Murine complement deficiency ameliorates acute cigarette smoke-induced nasal damage

Abstract

OBJECTIVE:

Exposure to cigarette smoke is a risk factor for chronic rhinosinusitis. Current literature confirms complement fragments are activated in human nasal mucosa. The mechanism(s) responsible for this activation is unclear. We investigated the effects of cigarette smoke on nasal mucosa in vitro and via a model of cigarette smoke exposure by using animals deficient in complement components.

STUDY DESIGN:

Prospective, controlled animal and in vitro human cell line study.

SETTING:

University laboratory.

SUBJECTS AND METHODS:

Human respiratory epithelial cells were exposed to five, 10, and 20 percent cigarette smoke extract (CSE) in vitro in the presence or absence of human serum. Complement activation was assessed by enzyme-linked immunosorbent assay and immunofluorescent techniques. Complement-deficient (C3^−/−, n = 6; factor B^−/−, n = 50) and sufficient mice (wild type, n = 10) were exposed to the smoke of four cigarettes per exposure for two exposures per day for three days. Mice were sacrificed 12 hours after the last exposure, and the nasal cavity was surgically removed. Histological characteristics were analyzed by the use of a subjective scale and quantitative image analysis scoring systems.

RESULTS:

In vitro analysis of respiratory cell cultures demonstrated that exposure of serum to CSE resulted in complement activation. Furthermore, immunofluorescent staining for C3d could only be demonstrated in CSE-exposed cultures. In vivo analysis demonstrated that complement deficiency, either C3 or factor B deficiency, resulted in a significant reduction in histological evidence of damage as compared with wild-type control mice (wild type vs C3^−/−, P = 0.02; wild type vs factor B^−/−, P = 0.07; no significant difference between C3^−/− vs factor B^−/−).

CONCLUSION:

These data demonstrate that cigarette smoke activates the complement system. Furthermore, complement deficiency protected against smoke-induced mucosal damage in this small series.

Each year, chronic rhinosinusitis (CRS) is estimated to affect 14 percent of the adult U.S. population.¹ Not surprisingly, CRS is repeatedly found to have an adverse impact on the daily activities, productivity, and cognitive function of patients.

Despite these widespread negative consequences, the treatment of CRS apart from surgical intervention remains largely symptomatic and often does not induce remission of underlying mucosal inflammation. Expertly defined as “inflammatory conditions involving the paranasal sinuses, as well as the lining of the nasal passages that persist beyond 12 weeks,” a more complete understanding of the inflammation characterizing CRS and its mechanism(s) is regarded as the gateway to developing more definitive treatment options.² Furthermore, as new types of therapeutic options are developed for other diseases, understanding the pathogenesis of CRS is crucial to predict the therapeutic outcomes of CRS patients to these novel treatments and to research them appropriately.³

Until recently, much of the study of CRS pathogenesis focused on the adaptive immune system in isolation, including the relative balance of T helper cell 1 versus T helper cell 2 (Th₂) cytokine milieu in response to ubiquitous environmental stimuli, such as bacteria and fungi. These lines of evidence, however, do not explain why only some subjects develop CRS and other related conditions to pathogens to which nearly all humans are exposed. Environmental stimuli presented to a person’s nasal mucosa are not unique pathogens; it seems plausible that variations in individual immunologic responses to otherwise-ubiquitous stimuli may contribute to the development of CRS. An emerging theory is the role of the innate immune system and, more important, how it interacts with and modifies responses of the adaptive immune system to induce deregulated immune responses leading to chronic inflammation.

