Strenuous exercise characteristic of endurance sports is believed to have a detrimental effect on the integrity of the structure of the respiratory epithelium. Such damage may lead to increased bronchial hyperresponsiveness to non-specific stimuli and contribute to the increased prevalence of exercise-induced bronchoconsriction observed in elite endurance athletes [1–4]. Endurance sports may be also associated with increased susceptibility to respiratory infections [5], which in turn, may exert a destructive effect on the airway epithelium.

Various proteins have been assessed as potentially useful tools in monitoring airway inflammation and epithelial damage [6–9]. Clara cell secretory protein (CC16), a protein secreted by non-ciliated cells of the bronchioles, has also been studied as an indicator of epithelial barrier disruption in the lower airways. Significant efforts have been undertaken to explain its role in airway inflammation and, although its function has not been fully elucidated, CC16 is known to have anti-inflammatory and anti-oxidative properties [10]. However, CC16 levels in serum depend not only on its production by Clara cells and possible leakage through the disrupted epithelial barrier, but also on renal clearance. Moreover, multiple factors may influence final CC16 serum concentration (e.g., diurnal variation, age, ethnicity, atmospheric conditions, allergen exposure, exercise). As a result of these concerns the utility of serum CC16 level measurements in monitoring of the activity of pulmonary disease, airway inflammation and chronic or acute epithelial damage is disputed (reviewed by LaKind et al. [11]).

Changes in serum CC16 levels were observed after acute, short-lasting exposure to exercise or provocative or irritating stimuli. Non-specific challenges with exercise and eucapnic voluntary hyperpnea (EVH) resulted in increased urinary CC16 levels [12, 13]. In adult swimmers, exposure to trichloramine, a by-product of swimming pool disinfection, caused transient increase in serum CC16 levels while no such changes were observed in children [14].

Chronic airway inflammation may also be associated with shifts in local and systemic presence of CC16. During pollen season, CC16 in nasal lavage fluid is decreased in subjects with birch pollen-induced allergic rhinitis [15]. Decreases in serum CC16 levels in asthmatics are independent of their atopic status and correlate positively with duration of disease [16]. CC16 is believed to have not only immunomodulatory and anti-inflammatory properties but also to possibly influence the anti-infectious response: an increased persistence of viruses and prolonged viral-specific gene expression were observed in CC16-deficient mice in response to RSV infection [17].

The aim of this study was to investigate the airway epithelium inflammation and potential damage in competitive athletes, as reflected by changes in serum CC16 levels. Secondly, we sought to identify possible associations of serum CC16 level with pattern of exercise, asthma or allergy, frequent respiratory infections and the immune response to respiratory pathogens reflected by pathogen-specific IgG levels in the serum.

This study documents that serum CC16 concentrations are significantly lower in top athletes than in control subjects from the general population. Decreased serum CC16 concentrations were reported in infants and children regularly visiting indoor pools [22, 23] but other studies brought conflicting results regarding the effect of single and repeated training session on serum CC16 [24, 25]. Discrepancies are believed to be due to intensity of exertion and breathing pattern employed during various types of athletic activity (reviewed by LaKind et al. [11]). Our study, including large and heterogeneous group has shown decreased serum CC16 levels in professional athletes, irrespective of the type or pattern of high-level activity. However, due to the limited number of athletes representing each discipline, it was impossible to assess the influence of each type of sport on serum CC16 concentration. We have not ascertained any influence of predominant exercise pattern (endurance versus non-endurance) on athletes’ serum CC16 levels. Although withdrawal of swimmers and triathletes from the serum CC16 analysis did not influence final result, these data - in our opinion – are too scarce to conclude that unfavourable training environment (e.g., trichloramine exposure) does not affect serum CC16 concentration.

CC16 is a protein of multifaceted characteristics. It may be considered both as a marker of the disruption of airway epithelial barrier integrity and as a co-participant in anti-inflammatory response. Degree of CC16 involvement and significance in the aforementioned processes is still a matter of research and dispute.

Our results suggest, that it is an intensive and regularly performed exercise per se that stimulates airway inflammation, irrespective of training intensity. Reduced presence of CC16, an anti-inflammatory factor, may additionally contribute to airway inflammation and hyperresponsiveness.

In addition, the control non-athletes asthmatic patients were found to have CC16 levels similar to those in healthy controls, which contrasts with the results of other report, documenting a significant decrease of serum CC16 in asthma [16]. In our opinion, the possible explanation for this discrepancy is the fact that our asthmatic controls group comprised exclusively subjects with mild disease well controlled on anti-inflammatory treatment. Therefore, it is possible that that changes in serum CC16 level reflect the grade of airway inflammation in asthma, similarly as it was observed in regard to sputum CC16 in subjects with COPD [26]. Our data suggest that the issue of CC16 in chronic asthma be extensively studied in a well-defined population.

