unaffected among Clara cells of CCSP-/- mice, or that CC26 is secreted through an independent pathway that is not affected by CCSP deficiency. The demonstration of altered airway lining fluid protein composition in CCSP-/- mice despite the lack of altered CC26 expression and secretion supports the notion of a Clara cell secretory defect affecting some but not all Clara cell secretory proteins, and suggests the existence of multiple independent pathways for Clara cell secretion. Another possible explanation for alterations in Clara cell secretion in CCSP-/- mice is that CCSP may interact with other proteins within the ER and thereby influence the routing or folding of other secretory proteins. Clara cell secretory protein has been shown to interact with microsomal and plasma membranes through binding to an integral membrane protein (39, 40), and has been shown to bind other secreted proteins such as fibronectin (30). Lack of these interactions may result not only in structural perturbations to Clara cells, but may also result in alterations both in the secretion of Clara cell proteins and stability of extracellular proteins derived from either Clara cells or other secretory cell types. Findings of the present study raise two possible explanations to account for ozone susceptibility of CCSP-/- mice; either ultrastructural changes to Clara cells compromise their integrity and innate resistance to ozone-induced injury, and/or changes in airway lining fluid composition in CCSP-/- mice reduce its ability to protect against inhaled oxidants. The finding of a dramatic increase in Clara cell necrosis in both proximal and distal airways of ozone-exposed CCSP-/- mice relative to coexposed WT mice supports the notion of reduced Clara cell tolerance to oxidant challenge. Comparative histopathology of ozone-exposed WT and CCSP-/- mice also suggests that ciliated cells, particularly of proximal airways, are more extensively injured among CCSP-/- mice after exposure to 1.0 ppm ozone. This may either result from a failure of Clara cells to confer protection to adjacent ciliated cells from the toxic effects of ozone exposure, and/or reduced protective qualities of airway lining fluid among CCSP-/- mice. Patterns of ozone-induced cellular injury are consistent with findings of the present study and suggest that structural changes to Clara cells, coupled with biochemical changes in airway lining fluid composition, result in altered airway function culminating in ozone susceptibility. Chronic lung diseases of humans such as chronic obstructive pulmonary disease (COPD) and asthma, and airway anomalies associated with cigarette smoking, are associated with changes in the abundance of CCSP in airway fluid and serum (18–22). Alterations to Clara cells have not been documented among humans with chronic lung disease. However, changes to the airway epithelium have been described for horses with COPD, which exhibit a similar pattern of altered Clara cell ultrastructure to that observed within Clara cells of CCSP-/- mice (41). In particular, equine COPD is associated with a loss of the normal differentiated characteristics of Clara cells, such as secretory granules, and the appearance of lamellar inclusions. Whether these changes to airways of horses are causally associated with progression of COPD has not been determined. However, the susceptibility of CCSP-/- mice to inhaled oxidant pollutants and virally-induced airway inflammation suggests that alterations in Clara cell function that are associated with chronic lung injury may play an important role in disease pathogenesis and may even predispose the lung to further insult by environmental agents. Further studies investigating roles for CCSP in the regulation of Clara cell secretion and airway function will yield important new insights into mechanisms of airway defense in health and disease.
These studies were supported by NIH Grants ES-08964, HL-64888, and HL-70575. Received in original form February 19, 2002 Received in final form April 4, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