Innate immunity is the basic defense of the sinonasal cavity, including anatomical and physiological features of the epithelium, which functions to limit the penetration of environmental stimuli and pathogens, thereby lessening the induction of acquired immunity.⁴ Injury to the epithelial barrier has been implicated in the inflammatory response of asthmatic subjects by increasing exposure of triggers of innate and adaptive immunity. In recent reviews^3–5 describing the dynamic barrier role of sinonasal mucosal epithelium, authors characterize tight junctions, mucociliary clearance, the presence of pattern recognition receptors that recognize common microbial antigens, production of natural antimicrobial peptides and inflammatory mediators, the complement system, and interaction with the adaptive immune response all as mechanisms by which sinonasal epithelium curbs pathogenic inflammation.⁵ Therefore, implied is the possibility that a breakdown or dysfunction in the innate immune response may lead to overstimulation of the adaptive immune response by allowing exposure to unchecked stimuli, resulting in the pathogenic inflammation characteristic of CRS.

The complement system is proposed as one possible area of dysfunction in the mediation and regulation of the adaptive immune response. Complement is a tightly regulated system of proteins existing in plasma and on the cell surface that functions to identify and eliminate potential pathogens, as well as contribute to homeostasis of nonimmunosurveillance systems. There are three major pathways of complement activation that converge in formation of a complement protein, component 3 (C3) convertase, thereby leading to further cleaving and amplification of peptides to generate opsonins, chemoattractants, anaphylatoxins, and the membrane attack complex. The classical pathway primarily responds to antigen-antibody binding. The lectin pathway is induced by mannan binding lectin binding to microbes via surface mannose residues. The alternative pathway follows recognition of surface-bound complement component 3b (C3b), the product of spontaneous C3 hydrolysis, by factor B (fB); its subsequent steps may lead to amplification of the complement system or formation of the membrane attack complex. Because of its multiple mechanisms and roles in immunosurveillance and homeostasis regulation, dysfunction of the complement system is known to contribute to pathology and disease.

Complement has been shown to be produced and activated by airway epithelium after stimulation of innate immune toll-like receptors and has been repeatedly proposed as a contributing factor in the pathogenic inflammation of CRS.^6–8 Decreased levels of decay-accelerating factor, a regulator of the complement system, were recently shown to be associated with an increased susceptibility to allergic respiratory diseases.⁹ Further, C3 and mannan binding lectin levels have been shown to be increased in CRS patients as compared with control subjects.¹⁰ Therefore, the possibility of immune imbalance contributing to CRS may be attributable to a disproportionate relative increase in complement system components by more than one specific mechanism.

Experiments performed more than a decade ago demonstrated that extracts from cigarette smoke were capable of cleaving complement proteins in serum in vitro, with the mechanism of activation believed to be alternative pathway dependent.¹¹ Whether this occurs in vivo has been less well characterized. Exposure to cigarette smoke is an established risk factor in the development of CRS.^12,13 Floreani et al¹⁴ demonstrated increased airway epithelial cell expression of intercellular adhesion molecule-1, implicated in inflammatory cell binding and recruitment, upon exposure to a combination of cigarette smoke and complement component 5a (C5a). Furthermore, intercellular adhesion molecule-1 expression was greater with the combination as compared with exposure to either smoke or C5a alone, implicating that exposure to smoke may potentiate the functional responsiveness of the complement component 5a receptor (C5aR) to the forthcoming complement cascade.¹⁴ Given these observations, the complement system appears as a plausible target for pharmacologic intervention in the future treatment of CRS.

Given these data, we sought to investigate whether cigarette smoke extract (CSE) could activate the complement system in vitro and whether this activation led to complement deposition on respiratory epithelial cells. Furthermore, we assessed the role of the complement system on mucosal alterations caused by acute exposure of mice to cigarette smoke by using animals deficient in C3, the central component, and fB, an alternative pathway protein.

