Smoking-associated increase in mucins 1 and 4 in human airways

The mucus layer on the top of the airway epithelium forms the first line of defense against pathogens, toxins and foreign particles [1]. Mucus binds and clears the pathogens through mucociliary clearance. However, abnormal mucus production and clearance can contribute to respiratory diseases such as chronic obstructive pulmonary disease (COPD) [2‐4].

Mucin (MUC) macromolecules are believed to participate in the mucosal defense system by protecting the airway epithelium [5, 6]. Moreover, increased numbers of mucus producing cells i.e. hyperplasia of goblet cells, in the airway epithelium have been shown to associate with increased production of MUC [5]. MUCs are high molecular weight glycosylated proteins and are key components of most gel-like secretions [7]. Airway MUCs are major constituents of the secreted layer that comprise lung mucus in healthy airways and are a part of the mucociliary defense system that protects the lungs [8]. Many mucins, e.g. MUC1, − 2, − 3, − 4, − 5 AC, −5B, − 6, − 7, − 8, − 13, and − 19 have been detected in the lung. Among these, MUC1, MUC4 and MUC16 are transmembrane mucins with a large external polypeptide core protein [9‐11]. Human MUC1 is expressed in two different forms, as a transmembrane protein complex and as a secreted isoform [12]. Transmembrane MUC1 can also be cleaved from the epithelial surface by sheddases [13]. The initiation of mucin secretion involves a secretory cascade that results in a rapid release of MUCs both from the airway epithelial cells and secretory cells in the submucosal glands. MUC production is regulated by receptors for certain growth factors, e.g. epidermal growth factor receptor (EGFR) and 2 (EGFR2), also known as HER2 or ErB2 [14‐16].

MUC1 was originally thought to be expressed in most secretory human epithelial cells [17]. In histopathology, MUC1 has been commonly referred to as epithelial membrane antigen (EMA), or episialin, while its secreted form is also known as the Krebs von den Lungen antigen (KL-6) [17]. MUC1 has been shown to affect tumor progression in cancers in many organs e.g. a high expression of MUC1 has also been detected in lung cancer [9, 18]. Furthermore, MUC1 can act as a releasable decoy and limit the number of bacteria that access the epithelial cells [13]. Membrane-bound MUC4 also seems to be expressed in the lung [18]. The level of MUC4 is elevated in lung cancer and it has been speculated to play various diverse roles in tumorigenesis [19]. The β-subunit of MUC4 is able to interact with EGFR2 [20].

We hypothesized that the localization and expression of MUC1 and MUC4 in human airways would differ between smokers with or without COPD, and that a difference would also be evident in the levels of soluble MUC1. To test these hypotheses, we investigated bronchial mucosal biopsies and bronchial secretions from never-smokers, smokers with normal lung functions and patients with COPD from the Karolinska COSMIC cohort. For mechanistic purposes, we also evaluated the expression of EGFR1 and EGFR2.

We investigated both cell-specific immunohistochemical expression and the concentration of soluble MUC1 in human airways from age- and gender-matched healthy never-smokers (Never-smoker), current smokers with normal lung function (Smoker) and mild-moderate COPD-patients. We observed that smoking and a diagnosis of chronic bronchitis, but not airway obstruction, were associated both with increased MUC1 expression in cells suggesting basal cell phenotype, as well as the amount of soluble MUC1 protein in the airway lumen. The sources of the soluble MUC1 are likely to be secretions from the sub-epithelial glands, shedding of membrane bound MUC1 from epithelial cells or cells of hematopoietic origin [13, 38, 39]. Previous investigations of MUCs in COPD have focused mostly on MUC5A and MUC5B, whereas there are far fewer investigations of the membrane-bound MUCs such as MUC1 and MUC4. These present results and other studies have shown that the cellular localization of various types of MUCs in airways is variable whereas some MUC types, such as MUC1, seem rather to specifically originate from a certain cell type, namely basal cells.

Caramori et al. investigated MUC5A and 5B in bronchial rings of large airways from patients that had undergone surgery for lung cancer and detected different expressions of MUC5AC and MUC5B in central airways [40]. Kovalenko et al. utilized bronchial biopsies for the assessment of MUC2, MUC3 and MUC4 in large airways during an acute exacerbation of COPD, revealing that the expressions of both MUC2 and MUC3 were reduced [41]. O’Donnell et al. found that the levels of EGFR, ErB3 and MUC5AC but not MUC2 and MUC5B, were increased in bronchial biopsies of smokers as compared to non-smokers and COPD patients (n = 51) [15]. We found that MUC1 was localized in the cells suggesting basal cell phenotype of large airways whereas all the other epithelial cell types were negative. This was both an unexpected and a novel finding since the cell-specific localization of MUC1 protein in the large airways has not been previously published. The basal cell-origin expression for MUC1 was something of a surprise since this cell type is believed to act as a progenitor cell in the airways. The amount of KL-6, a mucinous sialylated sugar chain on the extracellular domain of human MUC1 protein, has been also found to be higher both in serum and in sputum of stable COPD patients [42] as well as during COPD exacerbations [43]. In peripheral lung, KL-6 is predominantly expressed in type II alveolar cells [44] and it has been shown to be highly expressed in bronchoalveolar lavage (BAL) fluid samples obtained from sarcoidosis patients [45]. Serum levels of KL-6 are known to be elevated in a variety of interstitial lung diseases including idiopathic pulmonary fibrosis; these diseases are characterized by alveolar epithelial cell damage [46]. An increased KL-6 level in lung carcinoma patients has been associated with a poor prognosis [47] and downregulation of the MUC1 was claimed to inhibit the non-small cell lung carcinoma progression [48]. In summary, it can be speculated that MUC1 may be a marker of lung progenitor cells both in airways and in alveoli, as MUC1 is predominantly expressed in the two cell types assumed to exhibit the properties of lung stem cells, namely basal cells of airways and type II alveolar cells [49].