Although recent data published by Jacobs et al [27] show that lower serum CC16 levels may be associated with allergic sensitisation, no differences in CC16 levels between atopics and non-atopics were observed in our population of athletes. Therefore, the associations of allergic sensitisation with markers of epithelial barrier disruption require further study.

Another novel observation is that athletes declaring frequent upper respiratory tract infections had significantly lower serum CC16 concentrations than those without frequent infections - Figure 2. Accordingly, a higher risk of frequent respiratory infections was associated with low serum CC16 level. Low serum CC16 was found to be also associated with patients’ reported symptoms of rhinitis, but not with physician-diagnosed rhinitis. This apparent discrepancy may be explained by the fact that symptoms declared as “rhinitis” were, in fact, symptoms of upper respiratory infections. This is congruous with observation that both frequent of URTIs and reported rhinitis symptoms are considerably more likely to be encountered in athletes with serum CC16 below 5 ng/ml.

The common belief that moderate exercise is beneficial to one’s health whereas excessive physical training may considerably increase the susceptibility to infection was confirmed by the concept of the U-shaped relationship between physical activity and resistance to disease [5]. Our study comprised top-level athletes and, although only 19.2% of them claimed to be suffering from frequent respiratory infections, those who reported infections had significantly lower serum CC16 concentrations. This interaction is further strengthened by significant associations of decreased serum CC16 levels with frequent declared respiratory infections and nasal symptoms and indicates, that the immune response against some respiratory pathogens may be modified by competitive sporting activity. Positive correlation of anti-PIV IgG and serum CC16 may be considered an indirect proof for immunomodulatory activity of the latter, but the lack of such correlation for the remaining pathogens is somewhat puzzling and suggests the need for further study of the influence of CC16 on the susceptibility to infection with particular pathogens.

Less frequent positive serological test results against PIV, RSV and M. pneumoniae in professional athletes compared to controls may be interpreted as indirect evidence for impaired immunocompetence resulting from continuous high exercise load. In a mouse model, CC16 was shown to influence the lymphocyte response of the airways after infection with RSV. Mice lacking CC16 demonstrated significantly higher concentrations of Th2 cytokines IL-5 and IL-13 in BALF for several days following the infection [17]. The stimulatory activity of RSV infection on Th2 cytokine synthesis and release may explain the association of atopy with more frequent occurrence of positive testing for anti-RSV IgG in atopic athletes.

The frequency of positive immune response to RSV was comparable between atopic athletes and healthy control subjects, but it was significantly lower in non-atopic athletes. In addition, correlations between serum CC16 level and intensity of immune response to respiratory viruses (RSV, PIV and adenoviruses) were only significant in atopic athletes. Given that serum CC16 levels in athletes were not seen to be modified by atopy status, it may suggest, that atopy per se influences the immune response to respiratory viruses.

Our study has several limitations. The study group was quite heterogeneous in terms of predominant pattern of exercise and training environment not allowing for identification of possible effects of any particular sport on CC16 or immune response against respiratory pathogens. Accordingly, no standardised and comparable measures of exercise load/intensity could be included into analysis. Furthermore, the AQUA question concerning respiratory infections does not determine their actual frequency, allowing only for the subjective assessment of whether the URTIs are frequent or not.

Another limitation of our study results from the inter-subject variability of CC16 levels. Overlaps in the serum CC16 are observed between the groups and it could be argued that outliers influence the overall result. Significant differences were shown through non-parametric tests which makes possibility of such influence less likely. However, this issue should be considered possible confounder and addressed in next studies on CC16 in larger populations.

For the purpose of logistic regression analysis, 5.0 ng/ml was set arbitrarily as the CC16 threshold value, based on the reference value given by the manufacturers of the ELISA kits. However, the CC16 levels in healthy subjects are variable and the accepted reference values for CC16 are actually lacking, which substantially hinders such analysis and signals the need for further assessment of CC16 levels in various clinical models [7, 24, 28].

Bronchial hyperreactivity was assessed only in few patients (due to their training schedules and low accessibility) and could not be analysed in this study. However, of the role of CC16 in the context of bronchial hyperresponsiveness may be important for understanding the pathogenesis of EIB, especially in view of the data showing association of CC16 gene polymporphism with airway hyperresponsiveness and asthma phenotype [29].

Our study demonstrated that regular high-load exercise training is associated with a decrease in serum CC16 protein levels and lower CC16 levels in competitive athletes are associated with increased prevalence of reported respiratory infections. Further studies are necessary to elucidate the mechanism of association between increased susceptibility to respiratory infections in top athletes and epithelial cell injury markers such as CC16.