Methods

Cell Culture

Normal human bronchial epithelial (NHBE) cells were purchased from Lonza Corp. (Walkersville, MD) and cultured in bronchial epithelial growth medium according to product specifications. NHBE cells were expanded on human collagen type IV-coated 75-cm² tissue culture dishes and plated on collagen type IV-coated 6-well plates for treatment, as described previously.¹⁵

CSE

CSE was made by bubbling the smoke from four 3R4F cigarettes (University of Kentucky) through 50 mL of tissue culture media by the use of a Shapiro cigarette smoke machine (Washington University, St. Louis, MO). Media were filtered through a 0.22-m filter (Corning Life Sciences, Lowell, MA) and used immediately. CSE was diluted with fresh media to create five, 10, and 20 percent final concentrations, as previously described.¹⁵

Experimental Treatment Conditions

NHBE cells were grown to approximately 90 percent confluency on human collagen IV-coated 6-well dishes. Epithelial cells were cultured in 0, five, 10, or 20 percent CSE in the presence of 10 percent pooled human serum (Quidel Corporation, San Diego, CA) for 24 hours. Supernatants were collected and stored at −80°C until analyzed. Cells were fixed in four percent paraformaldehyde and stained with C3d-fluorescein isothiocyanate (Dako, Carpentia, CA) by the use of standard immunofluorescent techniques and visualized with Zeiss Axiovert 200.

Complement Activation

Culture supernatants were assayed for the presence of the complement activation fragments complement component 4a (C4a), C3a, and C5a by commercially available enzyme-linked immunosorbent assay (ELISA; BD Biosciences, San Jose, CA).

Animals

Male C57BL/6 mice (The Jackson Laboratory, Bar Harbor, ME), C57BL/6 C3^−/− mice, and C57Bl/6 fB^−/− mice (in-house breeding colony) were housed in temperature- and light-controlled chambers on a 12-hour light/dark cycle and provided with rodent chow (Purina 5001) and water ad libitum.

Cigarette Smoke Exposure

We performed a preliminary study to access whether acute exposure to cigarette smoke resulted in nasal epithelial injury/damage. We used the acute exposure protocol previously described by Doz et al.¹⁶ In brief, C3^−/− (n = 6), fB^−/− (n = 5), and wild-type (WT) (n = 10) male mice, eight to 10 weeks of age, were subjected to the smoke of four unfiltered cigarettes per exposure, with mice receiving two exposures per day for three days. Control animals (C3^−/−, fB^−/−, and WT; n = 3 in each group) were exposed to room air. Twelve hours after the final exposure, the mice were sacrificed and nasal tissues were collected for histological assessment. All procedures were approved by the Medical University of South Carolina Committee for Animal Research in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals.

Histological Analysis

Mouse heads were removed for histological examination of nasal compartments. Skulls were skinned, fixed for 48 hours in buffered formalin (4%), and decalcified during two weeks by the use of a 14 percent ethylenediaminetetraacetic acid (Sigma-Aldrich, St. Louis, MO) solution. Coronal sections of the skulls at the middle third between nose-tip and orbit were made and stored in formalin before paraffin processing. After dehydration and embedding in paraffin, 5-μm sections were stained with Alcian blue and periodic acid-Schiff stain. An anatomical plane consisting of the nasal septum extending in view with nasal-associated lymphoid tissue was used to standardize the cut of each sample, as shown in Figure 1A. Sections were graded by an investigator blinded to the study groups and scored with the use of a two-component Likert scale and objective data collection.

Figure 1.

(A and B) Annotations highlight the sites where histological measures were made: (1) nasal septum extension, (2) inferior border, (3) nasal-associated lymphoid tissue, and (4) lateral border.

Subjectively, sections were first analyzed on a 0- to 4-point scale for overall subjective damage as determined by edema, ciliary damage, and epithelial disarray, as previously described by Lindsay et al.¹⁷ A score of 0 denoted no damage observed, whereas a score of 4 indicated maximal amount of histological insult to most respiratory epithelium. A second set of subjective measures regarding apocrine cytoplasmic secretion, or blebbing, was taken in the same plane and graded 0 to 2.¹⁷ No observed blebbing was scored 0, with severe blebbing scoring 2. The cumulative scores from both systems where then expressed with a score ranging from 0 to 6.