Caramori et al. observed that MUC4 was expressed in all cell types present in the airways [50]. Our results partially confirm this finding, i.e. basal and ciliated cells were positive for MUC4, but in contrast, in that publication cells suggesting goblet cell phenotype were negative in most of the cases. Here, immunohistochemical reactivity for MUC4 in goblet cells was evident. The level of MUC4 expression was positively associated with smoking history, but not with lung function. In contrast to the Caramori study, Kovalenko et al. were unable to detect any positivity for MUC4 in the airway cells of COPD patients with GOLD stage III [41]. Similar to MUC1, most of the previous studies on MUC4 have been directed at lung cancer whereas investigations into airway disorders are sparse.

Our results revealed that soluble MUC1 protein levels were higher in the bronchial secretions of Smokers than in Never-smokers. Interestingly, Ishikawa et al. [42] identified higher levels of KL-6 both in sputum and plasma in smokers as compared to non-smokers, these results are in line with our findings of elevated levels of MUC1 in bronchial wash [42].

The level of MUC1 expression correlated with lung function and was increased in individuals with chronic bronchitis, especially in females. Chronic bronchitis has been suggested to be more common among females than males [51].

Previous studies have revealed that the suppression of MUC1 led to a down-regulation of the expression of EGFR, and moreover, MUC1 regulated the nuclear location and function of EGFR [52]. Exposure to cigarette smoke up-regulated EGFR mRNA expression and induced the activation of EGFR-specific tyrosine phosphorylation, resulting in an up-regulation of MUC5AC mRNA and protein production in an in vitro-model; these effects were inhibited by exposure to selective EGFR tyrosine kinase inhibitors [15]. Anagnostis et al. demonstrated that EGRF1, 2 and 3 mRNA levels were higher in COPD patients as compared to healthy smokers [53]. The largest immunohistochemical study so far performed showed that the level of EGFR was increased in goblet and ciliated cells in large airways of smokers irrespective of the presence of COPD [15]. Our results are at odds with the previous findings. In our subjects, EGFR2 protein expression was actually decreased in the smokers as compared to the never-smokers and the decrease correlated with smoking habits, whereas EGFR1 expression did not reveal any demographic correlations. Similar to the present work, a recent study investigated both EGFR1 and EGFR2 with immunohistochemistry, showing higher EGFR1 expression in ex-smokers with COPD than in ex-smokers without COPD [54]. It has been reported that HER2, a.k.a. EGFR2 was more frequent in COPD patients in comparison with lung cancer sufferers [55]. One clinical trial examining inhaled EGFR reported negative results since a four-week treatment did not decrease the epithelial levels of MUC2, MUC5A and MUC5B in the bronchial biopsies [56]. A recent study using microarray analysis showed that the numbers of EGF receptors were increased in COPD as compared to smokers without COPD [53]. Interestingly, our results indicated that the level of MUC4 expression in goblet cells correlated with that of EGFR2 while the expression of EGFR1 did not correlate with that of MUC1, which is in line with a previous report where the level of MUC4 expression correlated with that of EGFR2 in lung carcinoma specimens [57].

It has been suggested that the phenotype of the disease differs between genders, and we have previously reported evidence of molecular-level sex differences in the Karolinska COSMIC cohort i.e. in the levels of the proteome [25, 58], eicosanoid [21], metabolome [23] and cytokine concentrations [22]. Based on reports of an involvement of the estrogen receptor in the regulation of MUC1 expression [59], we also performed sex-stratified analyses. We found that smoking increased MUC1 expression in the large airways to a greater extent in men than in women, and this correlated negatively with lung function only in males although the tendency was the same in female group. In female smokers, the concentration of soluble MUC1 in the airway lumen was higher as compared to the healthy never-smoker group. Female subjects with chronic bronchitis had higher levels of MUC1 and MUC4 in comparison with the subjects without chronic bronchitis. Interestingly same tendency was seen in males but the differences did not statistically significant.

There are some limitations to this study. The group sizes in general are relatively small in particular in the COPD group, where the numbers of current smokers are not balanced with ex-smokers. Also, a major limitation is that the study is based only on biopsies from the large airways. In spite of these concerns, the results are significant, and the gender differences are supported by other data blocks from the same cohort [21‐30].

In conclusion, we detected an association of MUC1 and MUC4 expression with smoking and with chronic bronchitis, but not with airway obstruction. Our results support the hypothesis that mucins are important in development of airway symptoms due to smoking. Further studies to elucidate the contributions of each MUC-type in the pathogenesis of COPD would be beneficial.