Objective data were collected in the same plane but limited to observation at three consistent anatomic locations on the right side of the murine sinonasal cavity, as shown in Figure 1B: the nasal septum, the inferior nasal cavity border, and the lateral maxillary sinus wall. Two independent sets of measurements were taken at each location, consisting of area of epithelium in μm², length of epithelium in μm, and goblet cell count in line with previously published methods.¹⁸ These measurements were computed to provide indices of average epithelial thickness and goblet cell count per μm of epithelial length.¹⁸

Statistical Analysis

Statistical analysis was performed by use of the Prism software package (Graphpad, La Jolla, CA) and SigmaStat 3.5, Sample Power 2.0 (SPSS, Inc., Chicago, IL). Data were compared with the use of paired or unpaired t tests, where appropriate. Values of P < 0.05 were considered significant. Given the small sample size of the murine model and the inherent sampling bias associated with histological analysis, nonparametric analyses of these data were considered appropriate, and Mann-Whitney U tests were used.

Results

Studies performed almost a decade ago demonstrated that exposure of human serum in vitro to CSE resulted in cleavage and activation of complement components.¹⁹ These pivotal studies demonstrated that, along with the classical pathways to complement activation (classical, lectin, and alternative pathway), complement could be directly activated by chemical or proteolytic activation/cleavage into biologically active fragments. Building on these observations, we performed similar in vitro experiments, but unlike previous studies, we performed these in a cell culture system to determine whether activation of the complement system by CSE results in complement fragments that could lead to membrane association and deposition.

We first sought to reassess whether complement activation occurred as a result of CSE exposure in respiratory cell cultures exposed to 10 percent human serum and graded concentration of CSE. Complement activation was quantified in cell culture supernatants by the use of complement anaphylatoxin ELISAs to demonstrate the presence of C4a, C3a, and C5a. Our data demonstrate that complement activation occurs in the presence of CSE at all concentrations (Fig 2). Although complement activation was observed in cultures that were not treated with CSE for C4a, C3a, and C5a, these levels were significantly decreased (C4a, P < 0.005; C3a, P < 0.004; and C5a, P < 0.001) as compared with treated samples and probably reflects normal complement serum “tickover.” Cleavage of C3 into C3a and C3b results in the exposure of a highly reactive thiol ester bond, which enables membrane deposition of C3b. This initial membrane association has the ability to amplify the complement cascade that could ultimately lead to cell lysis. To see whether CSE cleavage of complement generates biologically activate C3b that has the ability to membrane associate, we stained respiratory cell cultures with fluorescein isothiocyanate-labeled anti-C3d antibody. Our staining revealed the presence of deposited C3d in all CSE-treated cultures (data not shown), whereas C3d immunostaining could not be detected on untreated cells.

Figure 2.

ELISAs demonstrating presence of (A) C4a, (B) C3a, and (C) C5a in CSE-exposed epithelial cultures. Note the significant increase in complement fragments in CSE-treated cultures (P < 0.05).

In light of these culture findings and considering that smoking is a significant risk factor for CRS, we analyzed histological sections of murine nasal tissue that had been acutely exposed to whole body cigarette smoke. Histological samples from cigarette smoke exposed mice were scored for evidence of mucosal damage by the use of a modified Likert scoring system. Analysis demonstrated that complement deficiency, either C3 or fB deficiency, resulted in a significant reduction in histological evidence of damage as compared with WT control mice (WT vs C3^−/−, P = 0.02; WT vs fB^−/−, P = 0.07; no significant difference between C3^−/− vs fB^−/−) (Figs 3A and 4). To build upon these observations, we performed quantitative analysis to measure changes in epithelial thickness, as previously described.¹⁸ There was a significant increase in the epithelial thickness to epithelial length in WT mice as compared with nonsmoked control mice and both fB^−/− and C3^−/− (P = 0.02, P = 0.01, and P = 0.004, respectively) (Fig 3B).

Figure 3.

(A) Histological score per group based on a two-part Likert scale (0–6). (B) WT mice had a significant increase in epithelial thickness as compared with complement deficient mice (^#P < 0.012, ^##P < 0.004).

Figure 4.

Representative sections of WT nonsmoke-exposed (A), WT smoke-exposed (B), and fB^−/− smoked-exposed (C) mice from the inferior border of the nasal cavity.

Discussion

In the setting of inflammation, epithelial cells have previously been shown as an extrahepatic source of some non-terminal pathway complement proteins when challenged with various stimuli. The targeted focus of this study was C3, the convergence point for all three complement pathways, and fB, a protease of the alternative complement system that functions as an amplification loop. Recently, C4, C3, and fB gene expression were all shown to be up-regulated in the sinus mucosa of patients with allergic fungal rhinosinusitis and CRS without nasal polyps.¹⁹ Furthermore, immunohistochemical and ELISA studies that use human patient samples have demonstrated the presence of complement deposition for C3 and complement component 9 on the sinonasal mucosa and increased nasal secretions of C3a.¹⁹ Taken together, these data support a role for complement in the pathogenesis of CRS. However, it is difficult to ascertain, from these studies, the pathways to complement activation.

The nasal mucosa is a direct inlet for pathogens, noxious environmental stimuli, and particulate matter to interact with the innate immune response. There are plausible triggers for each of the three traditional complement pathways in commonly encountered microbes and pollutants. Additionally, allergen proteases, cigarette smoke, and induction of a Th₂ cytokine profile all have been shown to activate complement independent of the commonly described pathways, as we have demonstrated here for cigarette smoke exposure.

The complement system plays a key role in host immunity, but excessive or inappropriate activation of the system can lead to direct tissue injury and, via the production of C3a and C5a, excessive inflammation. C3a and C5a are potent chemotactic peptides for neutrophils and macrophages through either a direct effect or by the concerted up-regulation of adhesion molecules and chemokines.^8–10 Furthermore, neutrophil and macrophage-derived tumor necrosis factor-α and interleukin-1β have been shown to stimulate C3 production by respiratory epithelium, and the intrinsic production of complement by epithelial cells in response to cytokines may further perpetuate the proinflammatory milieu present within the nasal mucosa. Furthermore, the presence of C3 opsonins, as we have shown in our culture experiments, can act as ligands for complement receptors on immune effector cells, or as the building blocks for assembly of the cytolytic membrane attack complex.

Herein, we performed preliminary studies that show that acute exposure to cigarette smoke results in epithelial damage, as evidenced by edema, epithelial blebbing, and changes to the gross epithelial structure. Our in vitro analysis clearly demonstrated that exposure of serum to cigarette smoke in vitro resulted in complement activation and deposition of complement on respiratory epithelial cells. Given these observations, we performed acute cigarette smoke experiments by using animals deficient in complement components, C3 and fB, and have clearly demonstrated that complement deficiency protects against epithelial cell damage and disarray. Taken together, these preliminary data suggest that complement is activated by cigarette smoke and that complement plays a role in damage to the sinonasal respiratory epithelium as a consequence of acute cigarette exposure. An interesting addendum to the current model would be challenging the mice with stimuli known to evoke a Th₂ cytokine response immediately after the course of acute smoke exposure. If the inflammatory response were more pronounced, this would lend support to the role of complement in mediating acquired immunity. There are no large clinical studies linking cigarette smoke and CRS; however, clinicians are pivotal to developing these observational studies and trials. Understanding the implications of this murine model is likely to encourage future research in human CRS.

Conclusions

In conclusion, we have demonstrated that exposure of complement to CSE leads to complement activation that results in the production of biologically active cleavage fragments and C3 opsonins. Furthermore, we demonstrated, in a small series of animals, that complement deficiency protects against nasal epithelial and mucosal changes associated with acute cigarette smoke exposure.